At Dawn We Slept: The Untold Story of Pearl Harbor

But Oahu lost track of the carriers completely and became confused about the support ships. For instance, on November 15 Rochefort associated Battleship Division Three (Hiei and Kirishima) and Destroyer Squadron One with the “South Expeditionary Force.” And the next day Rochefort’s men placed Zuikaku, albeit with heavy reservations, in Jaluit in the Marshalls.²

Kimmel commented on the nonappearance of Japan’s carriers. Layton replied that “that happens frequently and it is a normal assumption that they were then in port.”³ As he explained later, “When carriers are not heard from . . . they are most likely in port, because there they are on low-frequency low-power circuits that cannot be heard. . . . It is only when they originate traffic themselves at sea that direction-finder bearings can be taken. . . .”⁴

Thus, Naval Intelligence on Oahu had a nagging feeling that something peculiar was going on in the Japanese Navy. The interceptor operators noted that Japanese communications became “very involved. They would send the same message time and time again. . . .”⁵ All this was part of the deliberate Japanese pattern of deception.

On the other hand, the Army on Oahu thought it had control of the situation. A test held on November 14 enabled Hawaii to detect carrier planes launching about eighty miles at sea. Within six minutes pursuit aircraft took off and intercepted the incoming bombers some thirty miles from Pearl Harbor. Lieutenant Colonel Carroll A. Powell, Short’s signal officer, wrote: “All the general officers present were highly pleased. . . .”⁶

One official who did not sing the customary hosannas over conditions in Hawaii was Assistant Attorney General Norman M. Littell. Writing on November 14 to Marvin O. McIntyre, secretary to Roosevelt, he expressed considerable doubt about the effectiveness of the leadership in Hawaii:

Appointments in peace time to the “Paradise of the Pacific” are one thing, but . . . in the increasing tensions of the Pacific, there must be able and fearless men in command, capable of making decisions and getting things done. I gained the impression of weakness on all sides. . . . The admiral in command at Naval Headquarters [Bloch] was a fine and widely experienced old gentleman past sixty, who, in command of one of the most exciting naval posts under the American flag, goes to bed at 9:30 P.M., because, as he told me, he “could not stay awake after that.” The General in command for the Army [Short] was also past sixty, an estimable man of great experience . . . ⁷

Kimmel resented Washington’s failure to act on his recommendations or to keep him posted. Later he insisted that these “repeated rebuffs . . . and constant insistence of the Navy Department that the major emphasis was to be placed upon operations in the Atlantic strongly contributed” to his own “estimate that an air attack of the nature and force of that delivered on 7 December was not to be expected.” On November 16 he once more wrote to Stark, obviously in an irascible mood: “In repeated correspondence I have set forth to you the needs of the Pacific Fleet. These needs are real and immediate. I have seen the material and personnel diverted to the Atlantic. No doubt they are needed there. But I must insist that more consideration be given to the needs of the Pacific Fleet.”⁸

At this very time someone missed another timely signal, this one from Captain Ellis M. Zacharias. Now in command of the cruiser Salt Lake City, he still took a keen interest in intelligence. Around this date a Curtis B. Munson came to Hawaii with instructions from Naval Operations (OPNAV) to “open everything to him.” He sought out Zacharias about what to do in case of an armed uprising by the local Japanese in Hawaii or on the West Coast in the event of hostilities. He also wanted to know whether such a contingency was likely. Zacharias categorically discounted the possibility because war with Japan “would begin with an air attack on our fleet, and for that reason it would have to be conducted with the greatest secrecy, and therefore no Japanese . . . in the United States or in Hawaii, would be aware of the fact that such an attack was coming.” Sometime after the Pearl Harbor strike Zacharias requested and obtained Munson’s confirmation of this conversation. Munson added that Zacharias had also suggested that “the attack would conform to their [the Japanese] historical procedure, that of hitting before war was declared.”

Nomura, too, marshaled his thoughts on November 14 and composed what may well be the most straightforward and penetrating of his summaries. He ended: “I feel that should the situation in Japan permit, I would like to caution patience for one or two months in order to get a clear view of the world situation. . . .”¹⁰ If, on November 14, Nomura could counsel patience for another month or two, obviously he had no idea that his country planned to initiate war on December 7 (Washington time), especially by a strike on Pearl Harbor. By this time Nomura’s continual harping on the United States’ long-range staying power, distaste for the Tripartite Pact, support of China and Britain, and the basic unity of its people must have made him the “Abominable No-man” to the Foreign Ministry.

The end of a session with Hull on November 15 found Nomura in such despair that he sent Togo suggestions for handling “the closing of the consulates and recalling of the Ambassador” should the talks break down and Japan “pursue an unrestricted course.” His message makes clear, however, that he did not expect a Japanese-American war to break out, certainly not immediately. Rather, he anticipated “the same situation as now exists between Germany and the United States . . .”¹¹—that is, virtual severance of diplomatic relations without a formal state of war.

But Tokyo wanted no part of the American attitude and so informed Nomura emphatically on the fifteenth, adding: “Whatever the case may be, the fact remains that the date set forth in my message #738 [November 25, Washington time] is an absolutely immovable one. Please, therefore, make the United States see the light, so as to make possible the signing of the agreement by that date.”¹² Asking Nomura to break the deadlock and persuade the United States to sign a blank check for the Japanese in ten days was a manifest absurdity.

While those in on Magic paid fascinated attention to the diplomatic messages, one of the most significant tip-offs of the year slipped by unnoticed. On November 15 Tokyo enjoined its Honolulu consulate: “As relations between Japan and the United States are most critical, make your ‘ships in harbor report’ irregular, but at a rate of twice a week. Although you are no doubt aware, please take extra care to maintain secrecy.”¹³

ONI’s fortnightly summary for the period ending November 15 reflected some of McCollum’s views: “The approaching crisis in United States-Japanese relations overshadowed all other developments in the Far East during the period.” No one apparently expected Kurusu’s mission to succeed, “the envoy himself reportedly expressing extreme pessimism. . . .”¹⁵ Nevertheless, if the rest of Naval Intelligence agreed with McCullum’s conjecture that Japan would strike the American Fleet, this long summary did not show it. On the contrary, it indicated that Japan was not yet “strong enough in Indo-China to attack Yunnan or even Thailand.” And the Combined Fleet remained “in home waters, nearly in full force in the Inland Sea.”¹⁶

If Tokyo’s message of November 15 to Kita passed over Washington’s head, Yoshikawa did not miss its full implications. Thenceforth he scouted the Fleet almost every day and made a special daily graph of the location and movements of all ships in harbor. Sometime in October he began destroying many reports from his secret file in Kita’s office and thereafter kept the file weeded. Later he claimed that he did so only because he needed the space. Of course, he acknowledged that a certain danger existed in having too much material lying around.¹⁷ Either way the massive destruction ensured that one month later he had the minimum of burning to do on short notice.

He gained the impression that the Pacific Fleet still operated on a peacetime basis. The ships did not appear to be particularly on the alert, and they paid almost no attention to camouflage. The Fleet returned to Pearl Harbor on weekends and granted shore leave to many officers and men. This was all to the good for the Japanese, although as a sailor Yoshikawa deplored the habit. He thought the U.S. Navy had no business adhering to a regular schedule with the political situation obviously so critical.¹⁸

Strangely enough, the worse the international situation, the safer Yoshikawa may have been. Sometime after December 7, Shivers told Fielder that he knew “the entire espionage ring centered around sic the Japanese Consulate, but diplomatic immunity prevented his investigation, and that anything he did might start the overt act which would create war.”¹⁹

The Japanese consulate used the J-19, not the Purple, diplomatic code. Rochefort was, in McCollum’s words, “the only officer in our Navy who is a top-flight cryptographer and radio man, and who also has a thorough knowledge of the Japanese language.”²¹ Since he received the J-19 keys within twelve hours of their solution in Washington, he and his unit would have had little difficulty in decoding and translating the messages between Tokyo and Honolulu.²²

The situation was like some cosmic practical joke set up by an unusually mordant-minded demon. While the Japanese Navy took the most elaborate precautions for communications security, the Foreign Ministry chattily gave itself away. In spite of the Japanese radio smoke screen, Hawaii’s Combat Intelligence still picked up significant patterns. Had Layton and Rochefort been able to sit down with the consular intercepts spread out before them, along with Rochefort’s communications gleanings, they might have cracked Yamamoto’s secret. If Rochefort had received the messages from MacKay Radio as well as RCA, by the latter part of November he would have been breaking down the latest in the “bomb plot” sequence. More were to come in early December, culminating in a single message which was a virtual giveaway.

Meanwhile, the State Department gave Kurusu, Japan’s special envoy, every cooperation, arranging to delay until November 7 the China clipper due to leave Hong Kong on the fifth so that Kurusu might embark. Kurusu looked like the complete American stereotype of a Japanese—short, slim, with reticent eyes behind the inevitable glasses, a mustache, and slanted brushstroke eyebrows imparting a perpetually surprised expression. What, if anything except camouflage, Japan hoped to gain by his mission, and how much—again, if anything—he knew about his country’s plans for the near future, who can say with complete authority?

Kurusu reached Washington at 1330 on November 15. That very day (November 16 in Tokyo) the Foreign Ministry sent Nomura a most disquieting reply to his excellent summary of the fourteenth:

. . . you may be sure that you have all my gratitude for the efforts you have put forth, but the fate of our Empire hangs by the slender thread of a few days, so please fight harder than you ever did before. . . . In your opinion we ought to wait and see what turn the war takes and remain patient. However . . . the situation renders this out of the question. I set the deadline for the solution of these negotiations in my #736, and there will be no change. . . .

Thus, once more Tokyo insisted upon November 25 as the cutoff date. And on the fifteenth the Foreign Ministry had sent instructions to Washington, as well as several other Japanese diplomatic missions, for destroying code machines in the event of emergency.²⁴

On the seventeenth, at 1030, Nomura presented Kurusu to Hull. The secretary of state took an immediate dislike to the new envoy.²⁵ Nevertheless, whatever his personal bias, Hull complimented Kurusu on the way he had “handled his relations with the public since coming to this country. He also spoke highly of the respect and confidence in which the Secretary and his associates hold the Japanese Ambassador.”²⁶ The three diplomats adjourned to the White House for an appointment with the President at 1100. Although they talked together for an hour and a quarter, nothing new developed. Nomura took a certain comfort from Roosevelt’s quoting William Jennings Bryan: “There is no last word between friends.”²⁷

When Hull, Nomura, and Kurusu met on the eighteenth, they faced the difficulty of working out a Japanese-American rapprochement while Japan was still tied to Germany. Hull even suggested “that Hitler would eventually, if he was successful, get around to the Far East and double-cross Japan.” Kurusu protested that his country could not “abrogate the Tripartite Pact but that Japan might do something which would ‘outshine’ the Tripartite Pact.” Eventually Nomura offered the possibility of restoring the situation to its status before the Japanese had moved into southern Indochina and the subsequent economic freeze. Hull was somewhat discouraging, pointing out that the troops withdrawn from French Indochina could be “diverted to some equally objectionable movement elsewhere.” But he promised to take up the proposal with the British and Dutch.²⁸ Although Hull retained his poker face with the two Japanese, obviously he saw in this modus vivendi the bare possibility of an interlude in which Japan could edge away from all-out militarism toward a more reasonable approach.

If the Japanese Foreign Ministry wanted to keep peace with the United States, why did it not snap up Nomura’s alternative? On the face of it the suggestion offered an acceptable breather. Each side gave up something tangible enough to provide evidence of good faith and in so doing removed two thorny obstacles: The Japanese pullout from southern French Indochina would take away the most immediate threat against the Allies in Southeast Asia, while the American financial thaw would cancel Japan’s prime grievance.

Unfortunately, however, Togo’s government wanted peace with the United States strictly on terms of full and simultaneous acceptance of all Japanese terms. What is more, Togo disliked Nomura’s excursions into the forbidden field of creative diplomacy. Even before Nomura and Kurusu had shot off several messages on the eighteenth explaining the modus vivendi in enthusiastic detail, the Foreign Ministry had dispatched to Nomura instructions concerning the repatriation of those Japanese who had not boarded the special ships.²⁹ What Tokyo did not confide to Nomura and Kurusu was the fact that this entire mission, purportedly a second evacuation voyage by the Tatuta Maru, was a blind. The vessel would leave Japan in early December, sail eastward for a few days, then turn back to Yokohama—a hoax equally cruel to the Americans aboard and to the Japanese eagerly awaiting the chance to go home. In any case, this message was enough to make any ambassador nervous and to alert any good intelligence officer.

The same was true of the message which Tokyo dispatched to Kita in Honolulu that same day, November 18: “Please report on the following areas as to vessels anchored therein: Area ‘N,’ Pearl Harbor, Manila Bay Honolulu [italics in original], and the Areas Adjacent thereto. (Make your investigation with great secrecy.)” “Manila Bay” had been crossed out; presumably Naval Intelligence meant “Mamala Bay.”³⁰

Receipt and processing of these vital consular messages did not permit the Army to translate this one until December 5. The fact that the Americans had broken the J-19 code did not mean that they could read the messages automatically. The keys changed daily, and in order to break them, the decrypters had to have a fair volume of traffic to work on. It was, in Mr. X’s words, “plenty tough.”³¹ Even so, the main delays resulted from transmitting the messages to Washington by mail. Here, then, was another valuable, if belated, hint to Washington that the Japanese were still interested in the exact location of Kimmel’s ships in Pearl Harbor. Nor could anyone miss the emphasis on “great secrecy.”

Kita promptly answered Tokyo with one of Yoshikawa’s meticulous reports on November 18. It was detailed, precise, and suggestive.³³ But the Japanese messages which really raised a fuss in ONI and G-2 were two which the Foreign Ministry dispatched to Nomura on November 29:

. . . In case of emergency (danger of cutting off our diplomatic relations), and the cutting off of international communications, the following warning will be added in the middle of the daily Japanese language short wave news broadcast.

(1) In case of a Japan-U.S. relations in danger: HIGASHI NO KAZEAME sic [“East wind rain”].

(2) Japan-U.S.S.R. relations: KITANOKAZE sic KUMORI [“North wind cloudy”].

(3) Japan-British relations: NISHI NO KAZE HARE [“West wind clear”].

This signal will be given in the middle and at the end as a weather forecast and each sentence will be repeated twice. When this is heard please destroy all code papers, etc. This is as yet to be a completely secret arrangement.

Forward as urgent intelligence.³⁴

When our diplomatic relations are becoming dangerous, we will add the following at the beginning and end of our general intelligence broadcasts:

(1) If it is Japan-U.S. relations, “HIGASHI.”

(2) Japan-Russia relations, “KITA.”

(3) Japan-British relations, (including Thai, Malaya and N.E.I.): “NISHI.”

The above will be repeated five times and included at the beginning and end.

Relay to Rio de Janeiro, Buenos Aires, Mexico City, San Francisco.³⁵

Tokyo’s “winds” dispatches went through the mill on November 28 and 26 respectively. From then on both services knocked themselves out to intercept the implementing messages. Washington directed Rochefort to establish “a listening watch on the most likely frequencies.” So he sent four language officers to Aiea, “where they covered on a twenty-four hour basis one or more frequencies in addition to all the known broadcasts from Tokyo. Results were nil.”³⁶ McCollum testified, “We were all looking for it. . . .” And so far as he knew, “we were continuing to look for that after the bombs had started falling on the fleet. . . .”³⁷

Why the “winds” message caused such a stir when the much more significant “bomb plot” series did not is another of the Pearl Harbor mysteries. Certainly no one in Washington needed the Japanese Foreign Ministry to tell him that diplomatic relations with Japan were “becoming dangerous.”

Kimmel and Bloch learned of the “winds” message as information addresses of a dispatch dated November 28 from the Asiatic Fleet, which also advised: “British and Comsixteen monitoring above broadcasts.”³⁸ The senior United States Army representative in Java also scooped it in and on December 3 forwarded the information to Miles. This was the only version to use the word “war,” his message including the sentence: “Japan will notify her consuls of war decision in her foreign broadcasts as weather report at end. . . .”³⁹ Either Java placed the worst possible interpretation on the Japanese wording, or the Foreign Ministry had alerted its Far Eastern diplomatic missions in much more forceful language than they had used to Nomura. Tokyo had done exactly that in connection with the imperial conference of July 2.

Consul General Walter Foote at Batavia advised State on December 4 in slightly less inflammatory phraseology—“When crisis leading to worst arises . . .” and “When threat of crisis exists . . .”—then threw cold water on the whole affair by adding prosaically, “I attach little or no importance to it and view it with some suspicion. Such have been common since 1936.”⁴⁰

No other aspect of the Pearl Harbor story has generated more heat and less light than the “winds” code. Was the implementing message received or not? It appears unlikely that Japan ever used the code. Normal channels of communication remained open between Tokyo and Washington until after the Pearl Harbor attack, so the need to broadcast “East wind rain” never arose.

Not that it really mattered. Looking backward down the corridor of time, one can appreciate that the joint congressional committee judged correctly when it reported: “Granting for purposes of discussion that a genuine execute message applying to the winds code was intercepted before December 7, we believe that such fact would have added nothing to what was already known concerning the critical character of our relations with the Empire of Japan.”⁴¹

American authorities would have been much better advised to pay attention to such dispatches as the one which Tokyo sent to the Honolulu consulate on November 20: “Strictly secret. Please investigate comprehensively the fleet—–bases in the neighborhood of the Hawaiian military reservation.” A garble left one word unintelligible; however, we may safely assume that the missing link is “air.”⁴² The Japanese took a particular interest in U.S. air power in Hawaii and the absolute necessity of pinning it down if the assault on the Fleet was to succeed. Even granted the garble, this message again deserved more attention than it evidently received, for nothing in it could be stretched to cover interest in Fleet movements.

The summary from McCollum’s office in Washington for the same date placed Hiei and Kirishima in the Kure-Saeki area: Akagi, Kaga, Soryu, and Hiryu in South Kyushu; Zuikaku in the Kure-Sasebo locality, with her sister, Shokaku, at Takao in Formosa, albeit with a question mark. Japan was winning the game of hide-and-seek.⁴⁴

Rochefort looked for a particular pattern. He believed that a major Japanese operation would reveal itself in “three definite stages”: first, “a large flurry of traffic”; then “a stage of apparent confusion . . .” caused by “the regrouping of the ships and the units”; third, radio silence would descend, “and when radio silence started then you knew something was up. . . .”⁴⁵

The Japanese placed surprisingly little communications security around the forces to move south, just as they had never made any secret of their expansionist designs on that part of the world. Therefore, as Rochefort later pointed out, the estimates made at the Pearl Harbor and Cavite Combat Intelligence units were correct in every respect but one—Pearl Harbor.

There is no indication that anyone else on Oahu equated the Japanese movements with peril to Pearl Harbor. Kimmel frequently dropped in on Rochefort to discuss matters, but nothing in their testimony hints that these talks touched on a possible attack on the Fleet in its anchorage. As Rochefort said later, “No one thought in terms of Pearl Harbor at the time.” Indeed, it was not Rochefort’s business to evaluate. He picked up the data, summarized them, and passed them to Fleet Headquarters to interpret; he had “no other information available” aside from that which he “gathered by radio.”⁴⁶

Kimmel had added Soc McMorris to the very restricted list of those in the know concerning Rochefort’s Combat Intelligence.⁴⁷ But the war plans officer, too, believed that “the Japanese interests lay in the Asiatic area and that they could more effectively utilize their full power in that area.” He did not believe an attack on the Pearl Harbor region would result in sufficient damage to make it worthwhile. And he thought the Japanese would reason as he did. What is more, he regarded an attack “by saboteurs or by submarines as extremely probable.” At no time, however, did he “envisage such an attack as actually occurred.”⁴⁸

But the Japanese planned to strike with terrifying power in both directions at once. In pursuit of this policy Togo gave short shrift to Nomura and Kurusu’s suggestion of a temporary solution to the diplomatic problems. He advised Kurusu directly on November 19: “The Ambassador . . . having received our revised instructions . . . will please present our B proposal of the Imperial Government, and no further concessions can be made.

“If the U.S. consent to this cannot be secured, the negotiations will have to be broken off. . . .”⁵¹

Here Nomura seems to have come perilously close to losing his temper. He suggested, among other things, that maintenance of the status quo or initiating military action would “bring about a situation full of dynamite and finally lead us to an armed clash.”⁵² He also begged Tokyo to hold up at least four or five days in announcing the evacuation ships: “. . . At a time when we are thus pressing for an early reply, I feel that it would do us great harm were we to announce that we are having ships, with all the accompanying dark implications, leave on or about the 25th or 26th. . . .”⁵³

Need we add that the Foreign Ministry declined these suggestions,⁵⁴ so indicative of Nomura’s good faith and good sense?

The next day, Thursday, November 20, might not have been the most tactful occasion for presenting Proposal B. As Nomura explained to Tokyo, this was “America’s biggest holiday. They call it ‘Thanksgiving.’ In spite of that, however, Mr. Hull not only agreed to talk with, but seemed glad to see both me and Ambassador Kurusu when we went to call on him.”⁵⁵

No longer able to stall off the evil hour in the face of Togo’s direct order, Nomura and Kurusu handed Hull the fateful document. Hull promised to give it “sympathetic study,”⁵⁶ but he knew from Magic “that this proposal was the final Japanese proposition—an ultimatum. . . .”⁵⁷

In exchange for providing Japan with all the oil it required, lifting the freeze, and discontinuing aid to China, the United States received only the promise of Japanese removal of troops from southern Indochina to the northern part of the country—the very thing Togo had just assured Nomura could not be done. In any case, such a move would have kept a dagger at the jugular vein of all Southeast Asia and the East Indies. True, the proposal agreed to no further Japanese moves into southeastern Asia and the southern Pacific, but that left a lot of Asia unaccounted for—notably Siberia and unoccupied China. Furthermore, the Japanese maintained utter silence about the Tripartite Pact. They offered no assurance that they would not supply Hitler through such a newly opened American trade channel. As Hull saw it, agreement would have meant

CHAPTER 44

“A SITUATION FULL OF DYNAMITE”