Pacific Crucible: War at Sea in the Pacific, 1941-1942

by Ian W. Toll

  The Aleutians operation, which had been tacked on at the insistence of the Naval General Staff, bore no clear strategic relationship to the attack on Midway. The forces there engaged, at such a great distance north of Midway, would be incapable of coming to the timely support of the main fleet in the south. The Japanese could take the islands of Attu and Kiska with little opposition, but their possession was meaningless. They were of little value to either the Japanese or the Americans. Lacking any nearby bases of support, the occupiers would be left powerless to beat back a determined counterattack.

  The Midway offensive seemed motivated, at least in part, by a desire to make work for the main Japanese surface fleet of battleships, which had spent most of the war swinging peaceably at their anchors at Hashirajima anchorage in Hiroshima Bay. The disparaging term “Hashirajima fleet” had been making the rounds among the Japanese carrier task forces. Moreover, the crews of the anchored ships were growing more restless as the weeks wore on. Admiral Ugaki fretted that “the morale of the Main Body is stale after a long stay in home waters. I have encouraged them, but we must study methods of training afresh and at the same time engage in some operational action.”

  With such a formidable armada, possessing a clear margin of superiority in both carrier and surface naval strength, why didn’t Yamamoto just concentrate his forces and overwhelm the Americans? Why not just fight and win the battle “by the numbers”? Apart from the Aleutians misadventure, a face-saving concession to the Naval General Staff, it appears that the Midway plan was an attempt to balance the Mahanian rule of concentration with the need to disguise the true scale of the Japanese offensive from the Americans. Yamamoto’s main purpose in attacking Midway was to lure the American carriers into battle. If Nimitz should discover that nearly the entire Japanese navy was hovering off Midway, he would likely keep his ships safe in port, even if to do so meant conceding the atoll to the enemy. Yamamoto’s intention was to keep the bulk of his forces safely beyond range of Midway-based reconnaissance flights, so that an unsuspecting Nimitz would send the Pacific Fleet out to give battle. But in that case, could the widely dispersed Japanese forces get into action in time to make a difference?

  The dilemma, write Parshall and Tully, was never satisfactorily resolved. “If the premise is accepted that the Americans would have to be lured from Pearl Harbor in order to create the needed battle, there was no way to construct an operational plan whose distribution of warships was both deceptive and mutually supporting. The two goals were antithetical. Yamamoto knew he couldn’t have it both ways, and he willingly sacrificed mutual support to the perceived need for stealth.”

  Underlying the planned Midway offensive was a casual and yet insidious assumption that Japan was fated to continue its run of victories, an affliction of mind that the Japanese later called, with the benefit of grim hindsight, shoribyo, or “victory disease.”

  During the first week of May, Yamamoto and ranking officers from throughout the fleet gathered on the battleship Yamato for a multi-day planning conference. “The whole ship was aglitter with brass,” writes Yamamoto’s Japanese biographer; “at mealtimes, everywhere was jammed with distinguished officers; the most junior among the vice admirals were turned out of the C. in C.’s wardroom and obliged to dine in the ordinary wardroom, while rear admirals went to the gun room, and mere captains and the like were reduced to eating standing on the deck.” Conference tables and folding chairs were set up in the forward mess halls for the larger meetings. Tabletop war-gaming exercises were held in the Yamato’s spacious staterooms and wardrooms, with model ships placed on maps of the North Pacific and throws of the dice to signify the omnipresent role of chance. These exercises were putatively intended to scrutinize the plan and shed light on its weaknesses, but it was soon made clear that Yamamoto and his key staff officers had no interest in scrutiny, much less criticism. The games’ purpose, put simply, was to certify and validate Yamamoto’s plan. He was a gambler, and he had decided that he liked the odds.

  During one tabletop exercise on the second day of the conference, carriers of the “red team” (representing the U.S. Navy) appeared on Kido Butai’s port flank and dealt it a sudden, crippling air attack. The table judge, acting in his godlike capacity as umpire, ruled that the scenario was unbelievable, and expunged the results. With self-satisfied assurance, the judge declared that if the games had yielded such a result, the games must be flawed—for it was clear that the American carriers could not get into battle until they had been alerted to Nagumo’s presence by the first Japanese airstrikes on Midway.

  In a second episode, Midway-based land bombers counterattacked Kido Butai. The dice determined that the attacking planes had scored nine hits on the Japanese carrier group, sending two—Akagi and Kaga—to the bottom. Admiral Ugaki, Yamamoto’s despotic chief of staff, was acting in the role of supervisor and chief judge. “Wait,” he said; “we’ll reduce the number of hits to three.” With that arbitrary revision, it was determined that only Kaga was sunk, and Akagi lightly damaged. Later, Ugaki arranged to resurrect the Kaga for the next stage of operations, involving attacks against New Caledonia and Fiji.* These were only the most egregious of several other rulings in favor of the Japanese side. “This kind of supervision,” said Mitsuo Fuchida, “was enough to disgust even the most hardened flying officers among us.”

  When Yamamoto himself asked the very salient question at the heart of the problem—what would happen if American carriers suddenly appeared within striking range of Kido Butai?—Lieutenant Commander Minoru Genda simply dismissed the threat, exclaiming: “Gaishu Isshoku!” (“One touch of the armored gauntlet!”), an idiomatic expression translated as “We’ll wipe them out!” Yamamoto’s one concession to the games was to instruct Nagumo to maintain a reserve strike on board the carriers after the Midway attack had been launched, in case the American carriers should appear.

  By revealing the weaknesses of the plan and demonstrating the threat to Japanese forces, the tabletop games had done what they were designed to do. But they had been sloppy and halfhearted from the outset, and in the end they were little more than a charade. They had been necessary for appearance’s sake, but any flaws they revealed were swept under the rug. Among the officers of the fleet, there was whispered criticism, but few dared voice open opposition. According to Fuchida, who was present at the planning sessions, overconfidence was endemic. He and his fellow carrier aviators “were so sure of our own strength that we thought we could smash the enemy fleet single-handed, even if the battleship groups did nothing to support us.”

  The plan’s most eminent critic was Vice Admiral Nobutake Kondo, commander in chief of the Second Fleet, who met privately with Yamamoto on May 1 and laid out his objections. Kondo urged that Midway be postponed in favor of a renewed westward push through the Solomons to New Caledonia. There, in the south, the fleet could count on shore-based air support. None was available in the attack on Midway. And even if Midway were successfully occupied, the Japanese would face daunting logistical and military challenges in keeping the atoll supplied. Kondo later wrote that he thought Midway a dead end, because it would “not be easy for Japan to maintain a powerful air force there because of a too-long supporting route.”

  Yamamoto was unyielding. He brushed off Kondo’s suggestion that the operation be delayed a month, to allow for further study and preparation. There were important meteorological factors favoring early June. Only then (it was believed) would high tide arrive at the right moment, shortly after dawn, to allow the amphibious boats carrying the invasion force to clear the reef fringing Midway. By July, the Aleutians would be shrouded in fog. Moreover, it was too late to postpone the operation—the resources of virtually the entire Japanese navy had already been set in motion.

  Six months earlier, Yamamoto had demanded the surprise strike on Pearl Harbor over the objections of more prudent critics, and achieved a tactically spectacular victory. He had predicted that Japanese airpower would prevail over Allied surface naval s
trength, and he had been proved right. Having never failed and often succeeded, he enjoyed an aura of infallibility. He was a national figure, who had won the adulation and trust of the Japanese people. He had twice used the threat of his resignation to cow his critics. He would not hesitate to do so again.

  In his diary, Ugaki confided: “I could not help being a bit tired after a week of continuous conferences starting with the war lessons, table maneuvers, and briefings.” On the other hand, spring had arrived, the weather was fair, and the flowers along the shores of the Inland Sea were in full bloom. May 5 was the Japanese holiday known as Boys’ Day, and though Ugaki could not see any of the carp-shaped koinobori banners on shore, he was moved by the “fresh green” foliage on the steeply ascending slopes of the islands around the fleet anchorage. He added a few lines of verse to his entry for that day:

  All day long, verdant isles fresh to see,

  Around us, floating on the sea.

  HIGH UP IN THE HILLY BACKCOUNTRY of central Oahu, amongst cane fields and lush flora, a U.S. Navy radio-monitoring station plucked coded Japanese transmissions out of the atmosphere. No secure communications link then existed between the station and the Navy Yard at Pearl Harbor, so the raw coded intercepts were typewritten and entrusted to the care of a courier, who drove them down the hill in a jeep and delivered them by hand to the basement of the Fourteenth Naval District Headquarters, home of Joe Rochefort’s cryptanalysis unit.

  Operation MO and the big clash in the Coral Sea, like every other large-scale Japanese naval operation, had released a flood of radio messages that could be linked to known locations, events, and fleet units. It was, in other words, a codebreaker’s bonanza. In May 1942, the Allies were intercepting about 60 percent of Japanese naval radio traffic, attempting to decrypt about 40 percent of the intercepts, and managing to break about 10 to 15 percent of the code groups in those messages.

  Except in rare instances, however, they could not read complete messages, but only a few phrases in messages that were otherwise gibberish. Rochefort spent his eighteen-hour days hunched over messages in which fragmentary phrases appeared alongside blanks. “My job was to fill it in,” he said, and for that he relied on his encyclopedic memory. “I could remember back three or four months when that command had sent a similar message.” The previously deciphered phrases shed light on the blanks, especially when the same phrase appeared repeatedly alongside the same blank. With the help of the IBM machines, a working hypothesis was tested against every other message in which a given code group had occurred. In this manner, the breakdown of JN-25 (the main Japanese naval code in use at that time) proceeded geometrically—as the staff recovered more code groups, they gained more leverage over the remaining blanks.

  Even before the Battle of the Coral Sea, Rochefort had suspected that the Japanese were planning another major operation in the central Pacific possibly aimed at Midway or even Hawaii itself. In early May, Allied listening posts noted a surge in radio traffic emanating from Saipan, in the Marianas, but no corresponding rise from Truk, Japan’s principal naval base in the South Pacific. That pointed to a new offensive aimed directly across the Pacific, north of the equator, rather than a renewed offensive thrust in the south. A request for a large quantity of fueling hoses to be sent to Saipan seemed to indicate a fleet operation ranging far from Japanese bases. Reconnaissance flights from Kwajalein, in the Marshall Islands, were growing in length and frequency. Traffic patterns gave the impression that the “pending operations” would be large in scale, probably involving the First, Second, and Fifth fleets. Evidently, something big was about to happen in the central Pacific.

  On May 12, Rochefort called Layton. “I’ve got something so hot here it’s burning the top of my desk,” he said. “You’ll have to come over and see it. It’s not cut and dried, but it’s hot! The man with the blue eyes will want to know your opinion of it.” The message included a recovered code group for Koryaku Butai, or “invasion force.” Another phrase, tentatively identified as “forthcoming campaign,” was linked to the geographic designator “AF.” On the following day, May 13, two significant messages were intercepted and partly decrypted. In one, a Japanese vessel submitted a request for charts covering the Hawaiian Islands area. In the second, a forecast of movements by a supply ship indicated that the future base of the Third Air Group would be “AF,” and discussed logistics of moving base equipment, ammunition, spare parts, and ground crews to that location.

  As early as March, Rochefort had postulated that “AF” was Midway. All geographic designators beginning with A had been shown to be in the Hawaiian Islands and other American-held islands in the central Pacific. But Rochefort did not believe the Japanese possessed the military forces or (especially) the shipping to mount an invasion of Oahu or any of the other main islands of the Hawaiian group. What else in the region would be a likely target? Midway was the only atoll large enough to support an air group; the others, such as Palmyra or Johnston Island, were thought too small to be worth a major effort. A simple process of elimination, therefore, pointed to Midway. By May 12, there was no doubt in Rochefort’s mind that Midway was the target of the pending attack. Layton agreed: “We knew that AH was Oahu and that AK was probably the French Frigate Shoals. Now that the base supplies for an air unit were being readied for shipment into ‘AF’ with the occupation forces, it had to be one of our island bases within striking distance of Pearl Harbor. Midway was the obvious target, since it was nearly 150 miles nearer the Japanese on Wake than the alternative, Johnston Island.”

  The deduction was sound, and the conclusion correct—but none of that would matter unless Admiral Nimitz could be convinced. There was no ironclad proof that “AF” was Midway. It was a supposition, relying on a mosaic picture made up of many disparate bits of evidence. But Layton and Rochefort, having correctly foretold the Japanese move against Moresby, now enjoyed the CINCPAC’s full attention. In the month between Coral Sea and Midway, Nimitz would be called upon to make the most high-stakes decisions of his forty-year naval career, and he was willing to risk everything, based on the best estimates of his intelligence advisers.

  But Nimitz, in this instance, was not the ultimate authority. Behind Nimitz, in Washington, loomed the imperious figure of Admiral King, who was anxious not to lose any more of the navy’s few remaining aircraft carriers and was taking an active hand in choosing where and how to deploy them. In the period immediately after the Battle of the Coral Sea, King remained uneasy about New Caledonia, Fiji, and the other island groups straddling the U.S.-Australia sea link. Analysts at OP-20-G, the communications intelligence unit in Washington, fed King estimates that reinforced his preexisting belief that the Japanese would aim their heaviest blows in the South Pacific. They were fixated on a series of decrypts referring to a fleet rendezvous at Truk, and argued that enemy naval forces were moving south. Rochefort agreed that a Truk rendezvous was in the cards, but believed it would not occur until after the capture of Midway. “The amazing part of the whole thing was that many people could not accept this line of reasoning,” said Rochefort later. “We were quite impatient at Station Hypo that people in Washington could not agree with our rationale, because they had the same information and should, without any particular stress on their brains, have come up with the same answer.”

  Some even suspected that the cryptanalysts were being conned—that Japanese radio dispatches pointing toward Midway might be an elaborate hoax. One of Rochefort’s strongest pieces of evidence was the May 13 intercept in which the Third Air Group had announced its future base as “AF.” To critics, that seemed too good to be true. No less a figure than General Marshall, the army chief of staff, remarked that “one Japanese unit gave Midway as its post office address, and that seemed a little bit too thick.” Army commanders were concerned about the safety of Oahu, and feared that Nimitz’s deployments would denude the island of its air defenses. When asked to send B-17s to reinforce Midway, General Delos C. Emmons, the local commander of Army Ai
r Forces, was reluctant to comply. “Japs may be practicing deception with radio orders intercepted by us,” he warned Nimitz. “Estimates should be directed at capabilities rather than probable intentions. [Enemy forces] have sufficient strength to make damaging raid on Oahu with view to wrecking facilities [in] Pearl Harbor and Honolulu.”

  As these conflicting points of view came to a head, the feud between Washington and Hypo became bitter, ugly, and personal. In secure radio transmissions between the units, insults were even woven into the message “padding”—the random phrases placed before and after message texts for security. Layton suspected that someone in Redman’s camp even took steps to destroy records of internal communications that would have exposed OP-20-G’s errors to scrutiny. “It was a mess,” recalled Lieutenant Commander Ham Wright, one of Hypo’s senior codebreakers. “We would fight with OP-20-G all the time; we could not get together.” The Hawaiians charged that the Washingtonians were too apt to jump at shadows—that they made too much of stray, fragmentary decrypts, often failing to apply basic common sense to their own predictions. Having worked through the immediate aftermath of the raid on Pearl Harbor, when an almost unbelievable array of lurid rumors had circulated, the Hypo team had learned the importance of skepticism. They did not believe their counterparts in Washington had absorbed the same lesson. As Jasper Holmes put it, “No one who has not experienced it can realize how difficult it is to track the shadow of truth through the fog of war.”

  Rochefort and his team were certain that “AF” was Midway, and that the attack would come in the first week of June. They had convinced Nimitz of their reasoning. But they lacked enough evidence to overcome Washington’s obstinate claims to the contrary. To that end, Rochefort had a timely brainstorm. He proposed a ruse. The local commander on Midway would be told to transmit a plain-language radio broadcast back to Pearl Harbor, reporting that Midway’s desalination plant had broken down, and that as a result the atoll was short of fresh water. This was done. A Japanese monitoring station on Wake intercepted the broadcast and immediately passed it on to Imperial General Headquarters in Tokyo, which in turn alerted the Combined Fleet to a freshwater shortage on “AF.” The latter message was intercepted and broken by Hypo and by the cryptanalyst unit in Melbourne.