10

Opening Act

The battle opened not in the central Pacific but among the fog enshrouded islands of the Aleutian archipelago some two thousand miles to the north. Part of the price that Yamamoto had to pay for getting the Navy General Staff to accept Operation MI was his agreement to continue with Operation AL—the occupation of several small islands in the western Aleutians. Though at the time the Americans assumed that this was a diversion for the Midway campaign, it was a stand-alone operation with quite limited goals: the occupation of the islands of Attu and Kiska in the western island chain in order to expand the empire’s defensive perimeter. To prevent the Americans from interfering with these landings, the Japanese planned to neutralize the American base at Dutch Harbor on the island of Unalaska, some four hundred miles east of Attu and Kiska. The overall operation was under Vice Admiral Hosogaya Moshirō, who commanded the Japanese Fifth Fleet, and the force assigned to strike Dutch Harbor consisted of two carriers and their supports under the command of Rear Admiral Kakuta Kakuji. One of the two carriers was the Ryūjō, which carried only thirty-seven planes. The other was the 24,000-ton Jun’yō, which carried fifty-three planes. Combined with two heavy cruisers and a destroyer screen, they comprised the Second Striking Force—a kind of mini Kidō Butai.

Though the entire Japanese operational plan for June of 1942 was characterized by a dispersal of force, the decision to send two carriers to the Aleutians seems particularly profligate. In fact, however, neither could have been used to reinforce the Midway-bound Kidō Butai. The Ryūjō was simply too small, and the Jun’yō, which had originally been laid down as a passenger liner and converted into a carrier only recently, had a top speed of only 24 knots, which meant she could not keep up with the Kidō Butai; even the plodding Kaga could sustain 28 knots. On the other hand, the fighters and bombers on the decks of those two carriers might have played an important—even a decisive—role in the Battle of Midway had some or all of them been transferred to the Zuikaku. This was not done mainly because the Japanese did not believe the Zuikaku was needed, but also because the pilots in the Jun’yō’s air wing were relative novices with little if any battle experience. Because of that, though the Jun’yō nominally carried fifty-three aircraft, Kakuta could count on only about thirty-three of those for combat operations.¹

The Americans had long been aware of Alaska’s vulnerability. The tail end of the Aleutian archipelago at Attu was only 650 miles from the northernmost of the Japanese Kurile Islands. The Japanese had a small base on Paramushiro in the Kuriles, but until 1937 the Americans had virtually no military presence in Alaska. That year, the Navy began construction of a seaplane base at Sitka, and soon afterward another at Dutch Harbor, though that was still some 1,400 air miles from Paramushiro. By 1942, these two American bases hosted two destroyers, three Coast Guard cutters, and a handful of long-range PBY Catalinas, all under the command of Navy Captain Leslie E. Gehres. The Army had twenty bombers plus forty pursuit planes under the command of Brigadier General William O. Butler.

In January 1942, Roosevelt had asked King about “operational readiness in the Alaskan area.” At the time King dismissed the idea of a Japanese assault there because “a landing in Alaska would be a costly undertaking, unproductive of immediate results, and would expose the occupying forces to strong counter attack.” Despite King’s skepticism, however, the United States did begin to build up its Alaskan forces, largely in response to Alaska’s governor, Ernest Gruening, who complained to FDR’s secretary of the interior, Harold Ickes, that “Alaska is far from prepared for eventualities.” What Gruening wanted was money—for airfields, planes, and equipment. And he got it. Ickes recognized that Gruening’s request was as much political as strategic, and he sent it on to the president, who approved the construction of a new 5,000-foot airstrip on the island of Umnak, just west of Dutch Harbor.²

For his part, Nimitz knew (thanks to the code breakers) that the Japanese planned to attack the Aleutians at the same time as they closed on Midway. He was not willing to weaken his carrier task force to defend those distant islands, nor was he willing to let them go by default. He appointed newly promoted Rear Admiral Robert A. “Fuzzy Theobald, a stocky 1907 Annapolis classmate of Ray Spruance, to command a surface force of five cruisers (two heavy cruisers and three light cruisers) plus four destroyers as Task Force 8, and gave him orders to defend the archipelago and “inflict maximum attrition” on the enemy attackers. ³

Without carriers, Theobald knew he had to depend on General Butler and the Army for his air support. In theory, at least, Theobald had command authority over Butler’s bombers, for in April Marshall and King had agreed that “when a state of fleet-opposed invasion is declared, unity of command is vested in the Navy.” The problem was that neither service had any practical experience with joint operations, and there was no clear chain of command or channel of communications that allowed the two services to work together efficiently. As there was no Department of Defense or Joint Chiefs of Staff at the time, the two services were entirely separate. Though Nimitz asked King to “inform Army that surface force will be almost completely dependent on them for air cover,” King could not order it. The only person with simultaneous command authority over both the Army and the Navy was the president himself. As a result, there was confusion and missed opportunity on the American side, though, as it turned out, this was matched by confusion and missed opportunity on the Japanese side, too.⁴

 Rear Admiral Robert A. “Fuzzy” Theobald commanded the cruiser-destroyer force dubbed Task Force 8 that Nimitz assembled to defend the Aleutian Islands. (U.S. Naval Institute)

Theobald and his task force reached Kodiak Island, some five hundred miles east of Dutch Harbor, on May 27—the same day that the crippled Yorktown appeared off Pearl Harbor. He met with General Butler at his headquarters and explained his plan to inflict “maximum attrition” on the approaching enemy, in conformance with Nimitz’s directive. Theobald’s surface force could not close with Kakuta’s carriers unless the Americans first gained command of the air. The Navy Catalinas were ideal for scouting, but they were flimsy and vulnerable and relatively useless in an attack, especially against carriers. Theobald could not even plan a night destroyer attack, because at that latitude in early June there was hardly any night. Butler would have to bring his Army bombers to the forward airstrip at Cold Bay and the new field at Umnak, where they would wait for a sighting report from the Catalinas and then attack. If their attacks sufficiently weakened the carriers, Theobald could then close with his cruiser force and finish them off with gunfire.⁵

Butler was less than enthusiastic about this plan. He objected to concentrating his air forces at Cold Bay and Umnak, more than five hundred miles to the west, instead of at Kodiak, where they could protect the city of Anchorage. Those western bases lacked support facilities and protective revetments; Butler worried that his planes would be sitting ducks. The conversation was courteous enough, but Butler stubbornly resisted the idea of staging his bombers that far west. Theobald considered asserting his newly established prerogatives as joint commander and simply ordering Butler to do it, but feared that if he did so it would “create an initial schism between the Army and the Navy that [would] adversely affect all [their] operations from then on.” So he tried to reason with Butler, pointing out the advantages of acting offensively rather than defensively. He reminded Butler that his own surface ships “could accomplish little until the enemy aircraft carriers were definitely accounted for” and reminded him of Nimitz’s orders to inflict “maximum attrition” on the enemy. By the end of their five-hour conference, Theobald thought he had convinced Butler, and on June 1 he returned to his flagship, Nashville, and led his task force back to sea, taking up a position four hundred miles to the south. There his ships were cocooned in a seasonal fog so thick that, as one officer on the Nashville recalled, “for three days we never saw the ship ahead of us.”⁶

Kakuta launched his first strike against Dutch Harbor early on the morning of June 3 (Alaska time). He sent off partial strikes from both of his carriers, but the inexperienced pilots from the Jun’yō got lost in the thick weather and turned back, and as a result only nine bombers from Ryūjō, plus three fighters, made it through to the target. The Ryūjō’s planes inflicted moderate damage, hitting the radio station, the oil tank farm, and an Army barracks, killing twenty-five Americans at the cost of two of their own planes.⁷*

Now was the time for the American Army counterstrike against the carriers. Navy search planes found Kakuta’s carriers a mere 165 miles away and radioed their coordinates; one of the snoopers even managed to drop a few bombs, though none struck an enemy ship. Theobald expected that the Army bombers would now sortie. Instead, the Army pilots insisted that “they had to await an order from General Butler,” who had apparently had second thoughts since agreeing to Theobald’s arrangements. He told another Navy officer that he doubted the Army planes could even defend their own airfields, much less damage the enemy. He therefore radioed Theobald from Kodiak, “Unless otherwise directed by you [I] will not advance bombing squadrons from Kodiak to Cold Bay Area.” Since Theobald was observing radio silence, he could not respond to this astonishing message. Instead, he sent the destroyer Humphreys racing back to Kodiak with a written order.⁸

Thus prodded, Butler released his bombers, though this did not result in an immediate strike. While en route to the target, the first group of Army planes received a radio report that the Catalinas had temporarily lost contact with the enemy. Rather than continue on in the expectation that contact could be reestablished—which it was—they turned around and returned to base. When a second group of bombers flew out toward the coordinates, the Army pilots fanned out to attack individually rather than attempt a coordinated strike. Most bombed from high altitude, some releasing their bombs blindly from above the stratus cloud layer, simply guessing at the enemy’s position “by calculation”—essentially by dead reckoning. As Theobald noted later, “such an attack could not be sure of hitting Kiska Island,” much less an enemy warship.⁹

A few of the bombers had been equipped with ship-killing torpedoes, but the Army pilots, inexperienced with such weapons, released them, too, from high altitude, all but ensuring that they would break apart upon striking the water. (One Army pilot claimed that he landed his torpedo square on the deck of a carrier, though that proved false.) Some pilots failed to locate the enemy at all and returned to base still carrying their heavy torpedoes. To avoid landing with such volatile cargo, they jettisoned them on the rocks offshore. Unaware that the torpedo warheads became armed only after the torpedo ran for a prescribed distance in the water, they reported them as “duds” because they didn’t explode. Theobald was willing to forgive their ignorance of torpedo ordnance but regretted the loss of the torpedoes themselves, since they were scarce and expensive. By the end of the day, the Navy had lost six PBY Catalinas; the Army had lost two bombers and two P-40 pursuit planes. The Japanese lost four scout planes—shot down by Army P-40s operating from the field on Umnak—yet suffered no damage to their strike force. Having conducted the required attack on Dutch Harbor, Kakuta turned west to cover the landings on Attu and Kiska. The ever-present fog delayed those landings, and he decided to send a second strike against Dutch Harbor on June 4. This time his planes destroyed several oil tanks and a few more buildings, though they again failed to knock out the base.¹⁰

Theobald was so disgusted with the performance of the Army bombers that he decided to return personally to Kodiak in his flagship to talk again with Butler. He arrived there on the morning of June 5 and immediately went to see a somewhat chagrinned Butler, who was aware of how little his bombers had accomplished. Theobald formally requested that Butler allow the pilots to strike without waiting for permission and suggested that they make concentrated and coordinated attacks rather than isolated high-altitude bombings. He left that same afternoon, believing, or at least hoping, that the problems had been resolved. Nonetheless, Butler’s planes never did manage to strike the Japanese fleet. Indeed, so feeble was the American attack on Kakuta’s carrier force that the Japanese commander may have been unaware that he had been under attack at all. As one senior naval officer reported to King regarding the Army bombers: “Either they were too slow in taking off, or the weather was too bad, or the distance was too great, or they couldn’t find the enemy.” In the conditions that prevailed off the Aleutian archipelago, some of this was not altogether surprising. Still, the Navy was inclined to attribute it, in part at least, to Army timidity.¹¹

Kakuta, too, was disappointed with the effect of his air attacks on Dutch Harbor. Far more distressing, however, was the radio message he received the afternoon of June 5 from Yamamoto. For the commander in chief to break radio silence at all was astonishing enough; the message he sent was even more so. Hosogaya and Kakuta were to break off their attacks in the Aleutians, cancel the landings, send the transports back to Japan, and close in on the Kidō Butai near Midway. Four hours later, Yamamoto reversed himself and cancelled those orders, telling his northern force commanders to complete their mission after all. But clearly, something had gone very wrong with the Kidō Butai.¹²

In fact, nearly everything had gone wrong, starting with the fact that the Japanese had failed to determine whether the American carriers they hoped to lure out to their destruction were even present in Pearl Harbor. They did have a plan to find out. Nearly three months earlier, well before the fateful conference in Tokyo at which Nagano and Fukudome had capitulated to Yamamoto’s blackmail and approved the Midway plan, the Japanese had conducted a long-range reconnaissance of Pearl Harbor using two giant Kawanishi flying boats. These remarkable four-engine seaplanes, called “Emilys” by the Allies, were 92 feet long (30 feet longer than the American Catalinas) and had an astonishing range of over 4,500 miles, which meant that, theoretically at least, they could fly from the Marshall Islands to Pearl Harbor and back without stopping. Such a flight would leave no margin for error, however, and so Commander Miyo (who a month later would strenuously oppose Yamamoto’s Midway plan) suggested that their range could be extended even further by refueling them at sea from submarines. This notion hinted at using them to bomb American cities along the continental West Coast. More immediately, it provoked discussions about a second attack on Pearl Harbor, a scheme that was code-named Operation K.¹³

Americans were very much aware of the possibility of long-range air strikes by seaplanes refueled at sea. Three months before Pearl Harbor, Hypo analyst Jasper Holmes, writing under the pen name “Alec Hudson,” had published a story in the Saturday Evening Post about American seaplanes refueled by submarines striking enemy bases three thousand miles away. In Holmes’s fictional tale, “twelve big bombers” attacked an enemy base “with machinelike precision,” wrecking an invasion convoy. Edwin Layton later speculated that Holmes’s story might have given the Japanese the idea for Operation K, but in fact the Japanese had begun experimenting with a seaplane-submarine partnership as early as 1939. After the war began, the Japanese planned to conduct a whole series of seaplane raids against Pearl Harbor—to keep track of the comings and goings of American warships, as well as to keep the Americans on edge and off balance by bombing them periodically. In the end, however, this dual objective undermined Japanese ambitions, for it focused American attention on the program and therefore compromised it.¹⁴

The first (and, as it turned out, only) seaplane attack on Pearl Harbor occurred in the first week of March 1942, before Yamamoto even submitted his Midway plan to the Naval General Staff. Two Kawanishis, each of them armed with four 500-pound bombs, took off from Wotje Island in the Marshalls on March 2 and in thirteen and a half hours flew 1,605 miles to an unoccupied atoll called French Frigate Shoals, halfway between Pearl Harbor and Midway. There they refueled from two prepositioned submarines, then flew on to Oahu, another 560 miles to the southeast, arriving just past midnight on the morning of March 4. By then the weather had thickened, and visibility over the American naval base was virtually zero. The pilot of the lead plane, Lieutenant Hashizume Hisao, could see a slight glow through the cloud layer, but not much else. Thinking that he had glimpsed the outline of Ford Island in Pearl Harbor through a gap in the clouds, he dropped his bombs. His consort did the same. Then both planes headed back for the Marshall Islands, another two thousand miles and fifteen nonstop hours away.

 The long range of the big four-engine Kawanishi H8K Type 2 seaplanes, called “Emilys” by the Allies, encouraged Japanese planners to consider long-range raids against American bases. (U.S. Naval Institute)

For all the effort and expended fuel, the raid did no damage whatever. Hashizume’s four bombs fell on the forested slopes of Mount Tantalus behind Honolulu, and the four from the other plane fell into the water near the entrance to Pearl Harbor. Moreover, the heavy cloud cover meant that Hashizume could not report with much certainty about what ships were or were not in the harbor, though he claimed to have seen at least one carrier. ¹⁵

The most important consequence of this raid was that it drew Nimitz’s attention to the threat. Nimitz asked Layton how the Japanese had managed to drop four bombs on Oahu (the four that fell into the harbor had disappeared altogether, and no one was even aware of them until after the war). Layton was fairly sure that they had done it with seaplanes refueled from submarines, and he told Nimitz about Jasper Holmes’s story in the Saturday Evening Post. Layton also speculated that the Japanese had used French Frigate Shoals to refuel. As a result, Nimitz stationed an American seaplane at French Frigate Shoals, sending the USS Ballard, a destroyer recently converted to a seaplane tender, there in late March.¹⁶

For a variety of reasons, the Japanese did not continue their planned series of raids on Hawaii, but when Yamamoto sought reassurance that the American carriers were still in Pearl Harbor on the eve of the Battle of Midway, his staff suggested a reprise of Operation K. Again Hypo was able to alert Nimitz to the Japanese plan. On May 10, Layton informed Nimitz about an intercepted message that referred to the “K campaign” involving both aircraft and submarines, and three days later he reported that “the K campaign [was] underway.”¹⁷

The Japanese committed six submarines to the project: two filled with aviation fuel, two as radio beacons, one as a plane guard, and one as a command boat. The first of them, the I-123 commanded by Lieutenant Commander Ueno Toshitake, arrived at French Frigate Shoals on May 26. When Ueno approached the atoll and peered into the lagoon through his periscope, he saw a U.S. Navy warship anchored there. When the two fuel-laden submarines showed up the next day, the American warship was still there. In fact, another converted seaplane tender, the Thornton, had joined her. The submarines were in no position to challenge them—the Type KRS submarine had been designed as a minelayer and did not have torpedoes or torpedo tubes. A surface attack would be suicidal, since each of the American surface ships boasted four 4-inch guns. Besides, the whole point of Operation K was stealth. The Japanese could only hope that the Americans would simply go away. Ueno radioed the circumstances back to his superior in the Marshalls and received orders to wait one more day. On May 31, several Catalina PBYs landed in the lagoon to join the tenders. Informed of this, Vice Admiral Tsukahara Nishizo cancelled Operation K. There would be no reconnaissance of Pearl Harbor before the Battle of Midway; the Japanese would simply have to trust that the American carriers were still there. Of course, the day before that, on May 30, the Yorktown had left Pearl Harbor to join Task Force 16 at Point Luck.¹⁸

The second thing that went wrong that week was that the Japanese were tardy in establishing the submarine cordons that were supposed to track the American carriers as they left Pearl Harbor in response to an attack on Midway. Seven submarines, constituting Cordon A, were to occupy a north-south line west of Pearl Harbor. Six more would constitute Cordon B north and east of French Frigate Shoals. Another six would occupy a line near Midway. The subs were to report the carriers’ movements and then inflict whatever damage they could as a prologue to the main event. All three cordons were to be established by June 2. They got a late start out of Japan, however, and also lingered a day in Kwajalein, so that they were late in arriving. In addition, several subs were delayed by their involvement with the aborted Operation K. As a result of all this, only one sub made it into position by June 2; the others did not arrive until June 4. By then, the American carriers were nearly a thousand miles to the north. Watanabe Yosuji, Yamamoto’s loyal logistics officer, blamed the submarine commander Captain Kuroshima Kameto. Watanabe insisted that Kuroshima was simply not energetic in pursuit of his duties. Whatever the merits of that assertion, Yamamoto and Nagumo steamed eastward unaware that the American carriers—their principal quarry—had already flown the coop.¹⁹

On June 2, Yamamoto’s battleships and Nagumo’s carriers, fighting their way eastward through rough seas, were blanketed by a fog so thick that the ships had to use searchlights to find one another in the formation. On the one hand this was a stroke of luck, for it hid them from the prying eyes of American long-range search planes from Midway. On the other it also prevented Nagumo from sending out search planes of his own, and it was stressful for the entire formation to execute the required zigzag course (to confuse American submarines) while maneuvering through a fog. A witness on board Akagi recalled seeing Nagumo and members of his staff on the bridge staring “silently at the impenetrable curtain surrounding the ship, … each face tense with anxiety.” Nagumo may indeed have been anxious. He had heard nothing from the submarines other than one report from I-168 off Midway, which relayed the information that, although the Americans were conducting intensive air search operations, the only vessel in sight was a picket submarine off Sand Island. Nagumo had to assume that no news was good news.²⁰

In fact, of course, the Americans were very much on the alert. On May 29, Nimitz had designated Navy Commander Logan C. Ramsey as the operational air coordinator for the airplanes at Midway, and Ramsey dramatically stepped up both the frequency and the range of the air search patrols. There had been some discussion within the American high command about Ramsey’s authority to send Army B-17 Flying Fortress bombers on such missions. Just as Theobald and Butler quarreled over their respective roles in defending Alaska, Army and Navy leaders at Midway squabbled over whose job it was to search for enemy warships. The Army insisted that the B-17s should be reserved for combat missions. The Navy’s position was that (in the words of one admiral) it was “criminal waste and stupid folly” not to take advantage of their two-thousand-mile range for air search missions. The discussion made it back to Washington, where George Marshall decided in favor of the Navy. As a result, the Americans were able to use not only the Navy PBY sea planes but also the heavy B-17 Army bombers in their air search pattern.²¹

The work was tedious. Every morning before dawn, the PBYs took off one by one at five-minute intervals from Midway’s protected lagoon while the B-17s took off from the airstrip on Eastern Island. They flew for seven or eight hundred miles out on their assigned vectors, then flew back again along a different axis to cover more area. After ten to twelve hours in the air, the crews landed, secured their planes, ate, slept, and then got up the next morning before dawn to do it all again. They flew mostly at low altitude—around 1,000 feet. That narrowed their search area, but the visibility was better and they were less likely to make mistakes in identification.²²

The first sighting report came in at 9:00 a.m. on June 3 from Ensign Charles R. Eaton, piloting a PBY about five hundred miles west of Midway. Eaton reported seeing “two Japanese cargo vessels” that fired on him with antiaircraft fire. Back in Midway, Captain Simard concluded, correctly, that these were only minesweepers patrolling ahead of the main body. Only minutes later, another report from a different search plane electrified the listeners at Midway, at Pearl Harbor, and out at Point Luck where Fletcher, Spruance, and the American carriers lay in wait.²³

The report came from Ensign Jewell Reid, flying another PBY out of Midway. Near the end of his plotted search area, some seven hundred miles west of Midway, he saw some tiny specks on the horizon. At first he thought it was dirt on the windscreen. His copilot snatched up the binoculars and stared out the windscreen. Reid postponed his turn for home and maintained his course. As the range closed, he saw that they were indeed ships—many ships. At 9:05 he sent the message: “Sighted main body.”²⁴

Back at Midway, Ramsey ordered Reid to amplify his report. Already near the edge of his plane’s maximum range, Reid dived toward the water, stayed low, and completed a wide circle out to the north. At about 9:30, he eased his plane up to about 800 feet and peered southward. At 9:35 he sent in a more complete report: “Six large vessels in column.” Again Ramsey asked for clarification: What kind of ships? What course? What speed? To get that information, Reid headed back to low altitude and maneuvered around behind the formation. He reasoned that the lookouts on the Japanese ships, which still lacked radar, would more likely be searching forward than aft. With the sun behind him, he crept back up to 800 feet to have another look. Finally he was able to deliver the information that Simard, Nimitz, and Fletcher needed: “Eleven ships, course 090 [due east] speed 19 [knots].” The formation included “one small carrier, one seaplane tender, two battleships, several cruisers, several destroyers.” He also requested instructions. By now, he was well past his optimum turnaround time for fuel use, and an entire Japanese fleet was between him and his base. His crew was therefore much relieved when the radio crackled out permission for him to return to Midway.²⁵

 Ensign Jewell “Jack” Reid (perched on the wheel strut) and his PBY crew were the first to make a visual sighting of the approaching Japanese force on June 3. (U.S. Naval Institute)

At Pearl Harbor, Nimitz was engaged in conversation with Layton when Arthur Benedict, who had just gotten off watch at Hypo, came running in waving a piece of paper. It was a copy of Reid’s sighting report. Nimitz had maintained his usual placid public demeanor through the past several days, though he confessed privately in a letter to his wife that his days were full of “anxious waiting.” Based on the initial Hypo intercepts, he had expected the enemy to begin the attack on June 3, and the absence of any sighting reports had been worrisome. Now, as he read Ensign Reid’s report, his weathered face broke into a broad grin. This must be Kondō’s “Invasion Force,” the “bait” that was supposed to lure the American carriers to their doom. Its composition was exactly what Rochefort had predicted, and it was almost exactly where Rochefort had said it would be. Nimitz handed the report to Layton. “This ought to make your heart warm.”²⁶

Nimitz knew, however, that this was not the “main body,” as Reid had reported it. Technically, the “main body” was Yamamoto’s battleship force, which the Americans still did not know about, though the real target—the key piece in the entire puzzle—was the enemy’s carrier force, the Kidō Butai, and so far there was no word as to its whereabouts. Aware of that, Nimitz decided to forward Reid’s report to Fletcher even though he was certain that Fletcher’s own communications team had monitored it. He did so because forwarding the message allowed him to add his own comment at the end. “That is not, repeat not, the enemy striking force,” Nimitz wrote. “That is the landing force. The striking force will hit from the northwest at daylight tomorrow.” Nimitz did not want to micromanage his operational commanders, but neither did he want them to go off half-cocked. At Point Luck, Fletcher was still well beyond striking range of this target, and the subtext of Nimitz’s forwarded message was unmistakable: Wait. Be patient.²⁷

Kondō’s force was, however, a perfectly appropriate target for the Army’s heavy bombers on Midway. If Fletcher’s position at Point Luck was still a secret, the location of Midway was never a secret, so launching an air strike from the atoll gave nothing away. As soon as word could be sent to the airfield, nine Army B-17s under the command of Colonel Walter Sweeny took off from Eastern Island and headed west to strike the first blow. It took them most of four hours to find the Japanese, and when they did it was not Kondō’s “Invasion Force” but the nearby “Transport Group” under Rear Admiral Tanaka Raizō, consisting of one light cruiser, ten destroyers, and thirteen transport ships filled with the 5,000 men of the landing force. Tanaka had outrun his air cover, and as a result all he could do now was try evasive maneuvers while his destroyers threw up as much antiaircraft fire as they could muster.*

The big American bombers dropped their ordnance from 10,000 feet. Each B-17 carried four 600-pound bombs—thus a total of nearly eleven tons of bombs fell among the ships of Tanaka’s command. The Japanese ships maneuvered radically under the rain of ordnance, most of which exploded when it hit the water. The flash of the explosions, the enormous geysers of water they generated, and the black smoke from the Japanese ships as they twisted and turned in the roiling water all looked pretty spectacular from 10,000 feet. Making accurate damage assessments is difficult in the best of circumstances, and especially so for Army pilots untrained in antiship operations. The returning pilots did the best they could. They reported five hits, one probable hit, and four near misses against two battleships and two large transports. Sweeny reported that one transport was on fire and that a battleship was on fire and sinking. Based on that report, American submarines were vectored toward the site to finish off the damaged battleship.²⁸

In fact, there were no battleships in Tanaka’s group, only a light cruiser and several destroyers, and none of them had suffered any damage. Despite all the sound and fury, no ship had been hit; no one, on either side, had been injured.

Nevertheless, the apparent success of the raid inspired Rear Admiral Patrick N. L. Bellinger, commander of PBY Patrol Wing Two, to attack as well. Bellinger was a career aviator who had sent the famous radio report that had informed the world of the Japanese attack back in December: “Air Raid, Pearl Harbor—This is no drill.” Now, eager to retaliate, he devised a way to use his PBYs to strike at the foe in a night torpedo attack. He had four of his Catalinas modified to carry the heavy Mark 13 torpedo, and he called for volunteers to fly them out to attack the enemy. As Gordon Prange remarked forty years later, this was an idea “straight out of a comic strip,” but it illustrated the American willingness and ability to improvise.²⁹

More important, it worked. Despite the darkness and the range, the four Catalinas, led by Lieutenant William L. Richards, actually found Tanaka’s “Transport Group.” At about 1:00 a.m. Richards saw “what appeared to be [an] endless line of Japanese ships” silhouetted against a bright full moon. He told his crew, “Drop that damn thing and let’s get the hell out of here.” They did, and even scored a hit—the first of the battle—on the small tanker Akebono Maru. The Japanese antiaircraft guns opened up, and, in the words of one witness, “it was like the fourth of July with tracers coming through the plane.” The PBYs escaped without significant damage, though the pilots had trouble staying awake during the long flight back to Midway. “One of us would fly the plane,” a crewmember recalled, “and the other would whack him in the face to keep him awake.” Their torpedo strike had killed eleven men and wounded thirteen more, though the damage to the tanker was limited. One Japanese officer concluded that the torpedo’s warhead had failed to detonate (not unusual with the Mark 13 torpedo), for the Akebono Maru stayed a float and even managed to maintain her place in the convoy.³⁰

These air attacks on Tanaka proved that the Americans now knew of the approach of the Japanese. Since there seemed little reason to maintain radio silence any longer, Kondō reported the attacks to Yamamoto. The commander in chief, however, did not pass the information on to Nagumo. For one thing, Yamamoto was maintaining radio silence so that the Americans would not be aware of him. Moreover, he very likely assumed that Nagumo was paying attention and had heard the report himself. As it happened, Nagumo had not, and he remained unaware that the Americans were alerted, or that they had already dispatched two air strikes to contest the approach of the transport force. Even had he known, it is not clear it would have made any difference. After all, the American air attacks had proved fruitless. The various elements of the Japanese armada remained essentially unharmed, and they continued to close in on Midway in accordance with the operational timetable drafted by Combined Fleet back in April. At noon on June 3, the Kidō Butai, still enshrouded by the convenient weather front, altered its course from east to southeast and increased speed to 26 knots to close on its predetermined launching point for a dawn strike the next morning. For their part, the American carriers at Point Luck moved westward at an easy fourteen knots in anticipation of an imminent sighting by the patrol planes from Midway. The real battle was yet to be joined.³¹

The American forces at Point Luck operated two hours ahead of the clocks on Midway. The ships of the Kidō Butai were still on Tokyo time, their clocks twenty-one hours ahead. Nevertheless, as all the clocks moved toward midnight on the ships on both sides that cleaved the waters of the Pacific Ocean, officers and enlisted men prepared for the changing of the watch: in the engine room, in the radio shack, on the topside lookout, and on the bridge. It was a ritual that had taken place a hundred thousand times before, and would a hundred thousand times again. By tradition, the man coming to assume the watch arrives early. On the bridge, he familiarizes himself with the ship’s course and speed and all other pieces of pertinent information. When he feels he has a firm grasp of the circumstances, he salutes the officer of the deck and pronounces the words that make him responsible for the ship over the next four hours: “I relieve you, sir.” His watch will last until 4:00 that morning when another officer will appear to take his place, but for now, at this moment, he is driving the ship. The night slips silently past. With the ship blacked out and all but invisible, it seems to exist in a world of its own. Of course, it is actually part of a complex pattern that stretches over half of the central Pacific, with more than 150 ships—carriers, battleships, cruisers, destroyers, transports, supply vessels, and submarines—each moving at its own speed toward its destiny.

In military time, 2400 became 0001 as a new day began. To the Japanese, operating on Tokyo time, it was now June 5, but for the Americans, it was now officially June 4,1942.

* After the raid, one Japanese Zero pilot attempted to land his crippled plane on a nearby island. His wheels stuck in the spongy tundra, and the abrupt landing broke his neck. The plane, however, was barely damaged, and five weeks later it was recovered by Americans and sent back to the States, where it was repaired and flight-tested. That helped American aircraft designers assess its strengths and weaknesses.

* There is some uncertainty concerning the identification of Kondō’s and Tanaka’s units. Some students of the battle assume that Sweeny’s B-17s struck at the same force that Reid had reported that morning. However, the makeup of each force, as well as the Japanese battle reports, suggest that while Reid saw Kondō’s “Invasion Force,” which included two battleships (Kongo and Hiei) as well as four heavy cruisers (“Six large vessels in column”), Sweeny’s B-17s actually struck at the “Transport Force” guarded by one light cruiser and ten destroyers.