Japanese Naval Aircraft and the Japanese Aircraft Industry, 1937–1941
I have been discussing the controversies that took place in the 1920s and 1930s over specific technologies, armaments, and tactics on the way to the creation of a modern air service within the Japanese navy. But it is important to keep in mind that during the interwar period these arguments took place against the background of a broader and intensifying debate within the navy as to the appropriate place of aviation in naval warfare generally. As in the United States Navy, the lines in this debate were drawn between the majority of officers in the fleet and the high command who held that the main batteries of the battle line remained the final arbiter of victory at sea, and the much smaller group of air power enthusiasts who argued that the range, speed, and destructive capability of the airplane had overturned the supremacy of the battleship. In charting this debate and its influence on the development of Japanese naval aviation, two facts must be kept in mind. The first point has been mentioned earlier in this study: In Japan, as in other nations, the capacities of air weaponry were consistently exaggerated by its advocates during this period. In particular, governments assumed bomb-delivery capabilities by aircraft that were not to exist until World War II. In its initial phase, therefore, the aircraft-versus-gun debate was between visionaries who claimed for the airplane all the things it could not yet do, and realists who noted the enormous disparity in capabilities between air and surface units.
Source: Fukui, Shashin Nihon kaigun zen kantei shi, 1:328.
The second point to note is that in its later phase, the debate was not between a blindly obdurate battleship faction and a small clutch of airmen struggling to find at least some naval role for aviation. Rather, it involved an argument between a battleship-oriented naval orthodoxy now sensitized to the potential of air power and a group of air power radicals who believed that all future naval warfare would be decided by aviation. The former held that the offensive power that aviation had by then acquired should be exploited to ensure victory by the surface fleet. The latter, centering their vision largely on the range and striking power of land-based heavy aircraft, advanced “the theory of air power omnipotence” (kōkū bannō ron), which called for the scrapping of all capital ships.
THE NAVAL AVIATION CONTROVERSY
In a sense, the opening shot in the argument had been fired at the very dawn of Japanese naval aviation when, in 1915, Nakajima Chikuhei, then a young engineer lieutenant, had written the air power declaration outlined in the first chapter. Yet as I have shown, in 1914 the prodigious technological gap between air power theory and air power fact made Nakajima’s manifesto the statement of a visionary, not a realist. Over a decade later, this gap still hampered aviation proponents. Speaking at the Kasumigaura Naval Air Base in 1927 before a group of General Staff officers, Lt. Comdr. Kusaka Ryūnosuke,† destined to be one of Japanese naval aviation’s most important figures, ticked off some of the advantages provided by aircraft in naval combat. Aircraft could operate over a wider combat area in a shorter period of time than surface craft. Aviators had a wide range of vision. Aircraft were difficult to shoot down because of their small size and speed. Aircraft in the attack were capable of dealing out psychological as well as physical damage. Yet, admitted Kusaka, the limitations of most aircraft in naval warfare were obvious: they could not stay in the air for long periods of time, they had difficulty operating with the fleet over great distances, their operations were at the mercy of the weather, and their bomb-carrying capacity was too limited to cause substantial damage to an enemy fleet.1
To an extent, of course, the development of carriers and their aircraft solved some of these problems, but at a time when the Japanese government and armed services were concerned about the possibility of an enemy bringing his aircraft carriers close to Japanese shores, the range of those land-based aircraft then in service seemed insufficient. Writing in the naval journal Yūshū early in 1930, Comdr. Matsunaga Toshio,† then executive officer on board the Akagi, argued for a larger land-based attack aircraft that could seek out and destroy an approaching enemy fleet while it was still far out at sea. Had Japan such bombers, capable of ranges of several thousand miles, possessing greatly increased bomb loads, based at strategic locations around the home islands, and supplemented by two or three aircraft carriers, these could prevent any enemy invasion force, even enemy carriers, from getting through, Matsunaga insisted.2
Rear Adm. Yamamoto Isoroku, who had become head of the Technology Bureau of the Naval Aviation Department in 1930, voiced similar arguments about the necessity of developing a new long-range capability for naval aviation. Yamamoto now had both the vision and the influence not only to make land-based air defense of the home islands a reality but to push it out into the far reaches of the Pacific. At the time, the small carrier aircraft possessed by both the Japanese and American navies were generally equal in the ranges at which they could seek and attack an enemy. Yamamoto, looking ahead to the day when Japan would be able to use its then demilitarized islands in Micronesia to create a chain of bases across the central Pacific, saw the tremendous “outranging” advantage of a long-range attack aircraft operating from those bases. Shifting from one island base to another, such a flexible bomber attack would upgrade the Japanese navy’s attrition strategy against the larger westward-moving American fleet. The Navy General Staff had also recognized the increasing value of Micronesia as a base for operations against the United States Navy and had ordered the design of the largest flying boats for use in the islands. But judging from the results of aeronautical engineering research, it became clear that land-based bombers were more likely to satisfy the navy’s operational needs than flying boats.3
With this concept in mind, Yamamoto initiated plans for the design of a land-based all-metal twin-engined monoplane with a range of 2,000 nautical miles and a 907-kilogram (2-ton) bomb load. Because of its experience in developing large all-metal flying boats, the Hirō Naval Arsenal was given the responsibility for developing the aircraft. The result was the Hirōshō G2H1, the largest naval aircraft of the time that was designed from the beginning as a land-based bomber. Mounting two water-cooled engines that were the most powerful available, the G2H1 was completed in April 1933 and, in Yamamoto’s presence, made its first flight over Yokosuka that May. Further tests at Kasumigaura made it clear that in many respects the aircraft’s capabilities fulfilled the Naval Aviation Department’s requirements, and in 1936 it was adopted by the navy as the Type 95 land-based attack aircraft. But in the meantime, continued evaluation had demonstrated that the aircraft’s engines were unreliable and actually underpowered for its size, and that its massively strutted fixed landing gear made it slow and cumbersome. Only eight G2H1s were therefore produced, and it was always regarded as a second-line aircraft.4
Yet the concept of a long-range land bomber inherent in production of the G2H1 did not die; indeed, Yamamoto was casting around for the development of another aircraft even as the G2H1 was being tested. As luck would have it, the army had also been experimenting with all-metal multiengined aircraft and had brought engineers from the Junkers firm in Germany to work with Japanese designers at Mitsubishi in developing a line of such aircraft that would incorporate the latest design features. Yamamoto recognized that the pool of technical skill now available at Mitsubishi was unrivaled anywhere else in Japan. He used his relentless energy and determination to press for assigning to Mitsubishi, on a noncompetitive basis, a contract for designing and developing a new reconnaissance plane. It would be capable of long overwater flights to determine American fleet movements in the Philippines and the Hawai’ian Islands. The aircraft was to surpass the G2H1 in performance but to be generally based on its airframe and engine type. In April 1934 Mitsubishi completed a single prototype of such an aircraft, the Ka.9, retroactively designated the G1M1. Its capabilities—an astounding range of 3,265 miles and a maximum speed of 165 knots, made possible by more powerful water-cooled engines (replaced, in later models, with radial engines) and a retractable undercarriage (the first in any Japanese aircraft)—delighted Yamamoto, who personally went up in the aircraft after its preliminary test flight. Further testing indicated that, more than being just a defensive long-range reconnaissance aircraft, the G1M1, if further developed as an offensive long-range bomber, had the capacity to give reality to the vision of Yamamoto and other air power advocates. Thus was born the first of the land-based twin-engined bombers, which led to the development of the G3M (see below).5
If the appearance in the mid-1930s of the land-based long-range bomber strengthened the arguments of the air power supremacists in the Japanese navy, their expectations were paralleled by the hopes that the navy’s so-called teppō-ya (“gun club”) placed on plans for the super-battleships Yamato and Musashi. It was the conjunction at middecade of these two opposing weapons systems—neither one was yet operational—that fueled the sudden aircraft-versus-gun debate in the Japanese navy, a controversy that was little reported in the West at the time and has been largely overlooked by Western naval historians in the decades since.6 The debate was the result of two proposed solutions to the same problem: the numerical inferiority of the Japanese battle fleet in comparison with that of the United States. Japanese naval war games and maneuvers during these years had confirmed once again the inevitable victory of superior numbers in a collision between forces of similar composition. The solution of the “gun club” to this tactical dilemma was the construction of capital ships far more powerfully armed and armored than those of the enemy. The solution of the air power advocates, on the other hand, was to change drastically the composition of Japanese naval forces by making carriers the navy’s main force, so as to circumvent and render irrelevant Japan’s inferiority in capital ships.7
Among the first of the naval air staff to seize upon the appearance of the large land-based bomber as the harbinger of air power supremacy was Capt. Ōnishi Takijirō,† destined to become one of the navy’s most forceful air power advocates, a prewar confidant of Yamamoto Isoroku, and the architect of the navy’s desperate kamikaze strategy at the end of the Pacific War. Beginning with an assignment to the seaplane carrier Wakamiya about the time of Nakajima’s circular letter on naval air power, which Ōnishi seems to have read with great enthusiasm, his early career had been mainly devoted to naval aviation, both as a line and a staff officer. In 1930, as a member of the Training Bureau of the Naval Aviation Department, he had written a memorandum calling for a drastic overhaul in the navy’s aviation structure and policies, including, among other things, an expansion of the number of land-based air groups. Now, four years later, as executive officer of the Yokosuka Air Group, Captain Ōnishi drafted an essay on naval air power that was far more strident. It called for the scrapping of all capital ships and a placing of the navy’s offensive capabilities mainly on aircraft. Repeating the arguments that Nakajima had used nearly twenty years before, Ōnishi inveighed against the stupidity of the rumored General Staff plans to construct super-battleships, claiming that the cost of one such monster could pay for a thousand top-of-the-line fighter planes. Conversely, Ōnishi applauded the advent of the new land-based bomber, declaring with more enthusiasm than evidence, “It is already dangerous for surface units to enter within the range of such a bomber group. Because battleships are fragile under enemy attack, it is wrong to make them the navy’s main force. For our naval armament we should shift from battleships to land-based air power.”8
Photo. 4-2. Capt. Ōnishi Takijirō (1891–1945)
Other, younger, air power advocates, restless under the battleship orthodoxy of the navy’s high command, began to weigh in with similar opinions. Asked to write on “Naval Armament Essential for the Effective Prosecution of War with the United States” as part of his study at the Naval Staff College, Lt. Comdr. Genda Minoru drafted an essay that proposed that the navy center its efforts on land- and carrier-based aircraft, with cruisers, destroyers, and submarines to be used in a supporting role. As for battleships, Genda wrote that they should be either scrapped or used as hulks for jetties. As he had been warned by Ōnishi when both were assigned to the Yokosuka Air Group, such ideas provoked a storm of criticism from gunnery and torpedo officers, and word went around the Staff College that Genda was a little crazy.9
For his part, Yamamoto regarded the plans for the projected battleships as the height of folly. At first he confined his opposition to sarcastic references to battleship orthodoxy in conversations with his colleagues and subordinates in naval aviation. About 1934, when he commanded the First Carrier Division, he remarked to a group of young pilots, “Even though the thick-headed teppō-ya have modified their outlook somewhat, they still don’t grasp the realities of air power, so you young men will have to renew your efforts in training and study [in order to convince them].”10 The next year, as head of the Naval Aviation Department, he disparaged the “big-gun” dogma with even greater ridicule:
As you know [he told a group of naval aviators at the Yokosuka Air Group], it is the custom for wealthy families to display splendid art objects in their living rooms. Such objects have no practical value, yet such households derive a certain prestige from them. Similarly, the practical value of battleships has declined, but the world’s navies still set great store upon them and they retain their symbolism as an indicator of naval power. You young airmen shouldn’t insist on the abolition of the battleship, but rather you should think of it as a decoration for our [navy’s] living room.11
But by this time the super-battleship project was under serious discussion as part of the “Circle Three” program. For that reason Yamamoto attempted to take the case for air power directly to the project’s powerful backers in the navy’s upper echelons. At the talks preliminary to the London Naval Conference of 1930, despite his misgivings, Yamamoto had followed his instructions and insisted upon parity in cruiser tonnage for Japan. But upon his return from London, he had met with Vice Adm. Koga Mineichi,† chief of the General Staff’s Intelligence Division, to register his objections to the plans for the construction of the super-battleships. Koga had earnestly defended the project, noting the apparent lesson of recent French naval policy. France had been one of the first major naval powers to start building aircraft and submarines in lieu of capital ships, Koga pointed out, but once it started doing so, France had dropped to a second-rate naval power.12 In vain Yamamoto tried to convince Capt. Fukuda Keiji of the Navy Technical Department, who had been given the responsibility for coordinating the design plans for the monster battleships, of the ultimate vulnerability and obsolescence of such warships. Despite his admiration for Yamamoto, Fukuda, fired by the technological challenge involved in designing ships of this unprecedented size, countered with explanations of their planned impregnability. Yamamoto ran into even more adamant resistance in his arguments with Vice Adm. Nakamura Ryōzō, chief of the Navy Technical Department and one of the super-battleship project’s most determined architects. Fruitlessly Yamamoto sought to demonstrate with detailed figures that the navy would gain far greater offensive power by spending the same amount of money on naval aviation.13
To an extent, of course, the resistance of the navy’s high command to criticisms of its super-battleship program reflected an irritation at having its collective wisdom challenged. When Adm. Takahashi Sankichi, commander of the Combined Fleet and before that commander of the First Carrier Division, stepped around to the General Staff and the Navy Ministry to express the view that perhaps the navy should concentrate on building its air power rather than building bigger battleships, his suggestion met with exasperation. “When Suetsugu Nobumasa held your command,” he was told, “he wanted us to emphasize submarines because he was a submarine man. Now, because you used to be head of a carrier division, you want us to plump for more aircraft. We simply can’t have Combined Fleet commanders giving vent to such personal biases.” Despite Takahashi’s lofty command and the fact that he still had an open mind on the air power–versus–sea power issue, his opinions cost him the trust of those overseeing the planning and construction of the super-battleships. From that time on he was cut off by the General Staff from further information about the project, an example of the arrogance of the navy’s bureaucratic fiefdoms.14
But the high command’s rejection of the arguments of Yamamoto and the other air power advocates was not just a matter of upper-echelon hubris. “Gun-club” officers believed that there were solid criticisms to be made of the concept of “air power omnipotence.” As they pointed out, aircraft were still unreliable machines, affected as they were by weather. Moreover, aerial attack had as yet failed to sink a battleship under way. Conversely, they pointed out, both the offensive and defensive power of capital ships had increased: new techniques in spotting the fall of shot had supposedly doubled the accuracy of battleship main batteries; and the number and effectiveness of ship-borne antiaircraft weapons had increased substantially.
It is, of course, the responsibility of military professionals not only to assay what is tactically necessary today but also to give thought to what may be tactically possible tomorrow. The state of technology and the rapidity of its progress are the variables in their capacity to do so rationally. These serve to set the military visionary of yesterday apart from the clear-headed military prognosticator of today. When Nakajima Chikuhei sent around his circular in 1915, the feeble capacities of aircraft lent an air of fantasy to his ideas. We can now see that by the mid-1930s, when Yamamoto Isoroku argued for the primacy of air power, the gap between air power theory and technological reality was closing with a rush.
Nevertheless, by the evidence available at middecade, the navy’s battleship orthodoxy seemed to have the greater logic in the controversy. The problem with Yamamoto’s arguments was that in the mid-1930s they still comprised prediction, not fact. Dive-bombing techniques and aerial torpedo tactics were still being worked out, and the new twin-engined bomber had yet to be tested in any sort of combat, let alone employed to attack moving targets. Moreover, compared with the radical theories of air power advocates like Ōnishi and Genda, who wanted to scrap all capital ships, the mainstream view, which argued for a balance of air and surface forces, seemed to the high command to be more rational. The mainstream view also seemed clearly confirmed by the adherence of the British and American navies to such a balanced force structure and doctrine. In any event, the weight of the high command was behind the plans to construct the super-battleships Yamato and Musashi. Thus, air power advocates like Yamamoto were unable to halt the project or to divert the same resources to strengthening the navy’s air power. In the years ahead, Japanese naval airmen would argue with increasing confidence for the primacy of air power, and a few even demanded its complete independence from the two established services.
The issue of an independent, unified air force had been raised as far back as the early 1920s, largely at the initiative of army airmen. A joint army-navy committee had been established at that time to look into the merits of creating such a force. Nothing came of the concept because both services remained essentially opposed to it, but in the mid-1930s, with the great strides occurring in aircraft capabilities, with the increasingly vocal claims being made in each service for the primacy of air power, and with the emergence of independent air arms in major European countries to serve as a model, the idea was brought up once again, initially by advocates in the army. The concept created such a storm of controversy in the navy that the Naval Aviation Department established another committee to look into the matter. In July 1937 the report of the committee, drafted by Ōnishi Takijirō, concluded that the idea was essentially unsound, and the scheme was dropped. Essentially, the navy’s opposition was based on two grounds: first, the suspicion that any unified air force would be dominated by army personnel and by army tactical considerations, and second, the belief that an independent air force would substantially weaken the navy’s own air power, seen as critical to the navy’s surface operations. The latter supposition seemed to be confirmed by the example of the weakening of the British navy’s air service after the creation of the Royal Air Force.15 So Japan, like the United States, entered World War II with its aviation divided between its two major services. Of course, the United States Army Air Forces came to acquire far greater autonomy than its Japanese army counterpart, and the large land-based air fleets of the Japanese navy (see chaps. 6 and 7) became more autonomous than American naval air forces, tethered as the latter were (except for Marine Corps air units) to their carriers.
JAPANESE NAVAL AIRCRAFT DEVELOPMENT ON THE EVE OF THE CHINA WAR
As we have seen, the development of an effective land-based long-range bomber had been the first of Yamamoto’s priorities. The impressive capabilities of the Mitsubishi G1M1 aircraft had provided a real basis for Yamamoto’s conception of such a bomber. Completed at Mitsubishi’s Nagoya plant as a shore-based reconnaissance aircraft, it had been tested at the Mitsubishi airfield at Kagamigahara in April 1934. The results, surpassing the navy’s performance requirements, had created a sensation in Japanese naval air circles. As a result, the navy issued specifications for a new long-range attack aircraft. Mitsubishi tested the first prototype in July 1935, and after a series of modifications, the Navy Air Arsenal finally adopted it in June 1936 as the Navy Type 96 attack plane, G3M1 (known to Allied air forces during the Pacific War as the “Nell”). The Japanese dubbed it chūkō, a contraction of an unofficial designation for the plane, chūgata kōgeki-ki, which meant “medium attack plane.” The plane was also called rikkō, from a contraction of a part of its official title, rikujō kōgeki-ki, “land-based attack plane.” It was a sleek, all-metal monoplane medium bomber whose slender shape was retained by excluding any internal bomb bays from the design, its 800 kilograms (1,764 pounds) of bombs or a single torpedo being slung from racks fitted beneath the fuselage.16 Powered by two air-cooled radial engines, the chūkō was armed with two machine guns mounted in dorsal turrets and one in a ventral turret.17
Once the chūkō had been accepted, the Japanese navy had to train a far greater number of aircrews (pilots, observers, and radio operators) and ground crews, since the new bomber required more personnel to man and maintain than any previous navy land- or carrier-based aircraft. The Tateyama Air Group was the first unit to receive the new airplane, and distribution to the Kisarazu and Kanoya naval air groups soon followed. When the G3M1 first went into service, the navy made it a practice to draw its crews from the most skilled and experienced of those who flew its carrier attack planes, men with seven to ten years in the navy and over a thousand hours of flight time with the fleet. In the coming first months of Japan’s air war in China, these airmen were to constitute the vital cadre of the navy’s medium-bomber forces on the continent. All through the spring of 1937, chūkō crews operating from these bases undertook intensive training in long-distance operations during flights to the Bonins, Saipan, and other overwater destinations.18 In range, speed, payload, service ceiling, and state-of-the-art radio equipment, the G3M1 was an unprecedented Japanese aircraft. When it appeared over Chinese skies in the late summer of 1937, it took international aviation circles by surprise and led some foreign observers to assume that it was simply a Japanese version of the German Junkers 86 bomber. In fact, the G3M1 was superior to the Junkers in every respect and at the time was surpassed only by the prototype of one of the greatest heavy attack planes ever built, the Boeing B-17. The skill of the aircrews who flew it, when added to the superb performance of the aircraft itself, created crack air groups possessed of tremendous morale and energy, convinced of their capacity to wreak horrific destruction, and unshaken in their belief that no fighter plane could touch them. In time the chūkō would prove to be a flawed aircraft, but its near-fatal defects—absence of protective armor, inadequate defensive armament, and unprotected fuel tanks—would not be revealed until it had been tested in the fire of combat over China.19 Even before the war, its crews worried about the fuel tanks catching fire, and from the first day of the raids on China their suspicions were confirmed, the bomber bursting into flames almost every time it was hit. While it must be remembered that at this time few military aircraft in the West had significant armor protection or were equipped with self-sealing gas tanks, it is testimony to the obduracy of the Japanese navy’s air leaders, or to their willingness to sacrifice aircrews, that these problems persisted in later navy bomber designs.
The appearance of the G3M1 medium bomber in the spring of 1937 coincided with the apogee of antifighter sentiment in the Japanese navy. The new bomber’s performance, particularly its speed relative to all operational Japanese fighter aircraft, seemed to underscore the impotence of the latter in the face of such capabilities. Bomber advocates like Lt. Comdr. Nitta Shin’ichi† believed that the fighter plane was no longer effective as either a defensive or an offensive aircraft. In his opinion it had become useless in an interceptor role because it could not maintain a superiority in speed, and it was of little use as an attack escort because it did not have the range to accompany the larger type of bomber.20
Yet in the months just prior to the China War, a revolutionary carrier fighter was coming on line that would restore the balance between fighters and bombers. Its origin lay in the search, early in 1934, for a new single-seat aircraft to replace the Nakajima A4N (Type 95) carrier fighter, a biplane that had from its inception been regarded as only a stopgap design. Even at a time when great strides were being made in aviation technology—in increased engine power and more streamlined configuration—the design of carrier aircraft posed particularly difficult problems, largely due to the inherent restrictions imposed by carrier flight-deck operations (including problems of visibility, landing speed, and the need to conform to the dimensions of the flight elevators). Given these restrictions, the design of a successful carrier fighter required a careful selection of performance priorities within the specifications set forth by the navy for any particular aircraft. Prepared by Lt. Comdr. Sawai Hideo of the Department of Engineering at the Naval Air Arsenal, the specifications for the new carrier fighter called for high speed, low weight, and excellent control.21
While air combat in China was to exercise an important influence in the design and development of certain aircraft, the specifications issued by the navy for its aircraft were nevertheless largely determined by the evolution of its missions at sea. This was especially true of fighter aircraft. Until about the time of the China War, the navy’s needs for carrier fighters had been limited to those of fleet air defense against enemy spotter planes and attack aircraft (a problem to be discussed in chap. 6). But around 1935–36, as part of its emerging concept of a preemptive strike against enemy carriers, the navy began to consider the use of fighter escort for such an operation. About that time, therefore, the performance requirements for the navy’s carrier fighters were significantly raised. Fighters no longer had only to drive off enemy reconnaissance craft but had to oppose enemy fighters, and this required a new level of performance. It was from this consideration, as much as from combat experience in China, that the navy’s two most successful fighters were designed and developed in the years immediately preceding the Pacific War.22
At Mitsubishi, one of the two firms selected to enter the design competition, Horikoshi Jirō, an experienced aeronautical engineer who ultimately gained a world reputation, was named to head the design team. While trying to meet all the navy’s requirements for the new aircraft, the Horikoshi group decided to place particular emphasis on speed and rate of climb. To maximize these, the team devoted intensive study to designing an aircraft with the minimum amount of air drag. Completed in January 1935, the product of this yearlong effort was a waspish-looking all-metal open-cockpit monoplane powered by a 500-horsepower radial engine and armed with two 7.7-millimeter machine guns mounted in the nearly elliptical wings. The careful design and smooth contouring of the aluminum monocoque construction of the fuselage, which included the use of flush rivets, made Horikoshi so confident that the aircraft would create the minimum amount of air resistance that he retained the use of fixed landing gear. While the retractable undercarriage was a recent feature favored among aircraft designers abroad, Horikoshi wished to avoid the extra weight and complex mechanisms it required.23
In flight tests at Kagamigahara in February 1935 the finished aircraft superbly justified the intensive effort the Horikoshi team had put into it. Attaining a speed of 243 knots, it exceeded navy specifications by 54 knots, and whereas the navy had required a time of climb to 5,000 meters (16,000 feet) of six and a half minutes, the Mitsubishi plane did that in five minutes, forty-five seconds. Since the tests also revealed some deficiencies in maneuverability, the incorporation of split flaps in the next prototype provided the sort of precise pilot control needed for dogfighting. Once out of Mitsubishi’s hands, the aircraft was given rigorous trials and evaluations by the test pilots of both the Naval Air Arsenal and the Yokosuka Air Group. By the time further modifications of wing and tail configurations were finished, the A5M, as the plane was designated, was an aircraft of unprecedented speed and maneuverability. In the fall of 1935, in a mock air battle carried out by the pilots of the “Genda Circus,” the A5M was tested against the latest models of various existing Japanese and foreign fighter planes and bested them in nearly all performance categories and tactical situations. In the autumn of 1936, the navy finally adopted the A5M as the Type 96 carrier fighter.24
The appearance of the A5M put a damper on the anti-fighter-plane sentiment within the Japanese navy. Even Capt. Ōnishi Takijirō, then chief of the Training Bureau of the Naval Aviation Department and the leading bomber enthusiast within the navy, spoke of deferring any decision on the utility of fighters until the A5M had been thoroughly evaluated. Its superb maneuverability, moreover, underscored the emphasis the navy seemed to have placed on dogfighting, as opposed to reliance on pure speed and firepower. This issue had been at the center of a heated controversy among the navy’s top fighter pilots in the first half of the decade, as discussed earlier. But most important, the appearance of the A5M carrier fighter, along with that of the G3M medium bomber, marked the culmination of Yamamoto Isoroku’s ambitious program of aircraft reequipment for the navy and signaled the entry of Japanese aviation into an era of self-sufficiency. In design, structure, and performance, the A5M was among the best fighters, and certainly the best carrier fighter, in the world at the time, and it gave Japanese aeronautical engineers the confidence that they could meet the highest standards.25
After the A5M was encountered by American pilots flying with the Chinese in their war with Japan (chap. 5), the fact that it was a Japanese-designed and Japanese-produced aircraft impressed American intelligence officers as much as the plane’s performance. Reporting to Washington, Claire Chennault wrote that flight tests against various Western aircraft confirmed that the A5M was “one of the best up-to-date pursuit airplanes in the world.” But, he emphasized, “the most striking significance about the Japanese Type 96 pursuit airplane is that not only is the airplane designed and constructed in Japan, but the motor, the metal sheets and tubing, the propeller, the instruments are made in Japan either under license or from their own designs.” He concluded, “Japan is self-supporting and independent in building airplanes.”26
JAPANESE NAVAL AIR TECHNOLOGY IN THE POST-TREATY ERA
Such was the accelerating advance of aviation technology that even as the A5M fighter was being distributed to the navy’s carrier- and land-based units in the spring of 1937, the Naval Aviation Department was initiating plans for a fighter design that would surpass it in performance. The specifications for the aircraft had been drafted by Lieutenant Commander Sawai of the Naval Air Arsenal, and the preliminary planning document for the aircraft had been drawn up in May after meetings between officials of the Aviation Department and representatives of the aircraft industry. But requirements for the new plane were afterward significantly shaped by the conditions of air combat in the first two months of the China War. In the opinion of the navy’s front-line pilots in that conflict, the navy needed a fighter with the speed and firepower to destroy an enemy bomber, the range and endurance to escort the navy’s G3M bomber on long-range missions, and the maneuverability to deal with any fighters it might meet on the way.
Based on these general requirements, the Naval Aviation Department issued to Mitsubishi and Nakajima specifications of unprecedented difficulty for a carrier fighter: a maximum level speed of 270 knots at 4,000 meters (13,000 feet), but a landing speed of less than 58 knots; a takeoff distance of less than 70 meters (230 feet); maneuverability equal that of the A5M fighter, but an armament of two 20-millimeter cannon in addition to two 7.7-millimeter machine guns. In any single category these requirements either surpassed or were the equal of those of the world’s outstanding fighter aircraft of the time. But what made the task of combining all these specifications within one aircraft so difficult was not the individual requirements but the fact that most of them were mutually incompatible. In particular, the obligation to incorporate a number of features, such as extra fuel capacity and heavy armament, for range and firepower, would inevitably increase the aircraft’s weight, while at the same time unprecedented speed and maneuverability were still required. Faced with these seemingly insoluble difficulties, the Nakajima company judged the specifications impossible and withdrew from the design competition.27
Not surprisingly, the Mitsubishi design team was headed by Horikoshi Jirō, who quickly realized that the basic problem was that the range and armament requirements called for a large and heavy aircraft, whereas the speed and maneuverability requirements argued for a small and light airplane driven by a light and powerful engine. Obliged to use existing Japanese power plant technology (and initially settling on an 870-horsepower Mitsubishi Zuisei engine), Horikoshi concluded that the solution to the problem lay principally in a relentless effort toward weight reduction, uncompromised success in aerodynamic design, and the inclusion of such innovations as would contribute significantly to satisfying any one requirement without materially affecting the others. Over the course of a year and a half, the Horikoshi team, meeting frequently with navy pilots, company representatives, and officials of the Naval Aviation Department, struggled to meet these objectives.
At one of these meetings, in January 1938, at the Yokosuka Air Arsenal, an intense argument broke out between two navy pilots with experience in air combat in China. The issue was the controversy over the priorities in fighter plane design described in chapter 2. Lt. Comdr. Genda Minoru argued that the most important capability for a fighter plane was its maneuverability and consequent close-in dogfighting ability. Lt. Comdr. Shibata Takeo countered that inadequate maneuverability could always be made up by pilot skill and that Japanese bomber losses over China made clear the need for escort by fast, long-range fighters. Although the Horikoshi team, in designing the new aircraft, attempted to satisfy both these priorities, judging from the structure and configuration of the final design the navy appears to have sided with Genda.28
The Horikoshi team solved the seeming incompatibilities in the design of the new aircraft one by one. A reevaluation of the standard Japanese navy stress tables dictating the strength of the structural components to be used in any aircraft led to the conclusion that the safety factors could be downgraded in certain instances and lighter-than-usual components substituted. In doing so, the team was aided by the fortuitous appearance of “Super Ultra Duralumin,” a new zinc-aluminum alloy recently developed by Sumitomo Metals that possessed far greater strength than existing copper-aluminum alloys. Air drag was reduced by streamlining the all-metal monocoque construction through the use of flush riveting, the adoption of retractable landing gear, the incorporation of a cockpit canopy, and the use of a “washout” configuration of the wings—a subtle downward twist at the wing tips. To bridge the incompatibilities of increased range and increased speed, the design incorporated the use of a detachable and aerodynamically shaped auxiliary fuel tank that could be carried beneath the fuselage and quickly dropped in combat, one of the first such devices in any air force. Generous wing area, large ailerons, and the adoption of a constant-speed variable-pitch propeller promised exceptional maneuverability and pilot control, and the wing mounting of two 20-millimeter cannon and two fuselage-mounted 7.7-millimeter machine guns provided formidable armament for a fighter plane of that time.29
The result of all these innovations by Horikoshi and his colleagues was a sleek and nimble-looking aircraft that was completed in March 1939 at the Mitsubishi plant at Nagoya. It was run through its company paces at Kagamigahara, where it surpassed all the standards the navy had set for it. After a few adjustments had been made, including minor changes in the wing design and the substitution of a more powerful Sakae 12 engine, the aircraft was turned over to the Naval Air Arsenal and the Yokosuka Air Group for navy flight trials that began in December 1939 and lasted nearly six months. In June 1940, impelled by the exigencies of the China War, the navy assigned fifteen of the new aircraft to the Twelfth Air Group already in China at the same time that the airplane’s carrier trials at sea were taking place aboard the Kaga. At the end of July, once the reports of the Kaga tests and the final conclusions from the Yokosuka test centers were in, the navy formally accepted the new aircraft as the Mitsubishi A6M (Type 0) Carrier Fighter.•30
The Mitsubishi Zero, as it came to be known worldwide, was undoubtedly one of the most ingeniously designed fighter planes in aviation history. While its configuration undoubtedly drew upon recent technological developments of the West, particularly those in the United States, there is no clear evidence that its overall design did not originate with the Horikoshi team.31 At the time of its appearance it was an aircraft of incomparable speed and phenomenal range (the latter being three times that of contemporary British and German fighters). Its low wing loading enabled it to turn inside any fighter plane then in production, and its large ailerons enabled its pilots to fly rapid rolls. The 950-horsepower Sakae engine with which the later models were equipped and the plane’s innovative wing design made it able to climb faster and steeper than any contemporary aircraft.32
At the time that it first entered combat, the serious flaws of the aircraft were not yet obvious, though in some cases they were the result of design trade-offs and were thus the obverse of its superior qualities.33 Like the design of a warship, the design of a combat aircraft represents a series of trade-offs between weight, speed, ceiling, range, maneuverability, armor, armament, and a number of other variables. Other defects were due to the restrictions imposed by Japan’s limited industrial base and inadequate strategic resources.
To begin with, its lightness, indeed its near fragility of structure, which helped to give the aircraft unprecedented maneuverability, meant that it contained too many single-point failure points (points at which structural failure would cause destruction of the entire aircraft), a weakness compounded by the fact that it possessed virtually no armor protection for either its pilot or its fuel tanks. While its rate and angle of climb were unprecedented for the time when the aircraft first took flight, its lightness meant that its diving speed was relatively slow, a defect that later in the Pacific War was to put it at a terrible disadvantage with respect to heavier and faster Allied fighter planes. While the Zero was a marvel of maneuverability below 4,500 meters (15,000 feet), its performance fell off sharply above that altitude, even though it had a flight ceiling of 10,000 meters (33,000 feet). Moreover, while the Model 21, the production model at the opening of the Pacific War, had a cruising speed of 180 knots and a maximum speed of 288 knots around 4,500 meters, at any speed above 260 knots the large ailerons of the A6M stiffened and the plane became incredibly sluggish and, here again, lost its deadly maneuverability.
There is no doubt that Horikoshi Jirō produced a fighter plane of outstanding performance, even though the Navy General Staff had presented him with an almost impossibly conflicting set of requirements and specifications. Yet after the war Horikoshi admitted that these deficiencies in performance—poor diving speed, poor handling capabilities at high altitude, and a comparatively low maximum speed—were all due to the aircraft’s small engine.34 This last imperative Horikoshi could do nothing about, since by the late 1930s Japan was restricted in its access to the strategic alloys essential to the production of high-tensile steel used in the larger American engines that were being developed.35 Moreover, a much larger engine would simply have shaken the Zero apart, given its structural fragility.
One last serious imperfection of the Zero should be mentioned: its armament. While its two 20-millimeter cannons in addition to its two 7.7-millimeter machine guns appeared to give it tremendous punch, this armament was to prove less than ideal for a fighter aircraft. In the first place, the cannon’s rate of fire and initial muzzle velocity were both low, which made it more difficult to hit a target than was the case with machine guns. Not only did the weight of the cannon and its heavier ammunition add significantly to the weight of the aircraft and thus affect its performance; the weight of the cannon ammunition meant that fewer rounds could be carried than machine-gun ammunition. As for its rifle-caliber machine guns, these were unlikely to inflict fatal damage on a heavily armored enemy, be it a bomber or a fighter plane. With twenty-twenty hindsight it is possible to argue that pilots of the A6M would have been better served if their aircraft had been armed entirely with heavy (.50-caliber) machine guns, as were most of the Allied enemies that it was to encounter in the Pacific War.
But during its appearance in the China War these ominous defects lay relatively far in the future, undetected by the Chinese and Russian pilots who flew the decidedly inferior planes that the Zero would encounter in the skies over the continent. During the Pacific War, later models of the aircraft were successfully modified to correct a number of these deficiencies, but midway through that conflict the heavy casualties among experienced Zero pilots, combined with the great increase in the performance of Allied fighter aircraft, rendered these improvements irrelevant in combat.36
Photo. 4-3. Mitsubishi A6M3 (Type 0) Model 22 carrier fighters
Source: Kōkūshō Kankōkai, Kaigun no tsubasa, 3:26.
In the more than sixty years since the Zero entered combat, its reputation has gone through cycles of admiration and condemnation. The object of popular awe and dread during the Pacific War, in postwar decades it wrongly came to be seen among publics in Western nations as a stereotypically flimsy product of inferior Japanese workmanship. By and large, pilots who came up against it in combat had a much more measured view of its strengths and weaknesses, and aviation historians in recent years have come to reinforce that middle view. But in any long-range assessment of the Zero, one thing must certainly be kept in mind: from its earliest conception, both its designers and the Japanese navy intended that at some point it was to be replaced by an aircraft of even greater all-around performance.
It was the profound misfortune of the navy and the nation that this never came to be. During the war the navy worked on several designs that could have replaced the Zero. The N1K2-J Shiden (“Violet Lightning”) was one of the best aircraft in the Pacific War and, by default, became the “successor” to the Zero, since the intended successor, the A7M Reppū (“Hurricane”), was not even designed until 1944. There was indeed a superlative aircraft that did, in a sense, “succeed” the Zero, the Mitsubishi J2M Raiden (“Thunderbolt”), a single-engined fighter provided with powerful armament and significant protection.37 The problem with all these aircraft was that they took too long to design, develop, and produce. By the time they came on line during the Pacific War, the tide of the conflict was running so violently against Japan that their appearance made little difference. The Raiden was a case in point. Developed by the overworked Horikoshi team, the aircraft was delayed by a series of design problems. Had it been available to meet the arrival of the second-generation Allied fighters in the southwestern Pacific in early 1943, it might have made some difference in that otherwise disastrous air campaign. As it was, the aircraft did not see serious combat until late 1944, and then only in small numbers.38
In any event, from 1937 to 1941 the navy developed not only outstanding carrier fighters but some of the best carrier attack aircraft of these years as well. The China War was to provide their first combat tests. In the fourth and fifth years of the navy’s involvement in that conflict, it introduced two excellent new bombers into its aerial operations on the continent: the Aichi D3A dive bomber and the Nakajima B5N attack plane.
The first of these had been conceived in the summer of 1936 as a replacement for the aging Aichi D1A2 (Type 96) carrier (dive) bomber. The specifications for the design competition had called for a two-seat monoplane design with speed and range capabilities equal to those that would become available in the Mitsubishi A5M carrier fighter. The winning entry by Aichi Tokei Denki turned out to be a rugged aircraft. Its design had been influenced to a great extent by contemporary dive-bomber designs of the German Luftwaffe. From the Heinkel HE 70, Goake Tokuichirō, Aichi’s chief engineer for the project, took the aircraft’s large elliptical wing, and from the Junkers JU 87, the famed Stuka dive bomber, he borrowed the configuration of dive-brakes under each wing to assist in control of the aircraft on dives and pullouts. Though Goake and his team were concerned with efforts to reduce drag, a consideration of the inevitable complexity and weight of retractable landing gear led to their decision to retain a fixed undercarriage. The original power plant, a 710-horsepower Nakajima Hikari radial, was deemed insufficient and was eventually replaced by a 1,000-horsepower Mitsubishi Kinsei 43 radial. In competitive trials with an aircraft of Aichi’s one rival for the contract, Nakajima, the Aichi plane proved far superior. Following service trials and numerous minor airframe changes, the Aichi D3A1 was accepted by the navy as the Type 99 carrier dive bomber Model 11.•39
The Type 99 was put through carrier qualification trials aboard the Kaga and Akagi in 1940, and in that year and the next it saw limited combat, operating from land bases in China and northern French Indochina. Roughly contemporaneous with the Stuka and the Douglas Dauntless, the D3A1 was essentially the equal of both in structural integrity, maneuverability, and speed, though in bomb load, range, and diving ability it was inferior to the Dauntless. Its early combat results were outstanding. Participating in all major Japanese carrier operations in the first year of the Pacific War, the Type 99 was to sink more Allied warships than any other type of Axis aircraft, but this may have been a function more of the outstanding aircrews who flew and fought in it than of the inherent superiority of the aircraft. In any case, the glory days of the Type 99 were brief. As early as the fall of 1942 the slow speed and poor armament of the aircraft made it shockingly vulnerable to Allied fighters, and its use in the land-based air war in the southwestern Pacific could only be termed suicidal.40
Even before the advent of the D3A1 dive bomber in 1940, the navy had issued specifications for a new carrier torpedo bomber, since none of the prototypes for such an aircraft developed in the early 1930s had proved satisfactory. An interim aircraft, the Yokosuka B4Y1 (Type 96),• the last fabric-covered biplane torpedo bomber in Japanese naval service, had played an active role in the China War. A more modern aircraft, with performance equal to the latest fighters, was obviously needed. Specifications had been issued as early as 1935, calling for a three-seat single-radial-engined aircraft of monoplane design, capable of a speed of 180 knots at 2,000 meters (about 7,000 feet), able to carry an 800-kilogram (1,764-pound) torpedo or an equivalent bomb load, and an endurance of four hours when carrying its full payload, or of seven hours for unarmed reconnaissance missions. Nakajima’s entry, the B5N, which eventually beat its Mitsubishi competitor for the navy contract, was a cleanly designed aircraft. It was equipped with such novelties as a variable-pitch propeller, a retractable undercarriage, and a hydraulically operated mechanism that folded the wings up and inward at approximately their halfway point for storage aboard a carrier. Completed in 1936, the Nakajima B5N1 made its first flight in January 1937, finished first in the competitive trials in November of that year, and was soon after adopted by the navy as the Type 97 carrier bomber.41
Though it almost immediately went into carrier service, the B5N was also allocated to the navy’s land-based air groups for operations in China, where, escorted by A5M carrier fighters, it performed well in support of ground forces. Combat experience in China and normal technological evolution led to a new version of the aircraft, the B5N2, driven by a more powerful engine (the 1,000-horsepower Nakajima Sakae 11). The B5N2 went into production early in 1940 as the Type 97 Model 12 and by 1941 had replaced the earlier version as the navy’s front-line carrier attack bomber. (A successor aircraft, the Nakajima B6N, was on the drawing boards in 1941 but was awaiting its flight trials when the Pacific War broke out.) At the time of Pearl Harbor, where it helped to destroy the American battleship force, the B5N2 was undoubtedly the world’s best carrier torpedo bomber, far superior to either the American Douglas Devastator or the ancient British Fairey Swordfish. But eventually, like the Aichi D3A1, its slow speed and paltry armament relative to U.S. fighter aircraft led to its wholesale destruction in the skies over the southwestern Pacific.42
Two other aircraft developed for the navy prior to the Pacific War—the Mitsubishi G4M medium bomber and the Kawanishi H8K1 flying boat—rounded out the family of aircraft that provided the material edge of the navy’s offensive air power on the eve of the Pacific War.
The Mitsubishi G4M had its origins in the navy’s desire for a replacement for the G3M bomber (even as the G3M was first entering combat in China), about the same time that the navy sought an eventual replacement for the A5M fighter.43 Not surprisingly, the navy wanted a bomber superior in performance to the G3M: a plane with increased engine power, greater speed, range, and payload, and better armament (which the slender configuration of the G3M could not accommodate). Development of such an aircraft by Mitsubishi—which in this case obtained the navy contract without going through the normal competitive process—took time, and it was not until September 1939 that a prototype was ready for flight-testing at Kagamigahara airfield. The twin-engined bomber that first flew in late October of that year was a cigar-shaped aircraft—indeed, the Japanese dubbed it the Hamaki (“Cigar”)—with an airframe that would be easy to manufacture and that would accommodate an internal bomb bay and more effective defensive armament: four machine guns, in the nose, the port and starboard waist positions, and the dorsal blister, as well as a cannon in the tail. Satisfied with its performance, Mitsubishi turned the plane over to the Yokosuka Air Group for its service trials in January 1940, following which the navy accepted the G4M as the Type 1 land-based attack plane Model 11.•44
Nevertheless, nearly a year was to pass before production of the new bomber was authorized, in large part because the navy wasted time in attempting to employ a modified version of the aircraft as a more heavily armed escort for the navy’s medium bombers. At the time, the G3M bombers were suffering comparatively heavy losses in the skies over China because the A5M fighter had insufficient range for long-range bombing operations and because the A6M fighter had not yet come on line. But the G6M1 Type 1 “wing tip escort aircraft” (yokutan engoki) proved to be a failure for three reasons: it handled poorly because the reconfiguration of the aircraft to accommodate increased cannon armament caused a shift in the aircraft’s center of gravity; the addition of partial protection for the wing tanks caused a reduction in fuel capacity and thus a reduction in range; and the eventual appearance of the Zero in the same month that the first two G4M1s were completed eliminated the need for such an escort bomber. For all these reasons, the production of the G6M1 was canceled.45
At last, in the spring of 1941, the G4M in a bomber configuration was allocated to the Takao Air Group in China. If the G4M was an improved version of the long-range bomber that Yamamoto Isoroku had envisioned early in the 1930s, it was, like its predecessor, an aircraft whose serious flaws would become apparent only in combat. Its two Mitsubishi Kasei radial engines, of approximately 1,400 horsepower each at cruising speed, helped to give it the range the navy desired for its land-based medium-bomber force—in theory, 3,700 miles. To achieve that phenomenal range, greater than that of the American B-17, Mitsubishi, again under navy pressure, sacrificed ruggedness, armor, and armament. Despite improvements in armament and an effort to include fuel tanks that were in part self-sealing, the G4M was an aircraft so lacking in armor and thus so vulnerable to explosion under enemy fire that in the Pacific War it was to earn the grim sobriquet “The One-Shot Lighter” among its crews and the dismissive nickname “Zippo” among American fighter pilots.
In this sense the G4M was a flying contradiction. Designed to work in tandem with the Zero fighter as escort, its maximum range was too far for even the Zero, though admittedly bombers with greater ranges than fighters comprised, of course, a worldwide phenomenon for much of World War II. Facing an enemy target unprotected by adequate fighter defense or employed against a target area first cleared by Zero escorts, the G4M was able to achieve some remarkable bombing runs. The bombing of Chungking and other missions deep into central and southern China in the spring and summer of 1941, the destruction of the battleships Repulse and Prince of Wales at the outset of the Pacific War, and the long-range attack on Port Darwin, Australia, in the spring of 1942 are all testimony to the extraordinary reach of the G4M. But without fighter escort it was a flying firetrap, and in the air war in the southwestern Pacific this frailty eventually forced its pilots to make bombing runs at altitudes that were sufficient to reduce enemy fighter attacks but also made bombing accuracy almost impossible.
By the years of the Pacific War, seaplanes—both flying boats and floatplanes—came to occupy a significant place in Japanese plans for amphibious operations in the southern Pacific. Advance seaplane bases to supply reconnaissance and air cover for invasion convoys were seen as easy to establish, given the large number of quiet lagoons and sheltered harbors of the tropical Pacific.46 For great distances, the flying boat was still the navy’s principal reconnaissance aircraft, and the navy continued to depend on Kawanishi to produce this type. The Kawanishi H6K had proved to be an outstanding aircraft, but in 1938 the navy was already thinking of a successor and therefore issued to Kawanishi specifications for a flying boat superior in performance to the British and American equivalents of the H6K, the Short Sunderland and Sikorsky XPBS: a cruising speed of 240 knots and a range of 3,900 nautical miles. Beginning in August 1938, Kawanishi began testing models of the projected design in wind tunnels and water tanks. These experiments led to the development of the H8K1, a clean but sturdy-looking high-wing monoplane with cantilevered wings that held four Mitsubishi Kasei engines of 1,530 horsepower, the navy’s most powerful and reliable power plants. Contrasted with the navy’s land-based bombers, the H8K1 was a stoutly protected and defended aircraft, equipped with extensive armor protection and armed with 20-millimeter cannon in the dorsal and tail turrets and 7.7-millimeter machine guns in the two beam blisters, the ventral and cockpit hatches, and the bow turret. A prototype had undertaken its maiden flight in January 1941, and after successful service trials late that year the navy accepted the aircraft as the Type 2 flying boat, Model 11.• The Kawanishi H8K1 had been conceived as a reconnaissance aircraft, but its speed and range led the navy to use it also, like its predecessor, as a transport and, on several occasions during the Pacific War, as a bombing aircraft. It was, in any event, an outstanding representative of its type. Because of its rugged construction and excellent armor and armament, Allied fighter pilots considered it the most difficult Japanese aircraft to bring down. For all these reasons, it was probably the finest flying boat of World War II.47
For shorter-range reconnaissance, as well as operations, the navy came to acquire two workhorse floatplanes for the fleet. Despite its obsolete configuration, the singular maneuverability of the Mitsubishi F1M1/2 Type 0 (“Pete”) observation seaplane,• a single-engined two-seat single-float biplane that went into production in 1940, led to its use not only as a catapult-launched aircraft aboard battleships and cruisers but also as an interceptor, a dive bomber, and a coastal patrol aircraft.48 Similarly useful was the Aichi E13A1 Type 0 (“Jake”) reconnaissance seaplane,• a single-engined three-seat twin-float monoplane that went into production in 1940. Operating from cruisers and seaplane tenders, it saw service in 1941 both in attacks along the China coast and in reconnaissance over Hawai’i.49
In this survey of Japanese naval aircraft developed from the mid-1930s to the opening of the Pacific War, it remains only to mention two attack aircraft that were in various stages of development by the opening of the Pacific War but that played significant roles later in the conflict: the Nakajima B6N (Tenzan), a three-seat single-engined carrier attack plane intended to replace the Nakajima B5N, but which proved to be only marginally better; and the Yokosuka D4Y (Suisei) dive bomber, with the only in-line engine ever used aboard Japanese carriers. Intended to replace the Aichi D3A, the D4Y ran into major technical problems, and it did not come into widespread service, either as a reconnaissance aircraft or as a dive bomber, until midway through the Pacific War.50
THE JAPANESE AIRCRAFT INDUSTRY
If the industry that produced these aircraft was “unimpressive,” in the dismissive view of a postwar American survey of the wartime Japanese economy and technology,51 it was only in comparison to the scale and progress of the aircraft industry in the United States during these same years. From 1937 onward, the growth in Japanese aircraft production was remarkable even as management of the industry was increasingly directed by the government.
The two largest manufacturers, Mitsubishi and its nearest rival, Nakajima, were the only companies that designed and produced aircraft for both armed services. By 1937, Nakajima was beginning to edge up on Mitsubishi as the nation’s largest airplane manufacturer. The next two competitors produced aircraft exclusively for the Japanese army and were trailed distantly by Aichi and Kawanishi. Whereas Japanese industry made a little over eleven hundred aircraft deliveries to the two military services in 1936, it delivered more than five thousand in 1941.52 By 1941, Nakajima’s airframe plant at Ōta and engine factory at Musashi were enormous, and Mitsubishi’s airframe plant in Nagoya was the second largest in the world.
The first half of the 1930s had seen the Japanese aircraft industry move out from a period of importation and imitation of foreign models toward domestic design as well as production of airframes and engines. As we have seen, the Japanese navy had taken the lead in the development of an independent Japanese aircraft industry with the establishment of the Naval Air Arsenal at Yokosuka in 1932. The arsenal soon began designing and manufacturing its own aircraft in competition with civil manufacturers, as well providing guidance to these same manufacturers in the production of aircraft under contract to the navy.53
By the end of the 1930s, therefore, except for its use of imported machine tools and presses and residual foreign design influences, the industry was almost entirely domestic in character. Yet the industry’s relationship with government—or, more exactly, with the two military services—was essentially still as it had been earlier: the army and navy awarded contracts to aircraft manufacturers in accordance with service needs but otherwise exerted little control over, and provided no material support for, these same contractors. The Japanese army left both engineering research and trial production of aircraft in the hands of private aircraft contractors. The navy, however, conducted the necessary research in its own laboratories, which after 1932 were located at the Naval Air Arsenal at Yokosuka. Though it contracted with civil manufacturers for the production of most of its aircraft, the navy also adopted some aircraft designed and manufactured by the Yokosuka Air Arsenal. On their part, Japan’s civil aircraft manufacturers were responsible for procuring their own labor and materials but were free to expand or contract on the basis of the volume of business they carried.54
Then, with the onset of the China War, the industry came under increasingly tight government control. In 1938 a new law compelled all aircraft companies of a certain capitalization to be controlled as to equipment, techniques and production plans. In the years that followed, the government, while not directly providing capital, did offer numerous incentives for the aircraft industry to expand, including loans made through industrial banks, leasing of government-owned machine tools, and other indirect forms of assistance.55
In return, by the spring of 1941 the two military services came to control all aspects of management over favored aircraft plants, establishing standards for acceptance of aircraft, setting up inspection procedures, and stationing their technical representatives and inspectors in the factories so that quality standards were maintained. By the end of that year, as the availability of certain strategic materials necessary for the manufacture of aircraft, particularly aircraft engines, became even more restricted, each of the services organized its own control of these raw materials, which it allotted to the aircraft manufacturers handling its contracts.
The positive qualities in the design and manufacture of the navy’s aircraft demonstrate the skill of a significant segment of the Japanese engineering community—near brilliant in the case of the Zero—and the reasonably effective overall planning and production methods of prewar Japanese airframe and engine plants. Yet as wartime pressures and exigencies were to reveal, there existed serious inherent problems in the Japanese aircraft industry on the eve of war, difficulties that were to become worse as the war went on.
There was, first of all, the problem of the limited pool of scientific and technical expertise in Japan and the modest scale of the research facilities that could augment that expertise. This situation would not have become so critical had Japan maintained access to scientific and technical developments in the West, particularly in the United States, during the prewar decade. As it was, this access increasingly narrowed during the 1930s, and Japanese industry, particularly its aircraft-manufacturing segment, was largely ignorant of the scale and progress of aircraft design and manufacture in the United States.56
The excessively paternalistic relationship between the Japanese government also aggravated the adverse consequences of this technological isolation. Without foreign competition, there was little incentive for technological innovation. Instead of the sort of price competition that existed in the West, the Japanese aircraft industry was characterized by price-fixing and excessive government patronage. In this hothouse atmosphere the industry failed to make breakthrough technological progress, and competition often halted at the development of an aircraft prototype, with subsequent industry efforts being placed largely on production.57
Japan’s technological isolation also compounded the problems caused by the nature of the aircraft industry’s workforce. While there existed a small grouping of experts like Horikoshi Jirō at the top of the Japanese industrial scale, for the most part the industrial workforce at the lower levels had less than a generation of mechanical experience behind it. This meant that there was a significant discrepancy in the relative efficiency of Japanese and American aircraft workers in their respective industries, with the average Japanese worker producing about one-third of the output of his American counterpart. There were several explanations for this discrepancy other than lesser mechanical experience, of course: the fact that half of all riveting and one-third of all sheet-metal processing in the Japanese aircraft industry was done by hand; the existence of bottlenecks caused by inadequate machine tools, which were sometimes worn, poorly designed, inadequately cast, or low in speed of operation.58
The Japanese government also created a serious bottleneck when, in 1938, it insisted that a significant percentage of aircraft manufacture be carried out by subcontractors, many of which were quite small and not that efficient, since they in turn often used home industry. This sometimes led to a lack of precision and standardization of parts, imperfections that could hamper final assembly efforts at the plant and handicap maintenance operations on aircraft in the field. For the most part, such subcontracting slowed the speed of production. In a large aircraft of 100,000 to 150,000 parts, this could become a serious problem, but it also manifested itself in the manufacture of the advanced design of the Mitsubishi Zero, which made it difficult and time-consuming to produce, since much hand labor was involved.59
Finally, the Japanese aircraft industry was seriously hampered by the standoff relations between its two military masters. Across a range of functions and activities, the army and navy demonstrated their inability or unwillingness to cooperate in the development of combat aircraft. Both services carried out aviation design and production research and development in their arsenals and depots, but neither shared the results with the other. Each service saw to it that particular aircraft were designed and produced only for that service. In the months immediately prior to the Pacific War, when a shortage in strategic raw materials necessary to engine and airframe production became critical, each service developed its own sources of supply for such resources with little thought to the needs of the other service.60
An example of the obdurate nature of such interservice suspicion and hostility was provided by Germany’s wartime naval attaché to Tokyo, Capt. Paul Wennacker. After the war Wennacker recalled that during a tour of the Nakajima aircraft plant he was first guided by several naval officers around the navy’s development and manufacturing division of the plant. At the conclusion of the tour, the navy men opened a door that had been kept tightly closed. Here the naval officers bade him good-bye, and on the other side of the door a group of army officers took him on a tour of their section of the plant, an area to which the navy officers had no access.61
During the immediate pre–Pacific War years, none of these defects within the Japanese aircraft industry was of major consequence, and collectively they were more than balanced by the significant growth of the industry and the quality of aircraft it produced. Yet even at the opening of the war, the navy possessed an imperfectly armed air service. The carriers of the Combined Fleet had all the first-line aircraft they could embark, but many of the land-based naval air groups had too few of the latest-model fighters and medium bombers.62 Once the Pacific War turned into a grinding conflict of attrition, the Japanese aircraft industry was so riven by the widening fissures I have described that it had little hope of matching the output of its enormous rival across the Pacific.