One Hundred Years of U.S. Navy Air Power

by Smith, Douglas V.

  With the Yorktown design BuAer accepted that an island structure was not merely a necessary evil, but that it presented a vital means of controlling the flight deck. By 1932, with Ranger nearly complete, the bureau badly wanted her modified to incorporate an island. It was far too late to lead smoke pipes up through the small new island, but the structure did provide a conventional bridge (that incidentally reduced congestion in the forward part of the hangar) as well as positions for two directors, which made the ship’s 5-inch guns far more effective. The Ranger island was in effect the island structure planned for the Yorktowns less their massive funnels.24

  In effect these ships were pre-prototypes of the mass-produced Essex class of World War II. Investing in them showed just how important naval aviation was to the interwar Navy. They proved extremely tough. Enterprise fought in nearly every Pacific battle and survived the war. Yorktown was sunk at Midway.

  There was not enough tonnage for a third ship, so the General Board suggested a 15,200-tonner (32.5 knots, like the 20,000-tonner) for the next (1934) program. Construction was delayed because the tonnage involved was absorbed by the obsolescent Langley. In 1937 she was de-rated to seaplane tender status, hence removed from carrier tonnage. A much-improved Ranger, USS Wasp, was built (by this time available tonnage had dropped to 14,500). Like the Yorktowns, she had a massive island. Again, limited tonnage meant limited speed and also limited survivability. In Wasp as in earlier carriers, there was interest in flying aircraft from the hangar deck to increase the rate at which they could be launched. She was completed with two hangar deck catapults in addition to two flight-deck catapults. As with other carriers, Wasp was given an enlarged fighter squadron (twenty-seven aircraft) in 1940. For the rest she had a standard air group at the beginning of World War II: a scout squadron, a scout bomber squadron (dive-bombers), and a torpedo squadron, but by 1942 the torpedo squadron was twelve rather than the usual eighteen aircraft. The scouts and dive-bombers were the same SBD airplane, but their roles were quite different. Although the 15,200-tonner conceived in 1931 would have had the power and speed of the Yorktowns, Wasp was laid down with a 70,000 rather than 120,000 SHP) plant and a designed speed of 29.5 knots (Ranger produced only 57,000 SHP.

  Toughness meant the carriers were difficult to sink, but it was widely understood, in the 1930s, that their flimsy flight decks could easily be destroyed by light bombs. Although the phrase “eggshells wielding hammers” was coined to describe contemporary heavy cruisers, carriers were best described in this way, particularly as their offensive power grew during the 1930s. It was assumed that whichever side found the enemy’s carriers first would put them out of action. After a war game, for example, Admiral King, commanding the Battle Fleet Aircraft, was asked by battleship officers why his ships were not providing them with air services (such as protection for their spotting aircraft). They seemed to fight a private carrier versus carrier war in every fleet exercise. King replied that unless his carriers defeated the enemy’s at the outset, the battleships would never have any air services at all. Given this perception, through the 1930s U.S. carrier officers asked why future carriers could not be provided with at least lightly armored flight decks.

  The answer says a great deal about contemporary U.S. thinking. An armored flight deck would represent considerable topweight, even with thin armor. On a given displacement (limited by treaty), that topweight could be accommodated on a smaller hull with a smaller air wing. The alternative, unavailable at the time, would be to provide much the same air group by growing the hull to balance off the topweight involved. The smaller-carrier solution was regularly rejected because it was so important both to find and to kill enemy carriers. U.S. carriers were expected to operate alone, so each carrier had to be able to conduct all these functions. Finding an enemy required a squadron of dedicated scouts. Killing an enemy required a squadron of dive-bombers, preferably accompanied by a squadron of torpedo bombers—and by fighters. Four squadrons—more than eighty aircraft—required a large flight deck, hence a large carrier. The scouting (S) and dive-bombing (B) roles were combined in single airplanes by the late 1930s, but they were still separate functions and they still required separate squadrons. Thus by the outbreak of World War II, U.S. carriers typically accommodated two squadrons of scout bombers, such as the Douglas SBD Dauntless.

  In the late 1930s the big Fleet Problem exercises awakened the fleet to the desperate need for direct fighter protection; there was only limited faith in anti-aircraft guns, and it seemed that only fighters (if anything) could deal with the dive-bombing attacks the U.S. Navy itself favored. Yet the single squadron of fighters aboard each carrier was considered part of that ship’s own anti-aircraft screen, or as protection for the ship’s own striking force (both the war in China and the Spanish Civil War seemed to show that such protection was essential). The standard carrier squadron of eighteen fighters already included nine spare aircraft to allow for attrition and to maintain full operating strength when some aircraft were being overhauled. Up to three of these “operating spares” were normally used that way. Commander, Aircraft Squadrons Battle Force Rear Admiral Ernest J. King (later the wartime CNO) proposed using spares to strengthen fighter squadrons on board the large carriers to twenty-four active aircraft. The Scouting and Battle Force commanders wanted a lot more. Ultimately the recommendation was to give the two slow carriers (Ranger and Wasp), which would probably work with the battle fleet, fighter-heavy air groups, leaving the big fast carriers, which would probably operate as independent striking forces, with their original air groups, tilted toward attack rather than defense. After all, if it was accepted that an air attack once launched could not be stopped, the only real fleet air defense against enemy carrier aircraft was to destroy them at source—at the enemy carrier. As for the strike carrier’s own fighters, normally they would form a continuous combat air patrol. Maintaining it would exhaust the pilots, who would be unable to escort the carrier’s bombers once the enemy fleet was found. Note that search radar could change this situation; the carrier might have to mount a combat air patrol only after enemy aircraft were detected.

  Ships already had the capacity for twenty-seven fighters; they just normally did not carry enough pilots. In June 1940 BuAer proposed that the fighter squadrons be boosted to twenty-seven airplane strength, and OpNav agreed. By that time the agreed characteristics for the next carrier class (Essex) already incorporated an interesting twist: they should have the usual four eighteen-airplane squadrons, but also space to accommodate a fifth (fighter) squadron of eighteen aircraft. Given pressure for more fighters, the bureaus interpreted this to mean five squadrons of planes with their personnel, stores, guns, and ammunition rather than the spare planes previously imagined. This new interpretation certainly comported with the twenty-seven fighter concept approved in June 1940; the new Essex had sufficient capacity for a double fighter squadron (thirty-six aircraft).

  Surviving papers do not mention radar as a factor in these changes, but it is difficult to discount, because it solved a serious problem. Through the 1930s hopes that the carrier’s own fighters could protect her against enemy attack had declined. About 1930, U.S. thinking envisaged a circular formation in which destroyers many miles from the carrier could spot approaching enemy aircraft in time for carrier fighters to take off and deal with them. Through the 1930s aircraft performance improved dramatically, to the point that it was no longer possible to set up a satisfactory screen (that proved difficult enough even with radar). Some suggested abandoning carrier fighters altogether. They survived mainly as a means of supporting strike aircraft (with light strike capability) and also as a means of supporting the Marines should they have to land on enemy islands (a role already being taken very seriously). The advent of radar in 1937–1939 changed the situation completely. Suddenly it became possible to provide effective warning, even against fast aircraft. Radar was first tested in the January 1939 maneuvers, aboard the battleship Texas. It did take considerable time for the U.S.
Navy to turn this raw technology into an ability to control fighters. At Midway a carrier versus carrier battle developed much as had been predicted. U.S. strike aircraft found three Japanese carriers before the Japanese found the Americans and inflicted fatal damage. Then the surviving Japanese carrier found an American carrier, USS Yorktown. Her aircraft managed to damage Yorktown fatally (she sank due to loss of power and inability to de-water after numberous Japanese hits). The U.S. ships had radar, but they also had only very primitive fighter control arrangements. Two years later, at the Philippine Sea, the Japanese found the American carriers first. However, the U.S. ships had highly developed fighter control based on the new technology of the Combat Information Center (CIC). The Japanese air striking force was massacred in the “Turkey Shoot.” It happened that most Japanese carriers themselves survived. One irony of the battle was that the U.S. aircrews thought they had done poorly; Samuel Eliot Morrison reports a feeling of depression on the hangar decks. In fact they had ended the Japanese naval air arm, because the Japanese could never train replacement aircrew—a fact not appreciated a few months later at Leyte Gulf. The core issue was aircrew more than carriers: the great strength of the U.S. Navy lay not only in its ability to multiply carriers and their aircraft but even more in its ability to multiply aircrew.

  The Washington Treaty (and its successor, the London Treaty of 1930) affected the U.S. Navy in a subtle way. Until 1934 U.S. warships were individually authorized by Congress. That year Japan withdrew from the treaty system, and a few in the United States realized that war might be coming. Congress passed the Vinson-Trammell Act, which authorized a “modern Treaty Navy.” This phrase avoided the appearance of expansion, but allowed considerable new construction, because the United States was well below the allowable tonnage. (Congress still decided how much money to appropriate, thus still controlled new construction.) By 1936 it was clear that Japan was no longer interested in any kind of limitation, but there was still hope that a treaty system might limit naval spending and indirectly avoid war. To that end the replacement London Treaty eliminated the tonnage totals on battleships and carriers, but it retained limits on individual ship size (the carrier limit was reduced to 23,000 tons, at British insistence). Signatories had to announce their naval building plans well ahead of construction, the idea being that they would restrain themselves to avoid touching off a building race. Given the form of the Vinson-Trammell Act, as the international situation worsened Congress authorized a percentage increase in various categories of treaty-limited warships in the Second Vinson-Trammell Act (1938).25 Because battleship and carrier tonnages were tied back to the Washington Treaty, the Act implicitly maintained much the same ratio between the two types—which had reflected the much earlier experience of World War I. The phrase “modern Treaty Navy” meant that new construction was approved to replace existing overage ships, replacement age having been set by treaty. In 1936 the only over-age U.S. capital ships were battleships. U.S. carriers were all far too young to replace in the near term.

  Particularly if carriers did not use torpedo bombers, it was not clear how effective they could be against other ships. An American attempt to multiply naval aircraft within the treaty limit shows how tricky this question could be. In the run-up to the 1930 London Naval Conference, Admiral William A. Moffet of the Bureau of Aeronautics suggested building a flight-deck cruiser, which could add to the fleet’s air strength without counting against carrier tonnage. At the conference, the chief American naval adviser, CNO Admiral William V. Pratt, managed to get the flight-deck cruiser into the newly limited total cruiser tonnage, of which the United States had a great deal available (the British, with whom the United States had parity, had a far larger cruiser fleet). However, when the flight-deck cruiser was tested at Newport, it failed. Its aircraft just did not have enough offensive power to balance off the reduction in cruiser guns; in duels the flight-deck cruiser was generally sunk. The design effort, which had gone to the point of preparing plans for bids by shipbuilders, was abandoned.

  The Second Vinson-Trammell Act provided enough tonnage to build a slightly improved Yorktown (USS Hornet). Her construction seems to have been pushed through by Congress against some naval staff opposition; the General Board felt that it was more urgent to build modern battleships. That did not necessarily mean that the senior navy was against carriers: one role of modern fast battleships was as escorts for carriers. Existing battleships could not keep up with carriers, and in a preradar era there was a real possibility that enemy capital ships would catch carriers—as actually happened in 1940, when the German Scharnhorst and Gniesenau sank HMS Glorious.

  In the late 1930s there was also a fear, at least on the General Board, that the best land-based aircraft grossly outperformed modern naval aircraft, hence that carriers might be moving toward obsolescence.26 That turned out to be a temporary consequence of the level of engine development. The most powerful engines suited to fighters produced about a thousand horsepower. The British and the Germans achieved high performance effectively by wrapping airplanes like the Spitfire around these engines, providing far too little fuel for naval purposes as then understood. The U.S. Navy found that similar power applied to a fully navalized airplane like the Wildcat (F4F) produced much less performance (the Japanese solved the performance problem in the A6M Zero by adopting a very light structure, limited armament, and no armor at all). Spitfire speed was limited more by what a propeller could handle than by engine output. By 1938 the Bureau of Aeronautics understood that the coming generation of two-thousand-horsepower engines would solve the problem. Aircraft could not gain much more speed, but they could achieve maximum speed with acceptable range and other features. The resulting fighters, like the Hellcat and Corsair, were quite the equal of their land-based contemporaries. It is not clear how quickly the conclusions reached at the technical end of the Navy reached the policymakers.

  U.S. naval exercises show how current fears played out. In the early 1930s, U.S. carriers regularly practiced attacking land targets, raiding Los Angeles and Pearl Harbor. In the mid-1930s the U.S. Navy adopted the high-performance (for the time) Catalina (PBY) flying boat. In line with the policy of using large numbers of flying boats (not limited by treaty) to provide an important component of naval air power, the Navy hoped that the Catalina could function not only as a scout but also as a bomber (hence the B in its designation). It seemed that Catalinas or their equivalents patrolling from Pearl Harbor could find an enemy carrier force before it got within attack range. The comment that land-based aircraft were reaching superiority meant that such detection would be fatal; the U.S. carriers stopped raiding Pearl Harbor in their exercises. This shift may have played into disbelief, in 1941, that Japanese carriers (presumably operating under the same limitations as U.S. ships) would attack Pearl Harbor. It turned out that scouting was far more difficult than had been imagined prewar, and that carrier attacks on land targets were generally entirely practicable; but that took time to learn. Again, the perceived vulnerability of carriers was key to U.S. naval thinking.

  CARRIERS FOR A TWO-OCEAN NAVY

  The story of the World War II carriers began when Germany, Italy, and Japan became Axis allies in the fall of 1938. By this time the Japanese attack on China, and German pressures for territory in Europe made war seem likely. Until this time U.S. national strategy was shaped by the requirements of the most likely war, a war against Japan. Once Germany had created the Axis, it was no longer possible for the United States to contemplate a war fought entirely in the Pacific. Too, Germany and Italy now clearly threatened to penetrate South America; the United States needed the naval strength to prevent any armed assault in the Western Hemisphere. The two problems were linked in that the United States seemed to need access to southeast Asian resources in order to maintain sufficient strength to beat off any Axis attack in the Western Hemisphere. By October 1938 the OpNav War Plans Division was investigating the building program required in a “war of maximum effort,” corresponding
to the World War II mobilization. In spring 1939 U.S. naval planners began to lay out the requirements of a “Two-Ocean Navy” capable of fighting simultaneous wars in the Atlantic and in the Pacific. The United States began secret staff talks with the British, who would probably also be involved in such a war.

  A two-ocean Navy would clearly require more ships, including more carriers. Battleship numbers were set by the need to match all potential enemy navies in both oceans. It did not yet seem that carriers could be dominant—and neither Germany nor Italy had a large carrier fleet the United States would have to match. It is striking in retrospect that this analysis did not take into account the possibility that carrier superiority would change the rules, as when a British carrier strike sank much of the Italian battleship fleet at Taranto in November 1940—or when the Japanese did the same thing to the United States at Pearl Harbor.

  The new ships were the Essex (CV-9) class designed during the fall of 1939. Because work began after the outbreak of war on 1 September, the previous treaty limit (23,000 tons) did not apply.27 However, it was possible to work quickly because the new carrier embodied many design features of the earlier 20,000-ton type; thus it was still somewhat limited by the defunct treaties (it displaced 27,200 tons). In drawing up characteristics for the new carrier, the General Board considered and rejected alternative proposals for relatively small carriers with armored flight decks, which would have supported many fewer aircraft. The way in which carrier flight decks had to provide sufficient length for landing-on, parking, and taking off dramatically limited the aircraft capacity of small ships.