One Hundred Years of U.S. Navy Air Power

by Smith, Douglas V.

  Illustrious suffered heavy dive-bombing in 1941 and was sent to Norfolk Navy Yard for repairs under the Lend-Lease program. There she made a considerable impression on U.S. officers—who did not notice that the German armor-piercing bombs actually had made such a mess of her flight deck that she could not keep operating. Those defending U.S. design practices said that the British ship was far better for European waters but that an American-type carrier with many more aircraft made much more sense in a lengthy Pacific campaign. U.S. officers wanted both an American-style air group and British-style protection, accepting that the result would be a larger ship (now unrestrained by treaty, with much relaxed financial limits, too). By July 1941 BuShips had details both of the Illustrious design and of the damage she had suffered. The Royal Navy had built a carrier with an armored hangar, not one with a full-length armored flight deck. A heavy German bomb penetrated her hangar roof (part, but not all, of her flight deck) and gutted her hangar. Because the hangar did not run the full length of the ship, had the bomb hit slightly further forward it would have penetrated deep into the ship, with disastrous consequences. On the other hand, the armor forced the Germans to use a large bomb, and that in turn reduced the scale of the attack, because not all German aircraft could have conducted the attack. Under similar circumstances an Essex would also have survived, although the hangar deck might have been penetrated. BuShips developed several designs showing what an Essex-like carrier would have to sacrifice to gain some degree of hangar protection. For example, a 28,000-ton carrier (CV-D) with 2.5-inch armor on her flight deck (Illustrious had 3 inches) would operate sixty-four aircraft rather than the eighty-three then credited to the Essex, and would make 31.5 rather than 33 knots. In November 1941 BuShips offered CV-E, enlarged so that in effect she made no sacrifices to gain a 3.5-inch armored flight deck. She had a 2-inch armored hangar deck. As in the earlier studies, adding armor so high in a ship made for much larger size; in effect the armored flight deck equated to the 8-inch guns of the CV-A study. Given her sheer size, she could operate six squadrons of existing types of aircraft.

  In March 1942 the General Board formally proposed building a big carrier based on CV-E with an armored flight deck, arguing that such tough ships were essential. She would be so massive that she could support the larger aircraft then being bought, such as the Grumman F7F Tigercat (which had been conceived, however, to operate from Essex-class carriers). This U.S.-style armored flight-deck carrier was soon called the Type B Carrier (CV Type B, then CV-B, and finally CVB, the B often interpreted as “Big” or “Battle”). Six of these Midway-class CVB were projected, but three were cancelled at the end of the war.

  The British managed to incorporate the heavy armored hangar by making it integral with their hull. U.S. designers could not adopt any such practice, so one great problem in the Midway design was how to support the vast weight of a deck treated as a superstructure. One lesson for the designers was never to do that again; in postwar carriers the flight deck was integral with the hull. It was a great triumph of U.S. structural design that large openings could be cut in the ship’s side (to provide something like an open hangar) despite the integral-deck structure. The sheer weight of the armored deck made it necessary to reduce freeboard, so the Midway-class carriers were far wetter than the Essexes. Another problem was how to provide some protection along the sides of the hangar, since surely an enemy would try to get weapons under the armored flight deck. No good solution presented itself, but the partial one was to place the ship’s 4-inch guns in protected mounts along the sides of the hangar deck, so that there was a good chance that any weapon fired at the hangar would hit them instead. As a consequence, these ships had clearer flight decks than the Essex class.

  In 1945 U.S. and British fleet carriers operated together in the face of Japanese kamikazes. U.S. officers noticed that kamikaze hits usually put their ships out of action (although many were repaired in forward areas). They considered the British armored-deck ships far superior, the saying being that after a hit they simply manned sweeps and cleaned up their flight decks. The British saw things very differently. They were painfully aware of how limited their aircraft capacity was and of how their aircraft could not be warmed up in their closed hangars (which were treated like magazines). When the British began to design a new carrier in 1943, they compared open- and closed- (armored) hangar designs. Eventually they selected a U.S.-style open hangar for their final wartime carrier design, HMS Malta. The ship was cancelled at the end of World War II, so until records were opened it was not clear just how impressed the British had been with U.S. design practices.

  CONVERSION

  The British developed another new carrier idea, a converted merchant ship to work with convoys. U-boats had to spend much of their time on the surface, submerging (if at all) only when near their targets. Only when surfaced could they develop sufficient sustained speed to shadow and then to intercept a convoy (very late in World War II the Germans introduced the snorkel, and the situation changed). Aircraft working with a convoy could catch surfaced U-boats. As yet there was no way that aircraft could detect submerged submarines, but the ability to frustrate U-boats trying to intercept a convoy was well worthwhile. British carrier operating practice had produced low-performance aircraft like the Swordfish torpedo bomber, which could operate even from a slow carrier with a small flight deck. By the time the British had their first converted merchant ship carrier (escort carrier), the U.S. Navy was working with the Royal Navy, although the United States was not yet at war. The U.S. Navy had no low-performance bombers, but it became interested in operating auto-gyros (predecessors of helicopters) from a converted merchant ship with a short flight deck. It converted a new freighter into the prototype escort carrier Long Island. By this time the U.S. Navy also had the new hydraulic catapult conceived for fleet carriers.29 These catapults made it possible to operate standard U.S. carrier aircraft from a converted merchant ship. Initially the new low-performance carriers were considered auxiliaries, designated AVG (seaplane tenders were AVs); later they were designated auxiliary carriers (ACV) and finally escort carriers (CVE). Late in 1941 the nominal U.S. requirement was twenty-five, soon cut to sixteen (with fifteen already under contract).30 The British were allocated another thirty, tentatively designated BAVGs, to be provided under Lend-Lease. Four more conversions were ex-tankers (Cimarron class), making a total of fifty such ships.31

  The Cimarron conversion was quite attractive, because it was larger than a C-3 and offered oil fuel for accompanying ships. The maximum effort program planned in 1942 (for 1943–1944) therefore included another twenty-four such ships, to be built as carriers from the keel up. None of these Commencement Bay–class ships was ready for combat before the end of the war, but they were used postwar. The earlier ships were reduced postwar to subsidiary roles, the most interesting of which were conversion of Thetis Bay into a prototype helicopter assault ship and conversion of Gilbert Islands into a communications relay ship (AGMR). In 1942 Henry Kaiser, who was already mass-producing Liberty ships on the West Coast, offered to build escort carriers designed as such from the keel up. The resulting Casablanca class was based on a Maritime Commission design for a small passenger ship (had it really been conceived as a carrier from the keel up, the design would have taken far too long). Fifty were built under a program authorized by President Roosevelt (hence not negotiated within the naval staff). Taking into account U.S. ships built for the Royal Navy, well over a hundred escort carriers were completed during World War II.

  Conversion did not produce a particularly efficient carrier. The merchant ships were short and slow. Their holds were ill-suited to the shops and other facilities a carrier needed. Carrier facilities were far lighter than the usual cargoes, so the ships had to be ballasted heavily. The island was cut to a bare minimum, smoke pipes being led up alongside the flight deck. This very austere conversion, however, could be carried out rapidly, and in 1942 it was urgent to produce ships to fight the Battle of the Atlantic.


  By 1943 the role of U.S. escort carriers had changed dramatically. To fight its war, the U-boat command had to maintain communications with all boats at sea. It directed them to targets, and it needed their reports of, for example, their fuel status and how many torpedoes remained. The Germans thought that their codes were secure and that reports from U-boats, using high-frequency (HF) radio could not be intercepted and certainly could not be subject to radio direction-finding. They were wrong on both counts. Code-breaking and radio direction-finding provided the Allied navies with what amounted to wide-area ocean surveillance of the U-boat force. In May 1943 the U.S. Navy began using this information to hunt down U-boats, rather than merely to screen convoys. Hunter-killer groups built around escort carriers attacked groups of U-boats whose position had been revealed by code breaking. Their aircraft ran down the shorter-range directions provided by the carriers’ HF direction finders. The U.S. offensive was unpopular with the British, who feared that the Germans would realize that their codes had been broken, but it proved quite successful (and the Germans never realized why). The U.S. Navy became interested in offensive ASW based on ocean surveillance, an important idea postwar.

  In the Pacific, escort carriers were used to ferry replacement aircraft (including Army aircraft for landings) and also to support the Marines directly in the later island battles.

  By fall 1941, it seemed obvious that the United States would need many more fleet carriers. An Essex would probably take three years to build. Any rapid expansion would have to exploit ships whose hulls had already been laid down. Fortunately the high performance built into new naval aircraft and the new catapults dramatically reduced the requirements any such converted carriers would have to meet. By 1941 the U.S. Navy was building numerous 10,000-ton Cleveland-class light cruisers. A study of a full carrier conversion showed that it would be both inefficient and lengthy. The idea was resisted by the General Board.32 After war broke out, the idea was reconsidered on the basis of the existing escort carriers. This time it seemed well worthwhile, and nine ships were converted into Independence-class light carriers (CVL). As measured by the number of aircraft per ton, they were far less efficient than the big Essex, but they were still a very useful expedient (the typical air group was twelve fighters, nine bombers, and nine torpedo bombers). Typically one light carrier worked with three fleet carriers in a task group. Once the night carrier concept had been developed, a light carrier offered a full complement of night fighters without complicating the operation of the big strike carriers. The light carrier was considered so attractive that late in the war two ships (Wright and Saipan) were designed from the outset for this role, using adapted heavy cruiser hulls (which had not been laid down as heavy cruisers).

  TRANSITION

  The period through 1945 produced a particular type of carrier for both strategic and technological reasons. The strategic reason was that, through this period, the U.S. Navy was designed to seize and then exploit sea control. It was therefore conceived much more to attack an enemy navy—the Imperial Japanese Navy—than to strike at land targets. Aircraft were shaped by that requirement. It was entirely reasonable to expect a dive- or torpedo bomber to take off within the 400-foot length available forward of parked aircraft. U.S. naval aircraft were expected to attack some land targets, particularly when supporting Marines going ashore, but that was very much a secondary role. Perhaps it should be added that even after the Royal Navy regained direct control of its fleet air arm in 1939, it still had to avoid encroaching on the land attack role espoused by the Royal Air Force; there really was a distinct U.S. Navy view of the role of naval aviation. That view would become very important after 1945. In 1945, all existing U.S. naval aircraft could operate from all existing carriers. By 1945 the U.S. Navy was interested in larger carrier aircraft, such as the twin-engine Grumman F7F Tigercat, which would not have been able to operate from small carriers; but it was still true that carrier dimensions and speed shaped naval aircraft, not the other way around. It is not, incidentally, true that some aircraft, such as the Tigercat, were conceived specifically to operate from the big Midways—they had not yet been designed when the aircraft specifications were laid out.

  The foundation built between the wars made it possible for the U.S. Navy to shift toward a carrier-centered World War II fleet. The huge prewar U.S. naval air establishment was relatively easy to expand to train tens of thousands of new pilots and other personnel. It also trained the senior officers to command a much-expanded carrier fleet. In 1941 the U.S. Navy had seven fleet carriers and one escort carrier. By the end of the war, the U.S. Navy had over a hundred carriers (although most were quick and relatively inefficient conversions of merchant ship and cruiser hulls).

  By the time the Imperial Japanese Navy was essentially gone, in 1945, the U.S. Navy had become interested in a new mission, strategic air attack. It was not entirely new: prewar fleet exercises did show valuable potentials for supporting amphibious landings and for attacking enemy shore installations (the U.S. carriers often raided the Panama Canal, Pearl Harbor, and Los Angeles), but they were secondary.

  That year the U.S. naval staff studied the contribution that carrier aircraft could make to the bomber attack on Japan by B-29s. It concluded that given the sheer number of carrier aircraft, and the short ranges from which they could be launched, carrier bombers could deliver up to 60 percent of the tonnage available from B-29s. Against this calculation, a January 1945 carrier raid on Tokyo was less than successful, the small carrier bombers finding the high winds over Japan too great a hindrance and probably also finding it difficult to identify targets from the high altitudes then used. They might have performed differently a few months later, when the heavy bombers shifted to incendiary raids mounted from relatively low altitudes. In that case the sheer number of carrier bombers might have made Japanese defense far more difficult. The other advantage of carrier attack was that a carrier could mount attacks from a far wider arc, making it much more difficult for a defender to guess where to place defenses.

  Navy interest had probably been spurred by the April 1942 Doolittle Raid, when USS Hornet launched sixteen Army B-25 medium bombers to attack Tokyo. They did relatively little damage, but the raid convinced the Japanese that they had to destroy the U.S. carrier force. Their attempt to force a decisive battle, at Midway that June, proved disastrous for them: they lost four carriers, which their limited industrial base could not easily replace. U.S. industrial capacity could more than replace the carriers lost in 1942; newly built U.S. warships dominated the Pacific War from 1943 on. Even before the Doolittle Raid (but probably inspired by planning for it) the U.S. Navy was sponsoring a Grumman design for a carrier-based medium bomber comparable to a B-25 (the TB2F). Although this project died in April 1944 (surprise air attacks were unlikely to succeed now that the Japanese had early warning radar), the idea of heavy land attacks mounted from carriers became very important postwar, largely shaping the new carriers.

  NOTES

  1.For more details of ships and of carrier designs see the author’s U.S. Aircraft Carriers: An Illustrated Design History (Annapolis: Naval Institute Press, 1983), and Thomas C. Hone, Norman Friedman, and Mark Mandeles, American and British Aircraft Carrier Development, 1919–1941 (Annapolis: Naval Institute Press, 1999). For British development see also this author’s British Carrier Aviation: The Evolution of the Ships and Their Aircraft (Annapolis: Naval Institute Press and London: Conway Maritime Press, 1988).

  2.On 7 October 1915 Captain Mark L. Bristol of the Office of Naval Aeronautics (predecessor of the Bureau of Aeronautics) reported that the British had converted several ships into carriers; he suggested both converting a U.S. merchant ship and considering building a special aircraft ship. Asked for characteristics, the General Board demurred pending the results of the North Carolina trials (NARA GB [General Board] 420-7 in RG 80; file 28 October 1915). The General Board was responsible both for war planning (until formation of the OpNav War Plans Division) and (until 1
945) for drawing up the characteristics of ships to be built. Given preliminary characteristics, the C&R preliminary design section (and then of its successor the Bureau of Ships) prepared “spring styles” (names after women’s clothing styles appearing in the spring) from which the board could choose as a guide to preparing final characteristics. The board also held hearings on characteristics, transcripts of which (almost all from after 1917) have been preserved.

  3.GB 420, 17 January 1925, enclosing the report of the special board appointed by the Secretary of the Navy on 23 September 1924 to review the needs of the Navy. The huge number overstated the importance of aircraft. It reflected their rapid development (often a model went from concept to production in a few months), which caused each combatant to buy generation after generation of aircraft. Aircraft were also quickly expended, so vast numbers built reflected much smaller numbers in service at any one time. By way of contrast, World War II was fought by no more than two or three generations of aircraft.

  4.The characteristics proposed by the Aviation Division on 23 June 1918 (in GB 420-7, folder 1916–1924) called for a hull at least 700 × 80 feet (about 15,000 tons) with a clear upper deck divided into a 250-foot forward part, a 150-foot amidships part, and a 300-foot after part, its bridge built over the deck so that airplanes could pass underneath, carrying a mast set amidships so as not to interfere with launching aircraft (as yet there was no thought of an island set to one side of the hull). Speed should be at least thirty knots so that the ship could work with battle cruisers. Uptakes should come up to the ends of the superstructure. Aircraft would be stowed, their wings folded, on the two lower decks and carried up to the main deck. Armament would be limited to four 4-inch anti-aircraft guns at the corners of the superstructure (these guns were suggested because they would be powerful enough to deal with both aircraft and with surfaced submarines; no such dual-purpose weapons then existed, but the idea suggests the extent to which a larger number of single-purpose [AA or surface] guns would have complicated the ship’s arrangement). A sketch produced by the Bureau of Ordnance (not the constructors) in July 1918 shows a substantial superstructure about three-quarters of the way forward, bridging the upper deck. This seems to have been much the concept embodied in HMS Furious, then being rebuilt as a full carrier with separate landing-on and flying-off decks connected by awkward paths (the U.S. solution, the clear path under the bridge, would have been better). The power plant shown was a big diesel, which would produce less exhaust than a steam plant. A seaplane-handling crane is a prominent feature. Formal characteristics for a FY20 carrier, submitted by the General Board on 10 October 1918, repeated much of what the Aviation Division wanted, but called for funnels and masts to be placed out to the sides of the hull, perhaps hinged outboard. The bridge should be a special design, perhaps a fore and aft bridge outboard (i.e., an island) clear of the flight deck. Full speed was now set at thirty-five knots, the speed planned for the new U.S. battle cruisers and scout cruisers. Reviewing the proposed characteristics, Goodall pointed out that the ship should have good underwater protection and protection against cruiser fire, that her guns were too weak, and that she should have a torpedo battery to deal with heavy enemy ships that she might unexpectedly encounter (as she could not possibly fly off aircraft quickly enough). “Although such a ship should not by any means be regarded as a fighting ship, it should be sufficiently powerfully armed to be able to brush aside light vessels of the enemy, so that its machines can be flown off in comparatively advanced positions.” Goodall guessed that she should be about 800 feet long (22,000 tons). By May 1920 characteristics had been rewritten to show sixteen 6-inch/53 (i.e., light cruiser) guns in anti-torpedo (anti-destroyer) batteries, plus one twin torpedo tube on each side. Aviation chief Captain T. T. Craven wanted the ship designed so that “the entire allowance of airplanes can be placed in the air from stowage as quickly as possible.”