The best aircraft carrier of World War Two?

I enjoy doing these little discussions on topics that ‘everyone knows’ the answer too. (And frankly, arguing a highly debatable issue is always fun.) I particularly enjoy challenging what ‘everyone knows’. I may not get the perfect answer, but I often get a good discussion.

My favourite example of this is those who try to suggest that the M4 Sherman was the ‘best’ tank of the war. It was, at best, a functional design when it appeared in 1942, but had fallen behind the pace by 1943, and was certainly a deathtrap against the infinitely better German Panthers and Tigers. The only positive thing you can say about it from 1943 on was that it was available in such vast numbers that you could sacrifice several of them to get each Panther (pity about the crews…) But you can also say that about the Russian T34, which – as ‘everyone knows’ – was a contemporary and much better tank (certainly a better candidate for ‘best’ tank of the war). In fact the continued production of an obsolete model like the Sherman almost certainly had a negative effect on the prosecution of the war. Which means that the Sherman was a worse choice than the alternatives the Americans could have produced for 1944 and 1945 – not least for American crews.

So moving on to aircraft carriers. ‘Everyone’ knows that the Essex class was the best carrier of the war. Why do they know this? Because there were lots of them and they seemed to work pretty well. Oh, and none actually sunk, even when put out of action.

Did they have the biggest airgroup? No that was the Midway class (which in fact had a designed air group - 130 - that was soon realised to be too big for a single ship to manage in combat). Could they take the most punishment and remain operational? No that was the Illustrious class? Were they the longest surviving in service? No that was the Colossus and Majestic classes (some of which served into this century).

So what makes them possible contenders for ‘best’? Well they were bigger than anything completed to fit into the interwar treaties, because they were started after those treaties lapsed. This gave them an advantage over British or Japanese designs in pure space. They were fast, and long ranged. They were flexible, and could adapt to bigger aircraft. And they tended to survive bomb or kamikaze damage, even when gutted by fire, or after having their combat capacity removed by a single bomb damaging the deck or burning the air-wing.

What makes them not contenders? Well they could be put out of combat operations by a single bomb. They could be reduced to a gutted wreck that would need six months in drydock by just about anything that happened to hit them. (On 12 occasions Essex class vessels were seriously damaged by air attack, and what happened to USS Franklin and USS Bunker Hill are good samples of the weaknesses of the design.) For something so much bigger than the British or Japanese equivalents, they were suprisingly innefficient by comparison.

So what alternatives are there? 

Once again, that depends on when and where.

I hate statements that suggest that the best tank of the war was the Panther. It wasn’t there until 1943? By 1945 the Stalin, Patton and Centurion were all better. What can you sensibly say about best?

There is also the issue of where. British carriers are usually decried for having too small airgroups in European operations. But this is because the Pacific practice of having 50% of aircraft in permanent deckparks could not be considered in European waters. Once British carriers arrived in the Pacific they too had 50% deck parks, and HMS Indomitable for instance, with a hangar area 85% of that of USS Yorktown, operated 72 aircraft in the Pacific compared to the Yorktown’s 80 odd. Which means that British carriers could have operated bigger air groups in Europe, if anyone had considered that wise. 

No one did.

The difference being that if a bomb hit a deckpark, the planes tended to catch fire. In British carriers, even with Pacific style deck parks, that tended to be the end of the problem, and the other planes below deck could usually be used to continue operations after some concrete had been poured on any dents in the deck. In American or Japanese carriers, burning deckparks usually led to exploding ships, or at the very least months in the dockyards. Certainly continued air operations were not very common (though it did happen once or twice, which just shows that anything is possible, not that anything is actually likely).

So at what time and where are significant issues in comparing carrier abilities. Or perhaps, which designs were most successful for what purpose?

Early experimental carriers:

The carriers from the Great War and the 1920’s included HMS’s Argus, Hermes, and Eagle, INS Hosho and Ryujo and USS’s Langley and Ranger (and the French Bearn). All served during the Second World War, but mostly as escort or transport or training vessels. Those that were involved in operational duties tended to be easy targets, and HMS Argus in particular must have been a very lucky ship to have survived her various combat missions unscathed. The only one of these experimental models with a lengthy, and surprisingly successful, combat career was HMS Eagle, which fought for two years against the Italian and German navy’s and airforces, before finally succumbing to a torpedo in 1942. None of these vessels can count as very succesful as fleet carriers, but all were invaluable as escort, training or transport carriers. Given that they had to be operated within their limits, they were pretty successful carriers.

The interwar conversions:

Many of the early battles of the war was fought with interwar fleet carriers, a number of which were rebuilds of First World War battleships or battlecruisers. HMS’s Furious, Courageous and Glorious; INS’s Akagi and Kaga, USS’s Lexington and Saratoga. All of these vessels showed potential, but all had flaws. We will never know how effective Couragous and Glorious might have been, because both were early losses before their potential could be tested. This is a shame because the three half sisters had pioneered carrier group offensives in the Mediterranean in the 1930’s, and arguably three decks with 130-140 operational aircraft in Europe (would have been closer to 200 with deck parks in Pacific) was a better and more efficient (and survivable) option than most of the two deck alternatives operating similar numbers later in the war. Certainly Furious, with the smallest airgroup of the three, was still invaluable for most of the war.

The Japanese and American monsters were twice the size of British conversions, but all proved very vulnerable. They were too big and unmaneouvrable to avoid torpedo’s, and the sheer size of the air-wings made their refuelling and re-arming processes a disastrous weakpoint. Kaga and Akagi were incapable of withstanding bomb hits, and their aviation fuel systems were easily primed bombs. Lexington was lost not directly from a torpedo hit, but from the similar weaknesses in safety and damage control that were revealed thereby. Saratoga was astonishingly lucky to survive two torpedo hits and a Kamizaze attack, despite having been put into drydock for months on each occasion. It is a tribute to her improved safety procedures as the war went on, but it is notable that her aircraft capacity went down to about 70 (equivalent to ships less than half her size) as part of this improvement of safety. She also had only one working lift for most of the war, so her flexibility in combat was never as great as her size, or the size of her air-group, would otherwise suggest.

An interesting side point here is that the size of the two American super-carriers (and the tendency to put them on opposite sides in interwar naval games), means that the Americans failed to develop multi-carrier techniques until well into the war. At Coral Sea for instance the Americans ‘Task Forces’ were well separated, at a time when the British (who needed more smaller carriers for the same effect), or Japanese (who had bet everythig on a ‘hit harder and first’ strategy), concentrated their available carriers in defensive rings. The American ‘doctrine’ here was years behind the other two, which is an fascinating issue to grow out of such big conversions.

The ‘treaty’ carriers:

This group, mostly commissioned in the late 1930’s, are the most interesting. HMS Ark Royal, IJN’s Soryu and Hiryu, USS Yorktown and Enterprise. Again, all were quite competent vessels. Again all had weaknesses.

The Yorktown's were a good workmanlike design, but terribly vulnerable to fire. Unlike Japanese carriers, they did not blow up and sink when hit (in fact they were almost impossible to sink even when the USN tried hard to do so to damaged ones), but they could be reduced to un-flightworthy, and sometimes imobile hulks, by relatively minor damage.

The Soryu and Hiryu were fundamentally flawed in the weakness of their defenses. A single hit anywhere could convert them to floating bombs, just waiting for their own aviation fuel to finish their demise. They were the ultimate expression of attempting to use the biggest possible airgroup to hit your opponent first, and hope your oppnent never got a chance ot hit back. Unsuprisingly, they were revealed as time bombs when someone did get to hit back. Yorktown and Enterprise were somewhat tougher, but also capable of being put out of combat by a single hit. (Though with a much better chance of surviving to go into dock. )

Ark Royal was probably the best, and in many ways was the pinnacle of the interwar designs. A good airgroup of 60+ (even without deckpark!). Fast, maneouvrable, with an extremely powerful anti-aircraft armament, and enough armour to continue operations after the sort of the damage that usually sunk or drydocked Japanese or American carriers. Her main problem was the poor quality aircraft available early in the war (which was largely an issue of the Fleet Air Arm only coming back to naval control in 1938.) 

She served magnificently until late 1941, and, with her long term partners – the heavily modernised battlecruiser Renown and the cruiser Sheffield and their destroyer escorts – she pioneered the techniques later thought of as ‘Fast Carrier Task forces’. Had she survived until the availability of Wildcat's, Hellcat's and Firefly's instead of Fulmar's; and Barracuda's and Avenger's instead of Swordfish and Albacore's: her value would have increased even more.

She was unfortunately sunk by a single torpedo, possibly due more to the new Captain’s panicked attempt to evacuate the crew quickly rather than see what damage control could achieve. (Admittedly the suprisingly quick capsizing of the Great War design Courageous was in his mind.)

It would have been fascinationg to see what Ark Royal, with a deck park lifting her airgroup towards the 90-100 mark, could have achieved in the Indian Ocean or Pacific. She would certainly have been transferred to fight the Japanese in the Indian Ocean had she survived a few months longer, as she was more suitable there, leaving the heavily armoured Illustrious class in the more vulnerable Mediterranean fleets.  For a design limited to 22,000 tons by treaty, she was an astonishing achievement. All the sadder that her loss to a single torpedo revealed design and handling flaws that should not have been fatal.

The ‘compromise’ carriers:

The USS Wasp was squeezed into the American program to use up a few thousand spare tons of treaty allowance. Despite being a 1930’s effort, she suffered all the problems of the earlier experimental builds in being an overly optomistic attempt on inadequate tonnage. Her theoretical 70 aircraft capacity was based on not embarking any of the bigger torpedo aircraft, and in fact her airgroup was more like 60 in service anyway. Her protection was fatally flawed, and indeed, in the case of torpedo defense, practically non-existent. She served well as a transport carrier, but her use as a combat vessel was – as in the case of the early experimental designs – an act of unwarranted optimism by desperate superiors who should have known better. Torpedo hits caused the same uncontrolled aviation fuel explosions that bedevilled other Japanese and American carriers. She did not belong in fleet combat.

The Japanese equivalents of Wasp were the Zuiho and Shoho. Sneaky attempt to build carriers disguised as submarine depot ships. With small airgroups (no bigger than the old experimental carriers) they were really the forerunners of the American Independence and British Collossus class light fleet carriers, but without the survivability. As escort carriers they were successful designs. As fleet carriers they were not. By contrast their larger but slower sister Ryuho was treated as a training carrier, and can be considered a success in that role.

The ‘war is coming’ carriers:

The early wartime launches all tended to be modifications of designs based on the treaty limits. The HMS’s Illustrious, Formidable and Victorious, and the USS Hornet (a slightly bigger repeat of the 5 year old Yorktown design). The Hornet joined her two sisters in the glory of winning the battle of Midway. She also joined with Yorktown to demonstrated conclusively both the toughness of fundamental American constructions techniques (both carriers stayed afloat long after being abandoned, despite multiple attempts to sink them), and the vulnerability of American carriers to having their flight decks and air groups put out of operation by hits that would not sink them.

The Illustrious class were designed specifically to survive in close proximity to land based air power (ie European waters), and sacrificed airgroup to increase defensive capacity. The result was incredibly tough, and every one of the eventual six ships in the class shrugged off multiple bomb or Kamizaze hits, usually with little more effect than what one USN observer descrbed as ‘sweepers man your brooms’. (A huge effort by the Luftwaffe near Malta was the only time one of these carriers was put out of action, but she survived to fight many more campaigns.) On one occasion a Kamikaze attack caused fires which destroyed half the aircraft on one of the sister ships, but even then she still continued to mount operations. 

The cost was smaller air groups. Initially the class only operated 36 aircraft in the Mediterranean, though 52-57 was more common later in the war once deck parks were possible.

The ‘freed of limits’ carriers:

The later ships of the Illustrious class grew in size as treaty limits were removed, and Indomitable (72+ aircraft) and Indefatigable and Implacable (81+) were quite capable of holding their own with the Essex class in any battle-line. (Particularly given their superior fighter direction abilities, which the American Admiral  had noted when HMS Victorious served with the American Pacific fleet in 1943, and which continued to require markedly smaller CAP’s in 1945).

The Japanese response to the dropping of limits were the Zuikaku and Shokaku, ships often described as the best Japanese carriers of the war. (They weren't actually the best designs, but the better ones were sunk before firing a shot, so I suppose this is an acceptable generalisation.)

They were soon followed by the American equivalent, the Essex class. The Essex’s are renowned for their effectiveness, and (like the Yorktown’s) for their resistance to sinking. But they were remarkably easy to disable by fire, and could be put out of action by single bombs. By contrast the Shokaku class were actually well designed to absorb battle damage (for Japanese ships anyway), and Shokaku was bombed and survived for repair on two occassions (though surviving remained quite different from continuing combat operations after damage).

The Japanese proved that the Shokaku class were not a fluke by designing the magnificent Taiho (my choice for the best Japanese design), which was pretty much a non treaty enlarged copy of the British Illustrious class. The Japanese adopted the British style armour and enclosed 'hurricane' bow that had proved so useful in European waters (and that the Americans were later to adopt in their Midway class). The resulting vessel was probably a better design than the contemporary Essex’s, but unfortunately still had the damage control and aviation fuel issues that the Japanese never overcame. A torpedo could have been survivable had the ship not been in hard action, where, once again, the aviation leaks were not overcome.

It is interesting to compare the American Essex, Japanese Taiho, and British Implacable classes. They finally come to ships of a similar size, with similar capacities. The American ships are a bit larger, and were rated for 91 aircraft. The Japanese were middle sized, and carried between 65 and 84 (never actually served in combat so debatable). The British ships, despite being smaller, were much tougher, and (with deck parks) carried 81 aircraft. Of the three types, the Essex’s were best for Japanese style ‘hit and run’ operations (though Taiho might actually have been better), and the Implacable’s were best for ‘slug it out’ combat operations. (Ie: each nation achieved the best design for its preferred strategic approach.) However the limitation of the late expansion of the British ships design was lower hangers, which limited the types of aircraft they could carrry later in the war. So they had less ability to adapt than their earlier sisters which had higher hangers.

Wartime emergency carriers:

(We will leave aside the escort carriers. Good little ships for convoy escort and aircraft transport, but not suitable for fleet work. The most effective combat trole was to use them as floating airbases to cover invasions. This worked well for the British in the Meditteranean and Indian Oceans, and for the Americans in the Pacific. Some of the British, and most of the Japanese escort carriers were conversions, and most of the American ones purpose built. Within their limits, they mostly did quite good jobs. It is worth noting however that the British were horrified by the lack of fire safety on American escort carriers, and insisted on refitting them to higher standards before using them. This may have been a ‘European waters’ thing, but most American escort carriers when hit tended to respond in the unfortunate fashion of Japanese carriers – see Liscombe Bay for instance – not what was expected of American built ones. Only the 4 Sangamon class conversions from fleet oilers actually demonstrated combat survivability – see USS Chenango.)

So the American version of a ‘wartime emergency’ design was the light fleet cruiser conversions known as the Independence class. These though, were forced on a reluctant navy as a stopgap even before combat began. President Roosevelt became concerned that not enough fleet carriers would be available to cover combat losses while waiting for the arrival of the Essex’s, and demanded compromise vessels. They were too fast and expensive to be ecort carriers, and too small to operate on their own as fleet carriers (and had so little ammo storage that munitions were often carried on the hangar deck - even less protected than Japanese carriers!). But at a time when the Americans had to beg or borrow British carriers to stage diversions in the Indian Ocean or provide an extra carrier for the South West Pacific campaign, they were useful stopgaps. But stopgaps they remained – for good reason, see the loss of USS Princeton – and they were phased out as quickly as possible when real carriers arrived.

The interesting thing of course is that the ‘emergency wartime’ designs were decided by when the nations entered the war. Britain had to start earlier, and therefore reworked an earlier period design. Japan had more time to adapt, and a large part of the timing of Pearl Harbour was based on when the Shokaku class would be ready. The US by contrast didn’t finalise its post Yorktown design until too late to have new ships ready for war. (The name ship Essex, even with wartime pressure on construction speed, did not arrive to see action until May 1943!) As a result the Americans settled on the Essex design (or a slightly enlarged version) for most of the rest of the war.

The British and Japanese by contrast, experienced the shortfalls of their older designs early enough to start designing new vessels after actual combat and operational experience.

The Japanese emergency ship is the most amazing. They took an incomplete Yamato class battleship, and tried to build a huge version of the British ‘aircraft support carriers’ like HMS Unicorn. The resulting 66,000 ton monster – the Shinano – had a theoretical capacity of 120 aircraft in a hull armoured like a battleship, but in fact was designed ot operate maybe 50 aircraft, and have vast workshops and stores of supplies to support other carriers. The much smaller British versions worked brilliantly, so possibly Shinano would have too, except that she was torpedoed and sunk while still incomplete.

The British went in the opposite direction, and developed proper ‘light fleet’ vessels on the basis of the success of the Unicorn. They had discovered that 30+ knotts speed was not vital for most fleet work, and 25 knotts would be fine. Also, they knew that standing up for slugging matches was not necessary for most aircraft carrier work, so the short term ships could be built on merchant principles, not warship ones. The resulting 10 Colossus and 6 Majestic class carriers were an amazing success. A belief re-inforced by the fact that a dozen of them were in service into the 1970’s, and a couple lasted until this century. They became the ideal peacetime carriers for any nation that didn’t need or want (or couldn’t afford) large fleet carriers. They also became the model for the large number of amphibious assault ships, through-deck-cruisers, and V/STOL carriers built for many nations after these ships finally retired. Half a dozen of them were finished before the end of the war, including several active with the British Pacific and East Indies Fleets, but none actually engaged the Japanese before the surrender.

The late war carriers:

The need to rethink designs was caused by different things in different countries.

For the Japanese it was combat losses, and the recongnition of weakness that involved. The Shokaku class demonstrated toughness and survivability, but were too expensive and complex for emergency mass production. The Hiryu class were much simpler, and with a smaller and better protected aviation fuel supply, might have been more survivable. (Though the solution of pouring concrete over the fuel tanks of the resulting Unryu class does not inspire great confidence.) 17 Unryu's were planned, 6 were laid down, 3 were completed, but only 1 actually made it into service. Too little, too late of course, but that is more a reflection on Japanese industry than on the design. Nonetheless, as virtual repeats of the Hiryu class, they were really too light and vulnerable to take on even a pre-war Yorktown, let alone an Essex or Indomitable. They were possibly the best option, but they were never good enough for what was needed.

For the Americans, the need for changes also came from combat damage. As several Essex’s demonstrated the continuation of concerns over the flamability of American carriers, the fact that they usually survived to be repaired became less of an issue than the fact that a deck in a fleet action could be easily put out of action by a single bomb. The Americans started looking with envy at the British armoured flight decks, and Admiral Nimitz pressed to beg, borrow, or prefereably build, as many armoured carriers ASAP. (Particularly after the Kamikaze’s appeared). 

The obvious response was the Midway class, which – like the Japanese Taiho class – aimed to incorporate British design principles without the size limits of the treaty carriers. (A lot of the motivation being the inspection in the American repair yards of British carriers that had survived multiple hits by German dive bombers.) Unfortunately none were finished in time to see action. But these excellent designs served well for decades after the war, and would have undoubtedly been much more effective in surviving Kamikaze combat than their wooden decked predecessors, had Japan not surrendered earlier than was expected.

For the British, with the luxury of their armoured carriers surviving everything thrown at them, and living to fight again; plus an abundance of light fleet carriers entering service for the expected continuation of the Pacific war, the issue was simply one of getting bigger ships than the treaty limited or modified ships. 

First they enlarged the design of the Illustrious class to be equivalent to the Essex’s in size and airpower, but with greater weight to allow proper protection. The resulting Audacious class were not prioritised, and were put on hold when the war ended, and finished with many design changes at a leisurely rate post war. 

Then, they upsized the succcessful HMS Unicorn and Colossus designs to produce the Centaur class. Light carriers of the same weight as the treaty limit Ark Royal or Illustrious. (The Centaur’s were finished post war, but had long careers. HMS Hermes, now INS Viraat, is still flagship of the Indian Navy and potentially likely to stay in service until 2020 - more than 75 years after she was laid down!).

Finally they came up with their own version of the Miway class, but even a bit larger. The Malta class would have been the final development of wartime technology, and, like the Midways, would have been capable of adapting to steadily increasing aircraft sizes for decades. But they were cancelled uncompleted when it was accepted that there was non chance of them being needed. (Britain had dozens of carriers to potentially use against post war nations who had none, and only a single contemporary who had carriers to fight… their major ally. Much as some Malta class might have been useful in the unlikely event they still been around at the time of the Falklands war, the post war government was probably sensible to think that more than enough capacity was available in 1945.)

Conclusion

So although the Midway and Malta classes were the best carrier designs of the war, neither were really relevant to the war.

Although the Collossus, Majestic and Centaur classes were fabulously successful post war, they were not important in the war.

Although the Yorktown, Essex and Shokaku classes had significant operational flaws in terms of the ease with which they were put out of action, they (and possibly HMS Unicorn) were the best of the unarmoured carriers of the war.

Although the Illustrious class were the best battle carriers of the war, they were constrained in offensive air capacity by the effects of treaty limitations. (I am tempted to think that the Taiho, with its inclusion of all the best parts of the highly successful Shokaku class, PLUS British style armour and hurricane bow, PLUS a large airgroup: might have potentially been even better battle carriers. But the appalling Japanese record with battle damage played true for her as well?’)

Although the conversion carriers served well, they all showed significant limits, and most were lost due to them (and in the American case these also had a baleful effect on carrier doctrine).

The experimental carriers actually served very well, as long as they were not used for actual combat…

The usual balance for successful battleships is between speed to engage, armour for defense and gunpower for offence. For aircraft carriers it turns out to be between speed to engage and escape, air-group size for offense, armour or toughness for protection and defensive firepower. The best carrier of the war appears to have been the best compromise. A carrier with the air group capacity to do well in the Indian Ocean or Pacific, and the armour to be capable of staying in action after combat damage, and firepower and toughness to have a reasonable chance of surviving in European waters.

Given that every one of the wartime carrier had major design flaws, the best compromise appears to leave the Ark Royal. (I think that despite the fact that her particular design flaws led to her loss to a single torpedo, because, on balance, I believe a non explosive torpedo weakness seems a better choice than a highly explosive torpedo or bomb weakness. For a ship I would want to serve on anyway.) 

Frankly, until you get to the Midway or the Malta, they all had their own design flaw problems that made them less than perfect, so the idea that any wartime carrier was ‘ideal’ is dubious.

It is amusing that the Ark Royal, possibly the ideal carrier for the Indian/Pacific conflict, served her whole combat career in the far more dangerous European waters that she was less than ideal for. That she did so well there just emphasises her flexibility.