Real Battleships for WWII - Part I - defining a battleship

 This is the first of several long and quite technical posts, but I needed to do some background research as reference for a series of future articles.

Real Battleships for WWII - Part I - defining a battleship 

I recently enjoyed comparing a run of excellent You-Tubes on naval history and warships by Drachinifel, Navyreviewer, and by Ryan Szimanski (the curator of Battleship New Jersey Museum and Memorial). If you are interested in naval history, particularly World War Two, and haven't seen them, they are full of great information.

I was particularly convinced by Szimanski's definitions of battleships versus battle cruisers (or 'cruiser-killers' as Szimanski prefers to label them). He classifies anything capable of facing a fair number of hits from a real battleship without becoming instant Swiss cheese as a battleships – regardless of its speed or armament – and anything not capable of doing so as a 'cruiser killer'. 

Which fundamentally means he regards adequate armour to face roughly equivalent guns as the thing that lifts ships into Battleship class, by contrast to any amount of firepower and speed not making it past battlecruiser definition if the armour is simply incapable of taking hits from similar guns. 

(In fact battlecruisers were not even designed to face other battlecruisers, as they weren't designed to face their own guns. Battlecruisers were simply a way to use big guns to kill enemy cruisers. Putting them into a slugging match with other battlecruisers, lat alone proper battleships  – whether at Jutland or at Guadalcanal – was NOT their designed purpose... with inevitable results.)

This is actually pretty close to Admiral Fisher's earliest definition of the design concept difference, and much better than the British Great War practice of calling anything a bit faster a battle cruiser regardless of it's protection levels. 

Meaning Szimanski puts Hood and Scharnhorst in the fast battleship class, and ships like Repulse, Kongo, Dunkerque, Conti de Cavour, Deutschland or Alaska in the 'cruiser killer' class. 

All quite sensible.

However...

The more I listened to comparisons, the more I kept coming up with.. 'surely that can't actually be right'.

This caused me to do a bit of review of some of the facts, and to re-visit some of the over 50 titles of books – dating from 1938 to 2019 – I have collected on capital ships, as well as swot up on the latest Wikipedia information additions, and check a few recent academic articles. 

Some of these books and articles are truly dreadful, though the reasons vary. Some, particularly the pre-war and wartime ones, are simply poor because they lack information we now know. Some later ones are bad because they seem to be confusingly based on the prejudices of the commentators (worse, they then seem to try and do incorrect comparisons). And some, particularly the 70's titles and some articles in Wikipedia, are clearly just people who have managed to dismiss facts that don't fit with their world view.

I will note that there is so much contradictory information between sources, that most of the below is just the best I can make of many many alternative offerings. I will hope to get some helpful referrals to better sources from anyone who spots anything where they can suggest better information?

As usual the worst books seem to be the 'poor little us, weren't we terrible' school for British academic historians, Some of whom like being disparaging about their countries efforts (in that they are just prescient of most modern academic 'historians' I suppose). And some of whom are consciously trying to suck up to the new trend...

A good example of possibly both attitudes being British academic H P Wilmott, who blatantly admits in one of his Forewards... 

"It would be pleasing to record that this book first saw the light of day as a result of a conviction that the story of the British contribution to the war against Japan in 1944-45, and specifically the story of the British Pacific Fleet, deserved an account that did both justice. Unfortunately this author cannot honestly make such a claim.This book took shape as a result of the realisation that a doctorate, and with it admission to the most mysterious Masonic order in the Western world, would be required if the author was to work in the United States" ... 

In other words, he approached his topic dismissive of it's value, wrote what the thought potential examiners wanted to hear to get his Doctorate, and managed some quite convoluted internal contradictions between his quoted evidence and his conclusions... (but see my review post of his dreadful book Grave of a Dozen Schemes  for a reflection on his approach). 

The second worst sources are the 'gee, weren't we great' American school (though we are now getting plenty of 'weren't we actually awful' revisionists in the US too... but so far their fixations are race and politics, not so much technical detail). 

Then there are the 'wow, look at what the clever stuff the Axis technical geniuses achieved' school. A lot of it really impressive fantasy, particularly relating to the Germans. (Though, interestingly, the many Italian technical successes are often actually under-rated)...

All of these schools get a reasonable, and well deserved, kicking from both Drachinifel and Szimanski. But even those two sometimes contradict each other. (Sometimes they even contradict themselves in their own videos. For instance I note that by his own definitions, Szimanski is not curator of Battleships New Jersey, but of Battlecruiser New Jersey... but more on that below)

So let's define what a battleship really was in WWII.

To investigate properly, we have to go a bit deeper into the traditional trade-off decisions made when building a capital ship. Specifically, the balance that could be achieved between Armament, Armour, and Speed.

Armament.

Let's start with armament, but not just by looking at how big the main guns were: which is often a bit pointless, as most guns above 13.4" or 14" could usually defeat most other people's 10" to 12" armour, on most ships, at most likely ranges. (Only British battleships from Nelson onwards, the German Scharnhorst's and the Yamato class, ever had the higher quality or heavier 14" to 18" armour – of variable effectiveness –  that – probably – made the effectiveness of most of the older 12", 13.4" and 14" guns a bit more dubious)... 

Capital ships could also only defeat their opponents armour IF they could find them and hit them accurately. Which was hardly a given, as only the British – with the 8 WWI 15" guns on Warspite – and Germans – with the 9 modern 11"! guns on Scharnhorst  – actually hit any target moving at 26+ knots at 26,000 yards or more. (Both in relatively clear conditions in daylight, and both well inside the theoretical 35,000-45,000 yard full range of most battleship guns. Which would appear to make the concentration of too many technical articles on angles of penetration effectiveness of potential hits at 30,000 yards range a complete waste of time. 20,000 yards is useful analysis. 30,000 is just pointless, even with radar fire control.) 

[3 years later, in foul weather at night during the Battle of North Cape, the Germans had their fears that British radar had advanced far beyond theirs in gunfire direction completely confirmed, when Scharnhorst was pounded to scrap by Duke of York and a few cruisers in a battle where even the British cruisers could engage and score hits at ranges that didn't allow Scharnhorst to reply accurately.]

No other navy came even close to hitting anything actually moving at any speed at 26,000 yards (though West Virginia managed within about 15% of it with 22,800 yards against a target limited to 18 knots using the latest radar at Surigao Strait. in late '44). Certainly not the radar deficient Italians and Japanese. 

At night, or in bad weather, that meant radar efficient nations had an unsurpassable advantage, particularly for fast moving targets at sea like those hit by Warspite and Scharnhorst. 

[Although one USN 16" battleship peppered stationary French targets in a port at even longer ranges, some USN engagements – like Guadalcanal – were at Jutland distances, if not closer. (5,000-8,000 yards, even though the US ships had radar... Which would have been fine if the USN had been consciously doing a Matapan style 3,500 yard ambush, but South Dakota's radar and other power went down to 'electrical fault', and she accidentally wandered within 5,000 yards and was battered at close range, "leaving the ship in Lee's words "deaf, dumb, blind, and impotent". Fortunately Washington's working radar allowed her to sneak up on the Japanese and win the battle.) ]

At Surigao Straits those USN battleships with more modern radar - the late war rebuilds West Virginia, California and Tennessee – spotted the enemy at over 30,000 yards, and opened fire at 22,000 yards, actually getting some hits with the opening salvo's! But some ships with less effective radar – Maryland (8 16") – had to wait for visual sightings of shell splashes before joining in, and Pennsylvania (12 14") with her older Mark III radar, failed to spot the enemy at all.

How many and how big your guns are, or what their range is, doesn't matter a damn if you never see your opponent!

Yamato/Musashi also get mentioned for 'accuracy' sometimes, mainly because they sometimes got straddles at long ranges, in clear weather, in daylight. But they rarely actually hit anything. Even slow escort carriers were apparently a bit difficult, even in daylight conditions, with the escort carrier Gambier Bay appearing to have been lost to damage from a near miss from Yamato, rather than to an actual hit...) So apparently the good optics that allowed straddles were not good enough to allow adequate corrections for many actual hits? 

[Though please note, in every battle of the war, the navy which has the choice – either through speed in daylight, or radar superiority at night or in heavy weather – always closed the range to their maximum advantage (if not to point blank where possible) before engaging.]

So instead of endlessly debating the value of 10 or 12 14" versus 8 15" or 16" or even 9 11" or 16"; or of 20 degree versus 30 degree or 40 degree elevation; or of heavier slower shells versus lighter faster ones: I tend to accept that most heavy guns could penetrate most armour if they could find the enemy and get accurate shooting happening... and just wonder whether they could only hit anything in good light and good weather, or if they were completely blind in the wrong conditions?

Armour.

A vastly oversimplified theory is that 12" armour is OK to stand up to 12" guns, and – if it is sloped properly for some ranges – perhaps even for 14" guns: but it can't really be adequate for 15", 16" or 18" guns at most ranges. Theoretically that means about 14" of armour for 14" guns, and 16" for 16" etc.

Vastly oversimplified.

An unrecognised fact is that most WWI, interwar and even World War II constructed battleships, only had about 12" belts or less.

That includes almost every Japanese ship (except the 2 Yamato's with a theoretical 16.1"); the oldest 3 USN battleships;  and – uniquely for Washington Treaty or post treaty – all 8 of the final 8 USN battleships ever built! 

The Italians didn't even make it to 12" plate, and had nothing over 10.8" on the Littorio's, and 9.8" on the older ships. Nor did any of the French ships (except the Richelieu's with 12.9"); Even the vaunted Bismarck's were really 12.6" (though the 2 Scharnhorst's went to 13.8").

So not one of these ships was theoretically suitable to take on 15" or 16" guns! (Though an adequate 'slope' – from say 12 to 18 degrees – on 12" armour gave at least a theoretical 14" to 15" protection, at certain ideal distances. Note that most WWII battleship vs battleship engagements were definitely not at these theoretical ideal distances – Matapan, Gudalcanal, Barents Sea, etc – and Szymanski admits that Bismarck's 'outdated' 12.8" belt and turtleback was far more effective at the ranges of her final engagements than the Iowa's supposedly 'more advanced' 12.1" internal sloped belt would have been.)

Only British battleships – and about half of the older American battleships – had more than 12" of armour on anything they called a 'battleship': with the WWI model Queen Elizabeth and Revenge classes on 13"; the USN's  9  WWI 'standards' and the South Carolina's on 13.5" (though it's actual quality makes that number less impressive); the Nelson's and Vanguard 14"; and the King George V class on 14.7". 

(The British 'Battlecruisers' Hood had 12" like many of the better protected foreign 'battleship', but the Renown class battlecruisers only had 9" like the lesser protected French, Italian and Japanese battleships/battlecruisers.)

But here we will do a slight diversion into armour QUALITY as opposed to merely thickness. 

There is considerable discussion on various You Tube articles, about British battleship armour being 10% to 20% better than most other people's armour. (Garzke and Dullin - Battleships of WWII Janes 1980 p.247) say British battleship armour was up to 25% better than USN Class A, while Szimanski likes to claim 12%-15% better. Let's stick with about 20%.

But that is also a vast oversimplification. 

If you want a one hour video on this topic made sensible for lay people then Drachinifel's A brief history of Naval Armour is apparently admired by many metallurgy professionals for its concise summary. 

Simply put, the British Case Hardened just the front of the armour (hard but brittle front), and left the majority of its thickness 'ductile' (flexible); while the Americans Class A armour Case Hardened 50%-55% of total thickness, and therefore had much less ductile flexibility. The Germans were not far off the British approach; the Japanese tried to case harden parts just under the surface; and the Italians – possibly more sensibly – varied their approach depending on the thickness of plate. 

Interestingly the American approach worked extremely well on cruiser grade armour stopping cruiser grade (6" or 8") guns, but was much less efficient for over 12" guns; and the British were most effective at battleships guns, with performance falling off at cruiser level. 

[Which sort of makes sense in one way, as the American's and the Japanese thought of cruisers as being central to supporting their line of battle, while the British thought of them as commerce protectors with a limited scouting and AA focus for the fleet. (Literally British heavy cruisers like County's and York's were usually off on isolated patrols, while lighter or AA focused cruisers like the Dido's and the Cerese/Carlyle's usually worked with the main fleets.) But in the other way it makes no sense, as the Americas and Japanese wanted their cruisers to help stand in the line against enemy battleships, whereas the British cruisers were specifically designed to face other cruisers.]

In practice this meant that for Cruiser armour, the American Class A was the best, followed by Italian, Japanese, German and British last. By contrast, for Battleships, British armour was best, followed by German, Italian, and probably American just pipping the Japanese. 

[In fact the Japanese found that the thicker the plate, the harder to use their 'under the surface' cooking process, and in reality the theoretical 16.1" plates for Yamato were actually quite brittle; and prone to crack... though USN tests post war point out that sheer bulk makes up for a lot of the weakness, so 16.1" brittle belt armour (Yamato) is probably at least as good as 12.1" Class A (North Carolina/South Dakota/Iowa), for stopping things, though probably considerably less effective than 14.7" British armour (KGV) which should, theoretically, have been proof against 16-18" guns... theoretically. (Apparently USN point blank range tests agains a 26" plate – read a Yamato turret front – post war showed that it cracked surprisingly easily when hit by heavy shells. Szimanski shows some nice pictures of such plates shattered by 16" shells at point blank range... But the USN apparently concluded that such plates were proof to their 16" guns at most battle ranges... but obviously belts and decks of much lesser thickness would not have been...) ]

Theoretical thickness is way less important than actual strength, particularly if the armour is layered and/or poorly 'cooked'. 

Which means that the supposed 16.1" main belt of Yamato was almost certainly vulnerable, at least to the 16", 15" and probably modern 14" shells of possible opponents, even if the turret faces were 'proof'.

As usual which armour type you would prefer to have depends what you are doing. British or Italian cruisers facing each other would have loved to have USN Class A, but American and Japanese battleships facing each other would no doubt have preferred to have British or German armour. 

So in practice the 12.1" plate on the North Carolina and Iowa classes (apparently equal to about 9" or perhaps 10" of British armour) were considerably weaker (at certain ranges) against 16", 15", or even 14" battleship gunfire, than: the 12" plate on the Hood or Nagato; 12.6" on Bismarck; 12.9" plate on Richelieu's – or indeed than the 11" plates of the Littorio's – and obviously appallingly inferior to the 13.8 plate on the Scharnhorst (which was more realistically equivalent to 16" of US class A). 

Given that even Scharnhorst's 13.8" main belt was penetrated by a modern British 14" shell from Duke of York at the Battle of North Cape, the inferior 12" belts of the vast majority of 'battleships' were probably not really good enough to face anyone else's 14", 15" or 16" guns in sustained combat.

Given that technical superiority in British armour production, even the Renown class's slightly upgraded 9" belt was apparently pretty close in effectiveness to the 9.8" - 10.8" of the older French, and Japanese ships, and the new build Littorio's, and perhaps more like 10.5"-11" in effectiveness compared directly to the 12.1" of many of the American and Japanese battleships. (Though the Admiralty - which did consider Repulse to be 'just a battlecruiser', would still have liked to keep the Renown's clear of anything with even WWI 14" or better, and Repulse's only half modernised deck armour was definitely weak for WWII conditions.) 

Still Renown's slightly more modernised armour stood up well to 2 hits from Scharnhorst's modern high velocity 11" guns in 1940, and there was probably not much greater risk in facing the 12.6" shells of the Italian rebuilt battleships.

I also note that, under the definition of Renown being a battle-cruiser because her 9" armour was not really good enough to face her own 15" guns (which I completely accept): then presumably you would have to say that the South Dakota's or Iowa's 12.1" Class A, (which works out at only about 10" effectiveness in British armour equivalent terms), was definitely not up to facing their own 16"/50 cal guns with super heavy shells? 

Szimanski repeatedly states that as a rule each inch of armour is required to meet each inch of gun (and that is without him discussing some nations inferior plate strengths). It is amusing to note that by his own definitions the Iowa's were most definitely Battlecruisers rather than Battleships!

Nonetheless, I would largely dismiss that, and say that just about any of the ships with 12" armour or better, in any reasonable layout, met the practical battleship standard for WWII – as opposed to battlecruiser vulnerability – when tested in actual combat. Frankly, with the exception of what Szimnski calls a 'golden beebee' shot – like the one from Bismarck that Drachinifel suggests lucked into a trough of water from Hoods bow-wave at just the right moment to get under Hoods belt – the common theme of battleships battered into submission in WWII is that they were usually reduced to floating impotence, with wrecked upper works, no fire control, and disabled main batteries, but still stubbornly afloat, and needing to be finished with torpedoes or scuttling charges. A sufficiently strong 'citadel' usually kept the ship afloat and the magazines protected, even whne the ship was battered beyond resistance. That defines a battleship as opposed to a mere cruiser killer. 

Which is why I go against Szimanski's self confusion here, and accept that 12" armoured ships like Hood and Iowa and Nagato are proper battleships even though their armour was actually pretty marginal against their own guns. (Not least because I don't fancy the shit-storm it would kick up to point out that only British battleships from Nelson on – and the Scharnhorst's of course – probably meet the test of being able to stand up to their own, or approximately equivalent, guns during WWII.)

To explore that a bit further.

Even the 13.5" plate of the USN 'standards' and the newer South Dakota's (equivalent to perhaps 11" of British armour) was probably still inferior protection to the 13" of the WWI Queen Elizabeth or Revenge class (equivalent to say 14.5"-15" of Class A?). Let alone inferior to the 14" on the Nelson or Scharnhorst (equivalent to almost 17" of USN Class A), or 15" on KGV (say 18" of Class A... that's 50% more effective armour than a South Dakota or Iowa).

[Though please note, again, that we are just comparing plates thickness, not layout, and the more modern ships usually had a 'more modern' armour layout, and sometimes angled belts, to help mitigate some of this. In theory the USN could drop the 13.5" thickness of the 'standards' and the North Carolina's to 12.1" in the South Dakota's and Iowa's because – like the 12" Hood and 14" Nelson designs – an angled belt gave slightly better protection than pure numbers... at some ranges. Against this of course the RN dropped the angled belt as technically dubious after the Nelson's, and the only real test of the 12" angled belt of ships like the Hood, Washington and Iowa, saw Hood sunk... Wonder what would have happened to South Dakota had Washington not snuck up behind Kirishima...]

In fact the 14.7" belt of the King George V was probably considerably superior to the 'poorly cooked' and brittle' 16.1" plate on Yamato. The Vanguard actually dropped to a 14" belt, but still probably still had superior armour protection to any other nations battleships, Yamato included. (Again, only the Scharnhorst's 13.8" belt really came close to modern British belts, and it is possible the Scharnhorst's too had better functional armour than the Yamato's? Though I probably wouldn't want to try taking on a Yamato with Scharnhorst's original 11" guns!)

Again, theoretical armour thickness is way less important than the quality, and layout, of the plate.

Speed.

In September 1939, or indeed in December 1941, Britain's 5 only partially modernised Revenge Class battleships, like the entire US 15 ship battle-fleet, and all the older French ships, would still be only 21 knot designs (if that, most older ships maintained a bit less real speed in practice).  

But I do not believe that this comparatively slow speed made them all 'useless'?

It is fair to say that all the older 12" and 13.4", and even many of the 14" ships still in service: were all outclassed by most rebuilds (I will cover this further in the next post), and certainly by any new builds larger than Duetschlands and possibly Dunkerques. 

But the USN's Big 5 (2 of them 14" and 3 16"), and the RN's 5 Revenge's (15") were all quite capable of standing up to most later battleships in a slogging match, particularly at closer ranges... 

As long as the faster ships wanted to engage...

All these 21 knot ships were simply too slow to have any say in whether they got to engage or not. Except for radar assisted night-time ambush (the battles of Cape Matapan, Gaudalcanal and Surigao Strait spring to mind), that was pretty much the choice of the faster enemy ships (Calabria for instance). However time and again all these slower ships still caused their faster juniors to beat a retreat rather than engage. 

From Atlantic convoy escort to Mediterranean fleet actions to the final battle of the Surigao Strait, the older slower ships repeatedly fulfilled their defensive missions despite their vastly inferior speed: by making faster Scharnhorst's and Littorio's and Kongo's back off rather than risk the sort of damage that might endanger their survival.

For comparison the 'mid range' speed Nelson's, Queen Elizabeths, Fuso's and Ise's were about 23-24 knots. (Modernised Queens dropped half a knot, due to greater weight and draft, and the Ise rebuilds increased a knot, mostly because they, like the Kongo's, only got limited armour increases... but all remained somewhere in the 23-24 'mid' range, except that apparently the Fuso's were down to about 18 knots by the time of their loss at Surigao Strait.) 

Only the Nagato's and rebuilt Italian ships of the pre-30's designs could get to 'a bit fast' 26+ knots, except of course for the very fast 32+ knot Renown, Repulse and Hood. 

Of the new build 35,000 ton treaty battleships - KGV, Washington & South Dakota were 'fast' 27-28 knots. The Nelson style Richelieu design actually allowed a good balance or guns, speed and protection (actually 32 knots despite considerably better armour than Washington & South Dakota). The much heavier (ie: not Treaty compliant) Littorio's (41,000) and Bismarck's (42,000), all hit the 'very fast' 30+ knots,  without managing such a good balance. And the super-heavy Yamato's (63,000) even dropped back to the 'fast' 27 knots. 

The 'Treaty Escalation Clause' upgraded Lion's (40,500) would have been 'very fast' 30+ odd knots too, and the post treaty Iowa's (48,000) of course threw all treaty limits out the window, but stayed fairly lightly armoured, to get to 32+ knots. (At least in good weather, or for short bursts. No USN battleship could manage it's 'Pacific' designed speed in heavy 'Atlantic' conditions, and Szimanski is clear the the Iowa's in particular suffered too much from excessive vibration in the long narrow bow to maintain top speed for long even in light conditions. Apparently there were proposals to rebuild the Iowa's with shorter bows to solve some of this, even at the cost of lower top speed.) 

Amusingly the Iowa's finally come quite close to matching the stats of the WWI designed Hood in a new iteration of a balanced fast battleship with size, speed, armour and firepower. for about 6,000 tons extra the Iowa's finish up with slightly bigger guns, similar speed (in good weather), but slightly weaker armour protection. (Given that both had the interwar experimental angled belts... though the Hood's deck armour and torpedo protection definitely needed its planned modification by WWII, which would have lifted it to closer to the Iowa's final tonnage, to match the Iowas in those areas. Fascinating comparison.)

Unsurprisingly the Vanguard also came out with almost matching stats to the Iowa's... except that the British ship had considerably better armour, as well as superior seaworthiness compared to the American ships. (Meaning Vanguard's theoretically slower speed in flat calm actually became slightly faster speed in rougher conditions - see Vanguard and Iowa on post war NATO exercises!). 

Interestingly, like the 63,000 ton Yamato's, the design for the 62,000 ton Montana class would also have dropped to about 28 knots due to a 4th triple 16" turret and heavier armour. (Armour more akin to the Vanguard's in effectiveness).

Realistic Comparisons

It is interesting that the WWI Revenge's only received one interwar modification, not two, and are almost always referred to by British writers as 'old', 'obsolete', or 'too slow and vulnerable for modern conditions'. Many call them 'ships the admiralty tried to keep out of the front line'. 

Presumably this is what people's biases encourage them to think? Ships of this class engaged German, Italian and even French battleships and cruisers – both individually and in line of battle – on many occasions during the first 4 years of the war, and 4 of them working as a squadron actually came within a couple of hundred miles of engaging the 4 Kongo's in a squadron in April 1942. But the Wikipedia article on Royal Oak for instance actually states...

"Attempts to modernise Royal Oak throughout her 25-year career could not fix her fundamental lack of speed and, by the start of the Second World War, she was no longer suitable for front-line duty."

This is amusing as they had the same speed as any of the French veterans, and as all of the American pre-30's 'standard' battleships. In fact the Americans did not get any battleship faster than 21 knots into service until the 2 North Carolina's arrived  in mid 1941! Every American 'standard', and even the old 12" dreadnought Arkansas, were apparently considered suitable to face German raiders in the Atlantic, and they did face the IJN at Leyte Gulf. What supposedly made the Revenge's less capable?

In fact even relatively unmodernised Revenge's – which frankly the Admiralty didn't consider worth the investment of rebuilding in any major fashion – still matched speed with most potential rivals, and actually outclassed most of their foreign 'contemporaries' in both protection, and armament.

[Admiral Cunnigham was certainly right to say that the slower speeds and unmodernised elevation/ firing ranges of the Malaya and the Royal Sovereign made it impossible for them to catch the Italian ships fleeing at the Battle of Calabria (where most people apparently agree that Warspite's hits on Guilio Cesare that slowed her considerably would have left the Italians at his mercy had they been able to come within range). If he had had a Hood, Renown or even Nelson in company instead of those two, he probably wouldn't have decided to circle twice to let the slower ships catch up, and thus given the Italian fleet time to escape. Fair enough. But the key word is 'fleeing'. the two Italian cruiser killers were effectively being defeated by Warspite alone, and Admiral Campioni certainly could not risk letting the other two British battleships get close enough to destroy his fleet. Too slow to allow a decisive win, didn't mean too useless to make the enemy flee. Obviously Cunningham would have done better with a couple of faster ships in support, and he appealed for – and received – more modernised Queen Elizabeth's. But it's a bit unfair to rail against Royal Sovereign for this inadequacy of speed when there was still not a single US or French battleship that would have done any better in the same circumstances. In fact the new Dunkerque cruiser killers were still – 9 July 1940 – the only other ships in either the US or French fleets that could have gone any faster... (though Richelieu was within a week of commissioning by then...)]

With the possible exception of deck armour (where only Royal Oak had completed a second upgrade pre-WWII designed to make their deck protection equal or superior to any likely foe), the 13" belt Revenge's were simply more heavily armoured than any foreign battleship until the quarter century later Scharnhorst, and Yamato classes. That armour superiority included advantages over every French or Italian battleship; over most USN ships including the latest South Dakota's and Iowa's; and over the Japanese Fuso, Ise or Nagato class (the last two named classes the only other navies WW1 designs actually 2 and 5 whole knots faster respectively than the 'too slow' Revenge's). 

The Revenge's 8 15" guns were also superior to the 12", 12.6", 13.4", and most 14" designs in other navies (though the 12 14" gun designs of the latter USN WWI vets had theoretically heavier throw than 8 15"), and probably outclassed in effectiveness everything except the 8 16" guns of the post war Nagato and Colorado classes (and arguably, there was simply not enough difference there to warrant expensive works). The fact that their main elevations was never raised is the main issue the Revenge's faced, but they were still quite competitive in the many Mediterranean sea battles they fought. The Wikipedia article on the BL 15" Mark 1 /II –used on the Revenge's, Queen Elizabeth's, Renown's and Vanguard – states:

"According to an American report produced after World War II, the British 15 inch Mk I was the most reliable and accurate battleship main armament of the war, though other guns and mountings had superior individual features."

Which means, again, that the issue is really the ability to hit the enemy when they possibly can't hit you at the same range (or perhaps even see you). And here British/American style radar superiority over most likely opponents also gave even the 'obsolete' Revenge's a clear advantage over likely Italian and Japanese opponents in the Mediterranean and Indian Oceans. (Even over their German opponents after 1942.)

So I think you can dismiss the 'obsolete' argument about the Revenge's with the same contempt you can show for the argument that the Iowa's and Bismarck were really only battlecruisers not proper battleships... Even the Scahrnhorst's light 11" guns don't exclude her from being a proper battleships (though the planned upgrade to 15" guns – that Gneisenau actually started to receive in her mid war rebuild after damage – would have helped a lot, particularly if they needed to face a Nelson, KGV or Yamato).

But you can't dismiss the problems with the Renown or Kongo's or Courbet's or Dunkerque's armour; or the Conte de Cavour's or Arkansas' guns; or the overall inferiority of the Deutschland's and Alaska's to any real battleship. These ships were simply not capable of facing any proper battleship in sustained combat, and do not belong on a list of proper WWII battleships.

Real Battleships in WWII included:

4 German - Bismarck's and Scharnhorst's (though the Scharnhorst's need the planned 15" gun refit to be really competitive); but definitely not the Deutschland's.

3 Italian - Littorio's, but not really any other Italian rebuild. (See next article again.)

4-5 French - Bretagne's perhaps, and Richelieu's; but not the Courbet's, and probably not even the Dunkerque 'cruiser killers'.

8 Japanese - Fuso's, Ise's, Nagato's and Yamato's; but certainly not Kongo's.

19 British - Queen Elizabeth's, Revenge's, Hood, Nelson's, and KGV's (and Vanguard); but not the Renown's.

22 US - Nevada's, Pennsylvania's, New Mexico's, Tennessee's, Colorado's, North Carolina's, South Dakota's and Iowa's, but probably not the New York's, and definitely not the Arkansas.

(Note that comes to about 45 real 'Allied' battleships versus about 15 'Axis'...)

But the interesting thing is which of these battleships was actually still competitive in WWII?

That will be covered in the next article, which looks at the value of the interwar and wartime re-building some of the WWI vintage ships.