Shattered Sword (the book) - some observations

[grumbler]

I know Berkut has read Shattered Sword http://www.amazon.com/Shattered-Sword-Untold-Battle-Midway/dp/1574889230, a 2005 book with yet another version of "the untold story of the Battle of Midway," and I suspect some others have as well, but I just finished it and thought it might spark some interesting conversation amongst the naval buffs hare.

The book really has three elements that the authors tie together to make a pretty compelling story:  (1) Japanese design and operational doctrine, which determined how they wanted to fight and how capable their ships, planes, and equipment was in both its intended and unintended uses; (2) what the Japanese specifically planned and did at midway to execute the doctrine to achieve real-world results, and (3) what happened to a number of individuals as a result.  The fourth element the book attempts, that of debunking earlier studies, is less successful, for reasons I will only briefly go into (reading the book will make this point more clear).

This book, rightly following in the footsteps of Lord, Prange, and Wilmott, focuses on the two key decisions that doomed the Japanese carriers in the battle: first, the decision to implement a scanty search plan on the morning of the battle, and, second, Nagumo's decision to hold off on attacking the Americans until he had recovered the morning Midway strike and completely rearmed the reserve planes he still had onboard.  The authors, rightly, I think, absolve Nagumo of the decision to rearm the reserve for a land strike.  That strike was needed if the absurdly tight Midway invasion schedule was to be kept, and Nagumo had no power to affect the schedule or the movements of the other forces counting on him to knock out the base.

The first decision was made by Genda, and approved by Nagumo.  It involved using just the six long-range floatplanes on the cruisers Tone and Chikuma plus a short-ranged plane from the battleship Haruna (which was basically looking behind them, and so didn't need to search as far).  The weather conditions required almost double that number of planes for complete coverage, but there were no more long-range  scout planes available.*  The only source for additional planes was the carriers, and Japanese doctrine said that carrier planes were for sinking the enemy.  Attack plane pilots objected to scouting, as it might cost them a shot at combat.  So, when conditions required the easy choice of inadequate scouting or the tough choice of telling pilots they might miss combat, Nagumo and Genda went with the easy choice.

* Tone and Chikluma should have had ten or even twelve long-ranged scouts between them, but production of those aircraft had not met expectations, and the cruisers started the war short one scout, and lost one each during the six months of combat operations.


More next post

[grumbler]

The question of exactly what Nagumo could have launched had he launched what he could as soon as the Americans were spotted takes up a lot of ink in this book.  The authors look into the re-arming matter more closely than any study I have ever seen.  They conclude that Nagumo would have had 30 torpedo bombers and 36 dive-bombers, plus whatever of his 40 Zeros he decided he didn't need for CAP.  They base this on their own idea that the Japanese would be very slow at re-arming their torpedo bombers, taking 35 minutes for each 1/3 squadron reaming (the max that could be done at once, based on the limited number of carts that could carry torpedoes).  If this were true, then only an idea would think this an insufficient force to throw at the enemy, and this whole debate represents the one place I think the book starts to break down.  Pretty much every other source on this, western and Japanese, accepts the estimates of Genda and Fuchida that more than half the planes had been armed with bombs, which then means that Nagumo's hesitation to strike makes sense logically.  This book, though, wants to explain things doctrinally.  The authors therefore argue that Nagumo failed to strike immediately because doctrine called for full-chutai launches, and he didn't have any full torpedo chutais to launch.

This argument is certainly persuasive in a number of ways, but relies on purely speculative estimates of how far the Japanese had come with their reaming and re-rearming, and the authors seem too anxious to stake out a claim for original research.  This particular answer made my spidey-senses tingle.

One of the things the authors do quite well, and that strikes a resonance in me, is their discussion of just how hard it would be for Nagumo to reach the optimum decision about what to do when his Midways strike was about to return, his reserve planes were re-rearming, and time was ticking away.  As they noted, Nagumo was not himself an aviator, nor was he happy in his command (whose technical features bewildered him).  However, Nagumo's staff was not in a position to guide him to the right decision because Nagumo's only command post was the tiny (15'x12') bridge of the Akagi, which was also, of course, the command post of the Akagi's CO.  Any conversation between Nagumo and his staff would, of course, be overheard by an Akagi crewman at their elbows, and so the staff was pretty much forced to say little other than "hai!" unless specifically asked a question by Nagumo – who was not a question-asking kind of guy.

There are lots of touches like this in the book; discussions about how the Vals could only use the center elevator on the Japanese carriers because their wings did not fold; the revelation that American carriers could operate aircraft so much more rapidly than the Japanese because American hanger bays were open to the outside, allowing the engines to be warmed up on aircraft that had not yet been sent to the flight deck.  The description of the Japanese concept of the carrier division (two carriers operating as a single unit) was not something new to me, but the discussion about what that allowed and implied (the Japanese had no concerns launching a strike using the torpedo bombers from one carrier and the dive bombers from another – in fact, at Midway CarDiv 1 provided the dive bombers for the morning strike and CarDiv 2 the torpedo bombers, while the fighters came equally from all of the carriers) was new.  Symmetrical numbers were important to them:  it was no accident that the Midway strike had exactly 36 fighters, 36 dive bombers, and 36 torpedo bombers.  I am not sure that this significance would amount to a compulsion, as the authors argue, against launching only 30 of 36 torpedo bombers, but it is interesting.

[grumbler]

Some other things struck me:  the authors are unaccountably kind to Frank Jack Fletcher, for instance.  They scrutinize every decision made by everyone else, and criticize where criticism seems warranted, but Fletcher's decision to short the morning Yorktown strike by a full dive-bomber squadron without even telling the strike leader was a major blunder, because it allowed Hiryu to come through the attack unscathed.  It is kind of ironic that Fletcher would lose his own flagship as a result of this decision, but the fact of the matter is that if Fletcher thought the chances of another carrier being out there warranted shorting the strike, he should have held back Yorktown's entire group.  The one squadron he did hold back, it was quickly obvious, couldn't do anything on its own.

The loss of Akagi to a single bomb hit has always been one of those issues that leave one wondering, and this book does far more than any other to explain just how that happened (and, in fact, why Scouting Six ended up mostly attacking the Kaga when Bombing Six was doing the same).

The authors spent a long time, rightfully, on Japanese carrier defense doctrine, which was one of the spectacular failures of the battle.  Their doctrine held that CAP was the main defense of the force, but they put almost no effort into making sure that their CAP could do so.  The radios in the Zero were weak and frequently broken, and could only tune in to one frequency (which was the same frequency every other plane in the carrier force used!)  Essentially, the CAP was launched and just expected to be where it needed to be on its own.  Further, though the planes had terrific range, it had only about six seconds' worth of 20mm ammo, and this was the ammo needed to bring down Devastators and Dauntlesses.  The Japanese carriers were constantly landing and launching CAP, mostly due to ammo limits.  Also, there was no one coordinating CAP levels for the force.  Each carrier maintained its own CAP levels.

If aircraft got through the CAP, there was no defensive formation of escorts to blast at them with AA – each Japanese ship was expected to provide its own AA, including the carriers.  The Japanese battleships were more than four miles away from the carriers.  The carriers had a fairly formidable set of AA teeth, but few directors.  Generally, a carrier couldn't expect to engage more than three aircraft at a time.  In the whole battle, only 2 USN planes would be downed by flak.

The real defense of the Japanese was maneuver.  That worked a whole lot better for Hiryu and Soryu than Akagi and Kaga, though.  Once the CAP broke down, CarDiv 1 was basically doomed, unless it had gotten its planes in the air – which it couldn't do while constantly having to tend to its CAP.

[grumbler]

Final bit:  the authors conclude that Japan lost the Battle of Midway because their doctrine wouldn't allow them to win it.  They say that, if someone other than Genda was in Gend's shoes, he would make the same decisions as Genda, and ditto for Nagumo.  This is almost a Marxist view of history, and while it makes a great deal of sense in the broader strokes, it is not so compelling in the specifics.  There were people telling Nagumo to strike the Americans as soon as they had been detected.  There had been times when the Japanese used strike planes to scout with.  There was no doctrine that called for perfectly symmetrical strike packages, and none that said strike planes could not be used as scout.

I think that the older explanations for Nagumo's and Genda's decisions is still the correct one: Genda and Fuchida's concept of "Victory Disease."  More scouts were not used because they were not needed, because the Americans were not out there yet, because in the plan they only sortied from Pearl Harbor when the attacks on Midway began.  There was no particular urgency in attacking the Americans on Nagumo's part because the Americans couldn't really hurt the Japanese fleet anyway.  The plans of the Japanese had always succeeded, and therefor they always would.

This is an excellent book, though, and I recommend it highly to anyone with an interest in the topic and a modicum of carrier savvy (enough to know what a CAP is and how the Japense designated aircraft by year marks).  The authors assume a bit of such knowledge.

Thoughts?

[Jaron]

TL;DR

[Jacob]

Very interesting.  I have nothing of substance to contribute, but I enjoyed reading the posts

[Berkut]

I thought it was really outstanding  - I don't presume to be educated enough about WW2 naval doctrine to really critique it, so it was mostly pretty convincing to me - of course, I recognize that I am easily convinced, since I am no expert.

I very much liked their focus on doctrine, rather than imagining what "could have" been done, if in fact military organization did NOT function on doctrine. It is an approach I imagine could be used elsewhere as well - most critiques of particular battlefield performance largely ignore the doctrinal constraints involved.

[Razgovory]

A good book report Grumbler.  You need to remember to to italicize foreign words and put foot notes in after you present a fact from the book.  Good job over all.  B.

[grumbler]

I thought it was really outstanding  - I don't presume to be educated enough about WW2 naval doctrine to really critique it, so it was mostly pretty convincing to me - of course, I recognize that I am easily convinced, since I am no expert.

I very much liked their focus on doctrine, rather than imagining what "could have" been done, if in fact military organization did NOT function on doctrine. It is an approach I imagine could be used elsewhere as well - most critiques of particular battlefield performance largely ignore the doctrinal constraints involved.

Agreed.  I think it is possible to overdo the "doctrine" approach, but even easier to under-appreciate the impact of doctrine.  Overall, I think the book got the balance about right.

[Berkut]

I thought another good point, or at least a point I did not know beforehand, that the authors made was that the supposed numeric advantage of the Japanese was largely a myth.

While they did have 1 more carrier than the US, the US had as many planes, since ours were larger, and then when you add in Midway, we actually had an advantage in actual numbers of planes. And of course we had a single focus for our operation, while the Japanese were divided in theirs (between dealing with the USN threat and neutralizing Midway).

So the idea that this was a desperate gamble on the part of the US was largely bogus - quite the opposite in fact, this was a fight the USN very much wanted, and was fought under conditions that were very favorable to the US in many ways.

The wargamer in me always imagines the Midway fight as a great roll on a poor odds battle - but in fact it was probably just good roll on a good or even odds battle.

The other point that I recall being made was that the entire Midway plan was supposed to include the entire Kudo Butai - but after Coral Sea, and the loss of two carriers, the plan was not modified at all. They just did the exact same thing with 4 carriers instead of 6.

[Grey Fox]

@Grumbler, are you sure? 

[grumbler]

I thought another good point, or at least a point I did not know beforehand, that the authors made was that the supposed numeric advantage of the Japanese was largely a myth.

While they did have 1 more carrier than the US, the US had as many planes, since ours were larger, and then when you add in Midway, we actually had an advantage in actual numbers of planes. And of course we had a single focus for our operation, while the Japanese were divided in theirs (between dealing with the USN threat and neutralizing Midway).

So the idea that this was a desperate gamble on the part of the US was largely bogus - quite the opposite in fact, this was a fight the USN very much wanted, and was fought under conditions that were very favorable to the US in many ways.

Yes, this is true in hindsight (though in numbers of planes in the striking squadrons, the Japanese still had an edge, and in pilot quality and aircraft capability the edge was enormous).

The wargamer in me always imagines the Midway fight as a great roll on a poor odds battle - but in fact it was probably just good roll on a good or even odds battle.

Disagree.  Hornet was worth, at best, half the value of the Yorktown (and, given that I don't think Hornet's aviators landed a bomb on target in the whole battle, maybe not even that), because she was brand new.  The Midway air group was a mixed bag of rookies in brand-new planes and veterans in obsolete planes, none of whom had ever worked together.  It was worth far less than a real air wing would have been worth.  In other words, there were only two moderately competent American air groups at Midway, and both of them were seeing real combat for the first time.

The other point that I recall being made was that the entire Midway plan was supposed to include the entire Kudo Butai - but after Coral Sea, and the loss of two carriers, the plan was not modified at all. They just did the exact same thing with 4 carriers instead of 6.

True, though in return they knocked out the two most veteran US air groups.  Under the circumstances, the Americans might have been better off having the Lexington/CVW-2 and Yorktown/CVW-5, even given that the Japanese would get Zuikaku and Shokaku. The problem the Japanese suffered from was not lack of aircraft.  In fact, it was in a sense the reverse:  they had too many aircraft on board their carriers when the bombs started to fall.

Had CVWs 2 and 5 been available, it is certain that AirPac would have sent CVW-3 from the damaged Saratoga to Hornet to replace the green CVW-8 (as, in fact, he did when he sent CVW-3 to replace CVW-5 on Yorktown due to losses at Coral Sea).  Arguably, no Coral Sea would have doubled American striking power in the battle, given that only two of the three US CVWs were really operationally effective anyway, while only increasing Japanese strength by 50%.

As it turned out, the situation the US faced at Midway was much better than the Americans had any right to expect, so I agree that, in retrospect, the battle was not as "incredible" as some have made it sound.  The decision to accept battle under those conditions, though, and to commit to positioning the carriers outside Midway's support range (so as to make the ambush more effective if it could be pulled off) took a lot of balls.  In a sense, the "incredible" element of the battle is that the US Pacific Fleet had a commander who could make that kind of decision.  Those guys are rare.

[KRonn]

Final bit:  the authors conclude that Japan lost the Battle of Midway because their doctrine wouldn't allow them to win it.  They say that, if someone other than Genda was in Gend's shoes, he would make the same decisions as Genda, and ditto for Nagumo.  This is almost a Marxist view of history, and while it makes a great deal of sense in the broader strokes, it is not so compelling in the specifics.  There were people telling Nagumo to strike the Americans as soon as they had been detected.  There had been times when the Japanese used strike planes to scout with.  There was no doctrine that called for perfectly symmetrical strike packages, and none that said strike planes could not be used as scout.

I think that the older explanations for Nagumo's and Genda's decisions is still the correct one: Genda and Fuchida's concept of "Victory Disease."  More scouts were not used because they were not needed, because the Americans were not out there yet, because in the plan they only sortied from Pearl Harbor when the attacks on Midway began.  There was no particular urgency in attacking the Americans on Nagumo's part because the Americans couldn't really hurt the Japanese fleet anyway.  The plans of the Japanese had always succeeded, and therefor they always would.

This is an excellent book, though, and I recommend it highly to anyone with an interest in the topic and a modicum of carrier savvy (enough to know what a CAP is and how the Japense designated aircraft by year marks).  The authors assume a bit of such knowledge.

Thoughts?

Seems that some of this is true about the Japanese not feeling the US Carriers were too much a problem. The Yorktown had been severely damaged at Coral Sea or elsewhere prior to Midway (did the Japanese think it had been sunk), but it was still made serviceable for the battle. The Japanese also expected their moves on the Aleutians to draw off some US strength, perhaps maybe even the carriers? So it seems to me that the Japanese expected as much trouble from Miday's air units and felt the US Navy was in tougher condition than it actually was, as I recall, but it's been a while since I've read very heavily on this battle.

As for scouting. My understanding that both sides did use carrier planes for scouting. Apparently at Midway the Japanese felt that wasn't so necessary, as they figured chances weren't so strong for the US Carriers to show up in much force. Also, fog of war I guess. One search plane, I think, had mechanical trouble delaying its launch, and that was the one that spotted the US carriers. Such a simple thing but which helped cause the Japanese commanders to have to make decisions at the worst time when that recon finally came in, when decks were full of armed or re-arming planes to attack Midway again.

[Berkut]

I look at Midway with the wargamers eye, and think that this is a bad fight for the Japanese to find themselves in.

ON the face of it, it isn't so bad - the Japanese tend to win these a lot more than not, and its not like there is any real reason to expect that they will just get hammered as they did.

But from a gaming perspective, it is one of those fights were the range of possible outcomes very much favors the other side. Sure, its not like it is a poor odds fight as far as strictly winning or losing the fight, but the consequences of the outcome of the fight very greatly favor one side over the other.

At one end, a crushing Japanes win hurts the US, but is not at all decisive. Reverse the losses (US loses all three carriers for 1 Japanese carrier lost) and does this change the outcome of the war? Nope.  Delays it certainly, until the US gets the Essex classes rolling out, but doesn't really change anything but timing, I think.

We know what the actual result was of course - a crushing US victory that very much sped up the war, allowing the US to surge forward in the South Pacific without concern about the KB showing up.

Something in the middle? Still favors the US. Any kind of even exchange of carriers would result in the Midway invasion probably failing in the long run, if not in the short, and japanese ability to hold Midway would be pretty tenuous anyway.

Of course, as much as you can look at doctrine to see that the tactical decisions made were hard to not make, you can look at Japanese naval culture and see that the "right" move at that point (hunker down, keep the KB in being as a credible threat, and make the US bleed) was probably just as unlikely. And maybe not even a winning strategy either.

[grumbler]

Seems that some of this is true about the Japanese not feeling the US Carriers were too much a problem. The Yorktown had been severely damaged at Coral Sea or elsewhere prior to Midway (did the Japanese think it had been sunk), but it was still made serviceable for the battle. The Japanese also expected their moves on the Aleutians to draw off some US strength, perhaps maybe even the carriers? So it seems to me that the Japanese expected as much trouble from Miday's air units and felt the US Navy was in tougher condition than it actually was, as I recall, but it's been a while since I've read very heavily on this battle.

The Japanese thought Yorktown, Saratoga, and Lexington had all been sunk, but were convinced the US had converted a bunch of merchant ships to carriers.

But it is true that they didn't expect the carriers to be north of Midway before the Japanese even revealed their intentions.

As for scouting. My understanding that both sides did use carrier planes for scouting. Apparently at Midway the Japanese felt that wasn't so necessary, as they figured chances weren't so strong for the US Carriers to show up in much force. Also, fog of war I guess. One search plane, I think, had mechanical trouble delaying its launch, and that was the one that spotted the US carriers. Such a simple thing but which helped cause the Japanese commanders to have to make decisions at the worst time when that recon finally came in, when decks were full of armed or re-arming planes to attack Midway again.

Actually, that scout's story is even more interesting that the myth.  The scout that was late was never supposed to fly near the Americans.  Another scout plane actually overflew the Americans without seeing them.  Tone Scout #4, because it was late, kinda cheated on their patrol area and thus stumbled across the Americans a long way from where Tone #4 was supposed to be (and where it thought it was, for that matter).  That late launch turned out to be good thing for the Japanese.