Saturday, September 26, 2015

Decisive Battles: Philippine Sea 1944

Battle of the Philippine Sea Dates: 19-20 JUN 1944

Note: This is another in the my series discussing Imperial Japan in the Great Pacific War. Up to this point I have largely concentrated on the relative incompetence of the Imperial Japanese Army of WW2 as an industrial armed force. How the Japanese military government came to launch a war that they should have known was hopeless, given the primitive abilities of their industrial base and their army has always intrigued me; it seems like such a fundamentally avoidable error.

The Imperial Navy seems to have been more technically competent than their army counterpart; indeed, until about mid-1943 the IJN was ahead of the USN in some critical tactical and technical elements such as torpedo and naval aircraft design. The IJN's failings seem to have been largely strategic, or mental, or perhaps even emotional. The Japanese admirals seemed to have an odd sort of defeatist outlook. Even when they won, when the USN took the harder pounding and the heavier losses, many of the IJN commanders seem to have hugged a lurking fear that their victories were mirages, their enemies hiding some dreadful power untouched by defeat.

The admirals had logistical and organizational problems, as well. As with the Imperial Army, logisticians were not particularly esteemed in a warrior culture and the IJN suffered from the sorts of logistical problems you'd expect. The 軍令部 (Gunreibu), the IJN General Staff, had issues as well. Fixated on the Mahanian notion of the "decisive battle" that (in the minds of the WW2 generation of senior officers, at least) had been vindicated by Tsushima they desperately wanted the USN to shove its collective head into that meatgrinder. That fixation was one of the driving forces behind what happened in the central Pacific in the early summer of 1944 and its part in Imperial Japan's remorseless slog towards destruction.

I've noted this in the earlier posts; the idea of a Japanese defeat as a "decisive battle" in WW2 is almost an oxymoron. Yamamoto was right from the beginning; Imperial Japan had no business making war on the United States. The empire was doomed from the moment the first anchor rose dripping from the greasy waters of Yokosuka harbor in 1941. If the counterfactual...perhaps; had the USN obliged and sortied into Philippine waters in March of 1942 they would very likely have been sent to the bottom and the war might have gone very differently.

The Battle of the Philippine Sea is of interest to me principally because it does such a good job of highlighting what the USN did right and the IJN did wrong that contributed to Japan's shattering defeat. It is a tale, written in steel and blood, in the sky and on the sea, of two organizations; one the learned and changed and succeeded...another that didn't.

Forces Engaged: United States Navy - The Philippine Sea was another of the Pacific War "over-the-horizon" engagements in which the only human eyes that saw an enemy warship were either in the air or beneath the sea. So regardless of the total numbers of warships engaged the important combatants for the USN were aircraft carriers of Task Force 58 of the Fifth Fleet and the submarines of the Fifth and Seventh Fleets.

TF 58 was the so-called "Fast Carrier Force" of Chester Nimitz's Pacific Fleet. As organized for the Marianas Campaign it consisted of four carrier task groups, TF58.1, TF58.2, TF 58.3, and TF58.4 and the "fast battleship" escorts of TF 58.7.

The carriers in each group are listed below with the number of aircraft embarked: F6F "Hellcat" fighters, SB2C "Helldiver" or SBD "Dauntless" dive bombers and TBF "Avenger" torpedo bombers.

TF 58.1 - 2 fleet carriers (CV12 USS Hornet [41 F6F, 33 SB2C, 18 TBF], CV10 USS Yorktown [46 F6F, 40 SB2C, 17 TBF, 4 SBD]) 2 light carriers (CVL24 USS Belleau Wood [26 F6F, 9 TBF], CVL29 USS Bataan [24 F6F, 9 TBF]), 3 heavy cruisers, 1 light cruiser, 9 destroyers under RADM Clark in Yorktown
Note: The USN in 1944 had three types of aircraft carriers, although the smaller two were somewhat similar to one another.

A "CV" was a "fleet" or full-size carrier. The two TF 58.1 CVs were typical of the Essex-class fleet carriers of 1944; about 870 feet long, 27,000 tons burden, capable of 33 knots and embarking 90 to 100 aircraft. Typically these consisted of a fighter squadron or squadrons (USS Yorktown had one in June 1944, VF-1's "High Hatters"), a dive-bombing squadron (Yorktown's was VB-1) and a torpedo-bombing squadron (VT-1 for Yorktown). Some fleet carries had a scouting detachment (identified as a "VS" unit) that at the time would still have flown the TBD Douglas Dauntless, the fleet dive bomber from 1942.

A "CVL" was a "small aircraft carrier"; an 11,000-ton, 600-foot-long, 31-knot flattop based on a light cruiser hull. Typically these smaller carriers embarked only fighters and torpedo bombers and only about 30 of these. The idea was originally to arm the CVLs with a miniature fleet carrier air wing of 9 aircraft of each type, but experience quickly showed that the fighter-torpecker mix worked better. The CVLs were an expedient, initially ordered because the fleet carriers were slow to build, but they did decent work as sidekicks to the big flattops. The Wiki entry points out that
"...(t)hese were limited-capability ships, whose principal virtue was near-term availability. Their limited size made for seakeeping difficulties in the many typhoons of the Pacific, and their small flight decks led to a relatively high aircraft accident rate. However, being based on a light cruiser, they were fast ships...(with) the speed necessary to operate with the main fleet carrier task groups."
The last carrier type didn't take a direct role in the naval engagements of the Marianas Campaign. A "CVE" was an "escort" carrier, a dinky 500-foot long, 7,800-ton, 21-odd-knot vessel often known as a "jeep" carrier or even worse names; "Kaiser coffin" or "torpedo junction" for their supposed fragility. The jeep carriers embarked about 25 aircraft that were typically divided into F4F fighters and TBF bombers (that were used as bombers - CVE Avengers typically carried either depth charges for antisubmarine work or iron bombs for ground attack).
Anyway, that was the brown-shoe Navy of '44.

TF 58.2 - 2 fleet carriers (CV17 USS Bunker Hill [42 F6F, 33 SB2C, 18 TBF], CV18 USS Wasp [39 F6F, 32 SB2C, 18 TBF]), 2 light carriers (CVL28 USS Cabot [26 F6F, 9 TBF], CVL26 USS Monterey [21 F6F, 8 TBF]), 4 light cruisers, 9 destroyers under RADM Montgomery in Bunker Hill

TF 58.3 2 fleet carriers (CV6 USS Enterprise [31 F6F, 3 F4U, 21 SBD, 14 TBF], CV16 USS Lexington [41 F6F, 34 SBD, 18 TBF]), 2 light carriers (CVL30 USS San Jacinto [24 F6F, 8 TBF], CVL23 USS Princeton [24 F6F, 9 TBF]), 1 heavy cruiser, 4 light cruisers, 13 destroyers under RADM Reeves in Enterprise
Note: the flagship of TF 58.3 is the old lady of the fast carrier force, the 1934 USS Enterprise, one of the victors of Midway and the sole survivor of her Yorktown-class. Note also that the Big E had embarked three F4U Corsairs. This fighter had some fairly significant deck-landing issues and by early 1943 was largely a) assigned to Marine fighter units that were b) operating from land bases. I'm not sure why the Enterprise still had these Corsairs aboard and would be very interested to find out.
TF 58.4 1 fleet carrier (CV9 USS Essex [42 F6F, 36 SB2C, 20 TBF]), 2 light carriers (CVL27 USS Langley [23 F6F, 9 TBF], CVL25 USS Cowpens [23 F6F, 9 TBF]), 4 light cruisers, 13 destroyers

TF 17 (5th Fleet submarine pickets) 20 fleet submarines, including 7 each of the Balao and Gato classes, two Sargo class subs, and one sub each from the Tambor, Porpoise, and Salmon classes.

7th Fleet submarines: 9 fleet submarines

It's worth noting that the 5th Fleet had an entire task group that included seven battleships (TF 58.7) including USS Iowa and USS New Jersey, the newest of the USN's big-gun capital ships and what less than five years earlier would have been the pride of the battleline. Their work for the next two days would be that of huge destroyers, guarding the carriers, their 16-inch cannons silent, their only prey the fleeting aircraft.

So 4 fleet carriers, 7 light carriers, embarking a total of approximately 470 F6F fighters, 190 SB2C divebombers, 180 TBF torpedo bombers, and about 60 SBD scout bombers with their escorting cannon-armed warshipe under the command of VADM Mitscher embarked in USS Lexington. In overall command of 5th Fleet was ADM Spruance in the heavy cruiser USS Indianapolis sailing with TF 58.3.
Imperial Japan - The 1st Mobile Fleet 第一機動艦隊 (Dai-Ichi Kidō Kantai) was also divided into task groups, in this case three main carrier squadrons. In June 1944 two of these (A and B Force) operated more-or-less together, while the third - C Force - worked as a forward or screening unit.

The Japanese naval air arm's fast carrier force, the 機動部隊 (Kido Butai) had been hammered hard at Midway and in the battles for the Solomons. The surviving carriers had been reorganized into the 1st Mobile Fleet and had begun to field some newer aircraft by 1944.
Although the Zero fighter (A6M and variants) still hung on the D3A "Val" dive bomber was being phased out in favor of the D4Y "Judy" and the B5N "Kate" torpedo bomber replaced by the B6N "Jill".

As we'll see, however, the problems that affected Japanese manufacture and design made this process slow and incomplete, and many air groups still flew the older aircraft. We'll also see that obsolete aircraft were often the least of the IJN's air arm problem...

C Force, 3rd Carrier Squadron - 3 light carriers (CVL HIJMS Chitose [21 A6M5b, 3 B6N, 6 B5N], CVL HIJMS Chiyoda [21 A6M5b, 3 B6N, 6 B5N], CVL HIJMS Zuiho [21 A6M5b, 3 B6N, 6 B5N]), 4 battleships, 8 heavy cruisers, 1 light cruiser, 14 destroyers under VADM Kurita in the heavy cruiser Atago

A Force, 1st Carrier Squadron - 3 fleet carriers (CV HIJMS Taiho [26 A6M5, 23 D4Y, 17 B6N, 2 D3A], CV HIJMS Shokaku [26 A6M5, 24 D4Y, 17 B6N, 2 D3A], CV HIJMS Zuikaku [27 A6M5, 23 D4Y, 17 B6N, 3 D3A]), 2 heavy cruisers, 1 light cruiser, 7 destroyers under VADM Ozawa in Taiho

B Force, 2nd Carrier Squadron - 2 fleet carriers (CV HIJMS Junyo [27 A6M5, 6 B6N, 9 D4Y, 9 D3A], CV HIJMS Hiyo [26 A6M5, 6 B6N, 18 D3A]), 1 light carrier (CVL HIJMS Ryuho [27 A6M5, 6 B6N]), 1 battleship, 1 heavy cruiser, 7 destroyers under RADM Joshima in Hiyo

24 mixed fleet (I-class) and small (RO-class) submarines (more about which later) the real pain in the ass...

Base Air Force, Marianas (1st Air Fleet; 第一航空艦隊 Daiichi Kōkū Kantai) - Originally supposed to be more than approximately 1,000 aircraft distributed across 17 airfields on 12 land-base locations, but in fact heavily attrited and close to ineffective as a force by 18 JUN 44.

In assessing the actual condition of the land-based naval aviation in 19 JUN 44 the big problem is that the number and type of aircraft at these land bases is not included in most general references for this engagement and what we do have is often confusing, contradictory, flat out wrong, or a mix of all three.

The most commonly used figures are contained in Morison (1947) and he just throws out numbers and locations without telling us what type of aircraft or the units assigned; 35 aircraft at Saipan, 67 on Tinian, 70 at Guam, and so on.

This is problematic.

For example, what were the aircraft on Saipan; fighters? Bombers? Both? Were they a single unit with some experience flying and fighting together, or a hotch-potch of odds and sods, a Zero here, a Judy there, a bunch of random guys who'd barely had tea in the mess once?

The problem with Morison's numbers isn't just the lack of information in the raw numbers alone. His relative numbers seem odd, too. He suggests that there were only 67 kites under the 1st Air Fleet flagpole on Tinian but 134 on Palau (presumably Pelilieu), a smaller post? Why 50 in Iwo Jima, far northwest of the fighting? Does the 35 aircraft on Saipan reflect that the airfields there had been hit harder than Palau, or that they had left Saipan to go elsewhere? And why; were they fleeing the USN or (as one of the commentors at Nihon Kaigun's aviation page tells us) because of an IJN tactical shift?

The pieces are all over the place, and I had to dig like hell to find anything. I found several sources; one that looks superficially promising is The campaigns of the Pacific war, a study published in 1946 by the United States Strategic Bombing Survey (USSBS, 1946). It's available online here and has a list of the unit postings, assigned aircraft number and types for the 1st Air Fleet in 1 JUN 44.

But the neat organization from this source looks more misleading the harder you look at it.

As we'll see, the USSBS (1946) order of battle appears to be the plan. The field intel suggests that the actual strengths of these units were far less and the actual locations are often far different from that plan and that makes it much more difficult for us to figure out who in the IJN Air Service (大日本帝國海軍航空隊, Dai-Nippon Teikoku Kaigun Kōkū-tai) land-based units was attacking from where with what on 19 JUN...but is a pretty good reflection of the degree to which those plans had gone sideways.

It's really unfortunate that this information is so difficult to obtain because it is fairly important to understanding this engagement. The original Plan A-Go relied heavily on land-based air attacks to supplement (and replace) the carrier airpower lost in the Japanese defeats of 1942 and 1943.
For what it's worth, here's some idea of what the IJN Naval air arm wanted to have in place in June 1944...with some comments about what we think really was there.

I've listed the units grouped by location then the USSBS (1946) roster of unit and aircraft type.
Note: The primary unit designation here is an "Air Group" (航空隊, or Kōkūtai). For the IJN this was roughly the equivalent of a USN carrier air wing, typically two or more squadrons (Hikōtai) that could be similar or identical type aircraft; all fighters, for example, or all A6M Rei-sen fighters ("Zekes" or "Zeros" as the Allies called them) or (when embarked) all the aircraft in a carrier task group, so several fighter, dive or scout bomber, and torpedo squadrons.

The Kōkūtai was, like its Allied equivalent, the primary administrative and tactical organization of the Navy air arm and at least in the Marianas in 1944 most of the Kōkūtai appear to have been single-type or, at most, two-aircraft-type organizations.
Chichi Jima (northwest of the engagement area) - a small seaplane force, probably no more than 8-10 "Rufe" seaplanes? This area was not included in the 1st Air Fleet AO so I don't really have a good sense of what the IJN might have had there.

Iwo Jima (northwest of the engagement area) - Morison (1947) reports 50 aircraft posted here, also outside the 1st Air Fleet AO. This may represent something he got from CAPT Ohmae as a locally based unit there, or it may be the remnants of an outfit he identifies as the "Hachiman Group" or "Hachiman Unit" that is supposed to have been formed from the naval aviation base at Yokosuka, attempted to shift over to Iwo, and was badly shot up in the process.

Saipan (in the Marianas chain) - originally approximately 80 A6M fighters from 261th Air Group (航空隊, Kōkūtai) A detachment of this unit is also reported from Guam, however, where a USN radio intercept dated 1 JUN 44 notes that the Guam detachment has a total of 12 A6M fighters, of which 11 were operational and 12 pilots - 6 with over 1,000 hours, 5 with more than 400, and a cherry with less than 10 weeks of stick time.
Tinian (in the Marianas chain) - originally approximately 125 aircraft, divided between five air groups over two airfields: 121st (10 x probably D4Y "Judy" scout bombers or possibly C6N "Myrt" reconnaissance aircraft),
(Note: the USSBS (1946) lists these aircraft as only as "Suisei (reconnaissance)" without giving the alphanumeric. Suisei is presumably a direct translation of 彗星, or Comet. This was the name the Japanese used for the D4Y that the USN called a "Judy" carrier dive bomber. Since this aircraft was also used for reconnaissance, and given the relative scarcity of the C6N, I suspect these were "Judys")
321st (15 x J1N "Irving" heavy fighter), 523rd (40 x D4Y dive bombers), 761st (40 x G4M "Betty" medium bombers), and the 1021st (20 x L2D "Tabby" transport aircraft).

Guam (in the Marianas chain) - originally approximately 215 aircraft from four air groups over two airfields: 263rd (80 x A6M "Zeke" fighters), 321st (15 x J1N "Irving" heavy fighters), 521st (80 x P1Y "Frances" medium bombers), and 755th (40 x G4M "Betty" medium bombers).

This is very clearly a massive overstatement and, as the radio intercept information shows, these units and aircraft were widely scattered by 18 JUN 44.

Truk (in the Caroline chain) - originally approximately 280 aircraft from seven air groups over three airfields: 151st (20 x D4Y "Judy" scout bombers), 202nd (40 x A6M "Zeke" fighters), 251st (20 x J1N "Irving" heavy fighters), 253rd (80 x A6M "Zeke" fighters), 503rd (40 x D4Y "Judy" dive bombers), 551st (40 x B6N "Jill" torpedo bombers), and 755th (40 x G4M "Betty" medium bombers).

Keep in mind that this is the assigned strength for the 22nd air Flotilla (航空戦隊 or Kōkū Sentai); the USN air attacks in February destroyed some 300-odd aircraft, many of which would not have been replaced by June, and many of the survivors were shifted about the engagement area.

Yap (in the Caroline chain) - approximately 40 A6M "Zeke" fighters from 265th Kōkūtai

However, a radio intercept dated 5 JUN reports the following units and aircraft at Yap: 202th from Truk (4 x A6M fighters), 261th from Saipan (1 x A6M), 321st from Guam (5 x J1N "Irving" heavy fighters), 503rd from Truk (6 x D4Y "Judy" scout bombers), 521st from Guam (2 x P1Y "Frances" light bombers), 523rd from Tinian (1 x D4Y scout bomber), 1021st from Tinian (3 x L2D "Tabby" transports).

Pelieliu (in the Palau chain southwest of the engagement area) - originally approximately 90 aircraft in three air groups: 121st (10 x D4Y "Judy" scout bombers), 265th (40 x A6M "Zeke" fighters), and 761st (60 x G4M "Betty" medium bombers)

Here, again, the IJN's intercepted radio transmissions tell the story of the chaos that reigned in the 1st Air Fleet in the summer of 1944. On 18 JUN a group of aircraft arrived on Pelilieu: 9 x G4M "Betty" bombers from 732nd Kōkūtai supposedly based on Halmhera, 4 more G4Ms assigned to the 753rd Kōkūtai from Manado, and 10 A6M fighters assigned to the 153rd Kōkūtai out of Sorong
Mereyon (reported to have been an "island fortress" in the Marshall chain) - originally approximately 40 x D4Y "Judy" dive bombers of the 202nd Kōkūtai. Given the pounding the Marshalls had already taken by June it is difficult to believe that many, or any, of these aircraft remained serviceable.

Halmhera (in the Indonesian archipelago) - reported to be approximately 40 x G4m "Betty" medium bombers of 732nd Kōkūtai at Wasile airfield. An intercept from 10 JUN 44 reports the follwing aircraft at Wasile; 17 x D4Y "Judy dive bombers of 523th Kōkūtai (of which 12 were operational and all of which were supposed to be on Tinian...), 18 x P1Y "Frances" light bombers of 521st Kōkūtai (supposedly posted to Guam), a total of 59 A6M fighters, including 10 that were non-mission-capable 25 from 261 Kōkūtai (assigned to Saipan), and 34 from 265 Kōkūtai (officially supposed to be on Yap).

Menado (or probably Manado, in the Celebes) - assigned 40 x G4M "Betty" medium bombers of 753rd Kōkūtai.

Davao (in the Philippine archipelago) - the problem at this location is that Morison (1947) only reports 25 aircraft based here . But the IJN records suggest two and possibly four Kōkūtai if the units reported at Lasong and Sorong are also Philippine air fields and his identification of the 201st Kōkūtai as at Cebu is wrong. The list of assigned Philippine-based aviation units includes the 153rd (20 D4Y "Judy" scout bombers and 40 x A6M "Zeke" fighters" at Sorong), 201st (20 x A6M "Zeke" fighters), 501st (40 x A6M fighter-bombers at Lasong), and the 751st (40 x G4M "Betty" medium bombers). Again, much higher numbers than Morison (1947) reports and very likely much attrited by the time of the naval engagement.

Adding up all the "assigned" strengths you get something like 1,240 aircraft; this is not supported by any of the published sources as close to the existing strength of the 1st Air Fleet on 19 JUN. The raids of late winter and spring had destroyed a very large number of the naval aviation strength at the Marshalls, Carolines, and Marianas bases.

There is some useful discussion here about the displacement of many of these aircraft to southwestern locations to support the defense of Biak. Commentor Rick Dunn remarks (about the aircraft list in the USSBS, 1946):
"The aircraft strengths are totally inaccurate and the bases are misleading. On June 1st much of the strength of the 1st Air Fleet was deployed to or en route to Wasile and other bases away from the Marianas in order to support the Japanese response to the invasion of Biak (May 27th). Just as the KON operation diverted part of the Combined Fleet so also was the Central Pacific's base air force diverted to the south. This was a major strategic blunder that went a long way toward making the base air force ineffective during the Marianas operations."
Later the same commentor includes the information from radio intercepts noted in the airbase paragraphs above.

So we can't be sure of the location, number and type of the land-based naval air units other than that they were considerably smaller, weaker, and more battered that they were supposed to be - or, more importantly, thought to be by VADM Ozawa - by 18 JUN 1944. My best guess is that the 1st Air Fleet probably had about 200-250 aircraft of all types spread out across the Marianas as well as all the way from the Philippines to the Volcano Islands, all at least notionally under VADM Kakuta on Tinian.

Altogether 5 fleet carriers, 4 light carriers and escorts embarking 220 A6M fighters, 97 D4Y divebombers, 16 D3A divebombers, 81 B6N torpedo bombers, and 17 B5N torpedo bombers supported by about 200-250 land-based aircraft of various types under the overall command of VADM Ozawa.

The Sources: The events of June 1944 are well-documented as is nearly always the case when two industrial societies fight. As always for the Great Pacific War, however, the caveat is that for an English-only speaker one side is much less accessible than the other.

The United States National Archives notes that:
"The major collection of Japanese World War II records accessible to those who do not read Japanese is in the National Archives in Washington. There are several series of intercepted Japanese diplomatic, military attache‚ naval attache‚ army and navy...with some of them running to tens and even hundreds of thousands, which are included in the Record Group 457 of the National Security Agency, the inheritor of the American wartime decoding records. These materials are of immense importance because they reflect not only affairs internal to the Japanese diplomatic and military services but because they report on the countries where Japanese diplomats were stationed. Furthermore, for the Japanese as for the German records, the destruction of war has in many cases left the translated intercepts the only surviving copies of documents of which no German or Japanese originals exist."
You'll note the important statement there - "for those who do not read Japanese".

For those who do there is a Japanese National Archives, that is reported to include a relatively full record of the Showa Period, the reign of the Emperor Hirohito that lasted until 1989.
This includes both Imperial Army and Navy records. Although some destruction took place (both by American bombing and Japanese incineration) in 1945 the Japanese war records are said to include
"Official records of the...Navy General Staff Office...(t)he military organs’ important records, which are crucial for understanding the Showa history, had been hidden away for years from the eyes of the public. Senshi Sosho (War History Series), which totals 102 volumes, was compiled by the then Military History Office (now Department) of the National Institute for Defense Studies, on the basis of records of the Imperial Army and Navy which survived in Japan, as well as ones seized and later returned by the GHQ. Most of the documents used for Senshi Sosho are now kept at the Military Archives of the NIDS (National Institue for Defense Studies)."
The problems inherent in this otherwise magnificent source are many. The materials do not circulate, so those unable to make the five-minute walk from Exit 5 of Ebisu Station by Tokyo Metro Hibiya Subway Line to the NIDS at 2-2-1 Nakameguro, Meguro-ku, Tokyo cannot access the records. And, unfortunately for the non-Japanese speaker many, perhaps most, of these records have still not been translated.

So for the casual researcher the Japanese sources are incomprehensible and the information that has migrated into English language sources is not always completely reliable.

Perhaps the biggest single open question is, as noted above, the condition and number of the land-based naval aviation units engaged in the battle. The commonly cited list of airbases and aircraft and the source for the order of battle listed above is from S.E. Morison's monumental History of United States Naval Operations in World War 2 (Volume 8: New Guinea and the Marianas, March 1944 – August 1944).

Morison's footnote on page 219 of that volume reads "Capt. Ohmae's list furnished 3 November 1951, compiled 'from every available document and interrogation of the officers concerned'"

This gentleman was CAPT Toshikazu Ohmae who was extensively questioned after the war (HyperWar has several of his interrogations here). CAPT Ohmae is described in his interrogation as Chief of Operations for the IJN's Third Fleet in June of 1944 and as VADM Ozawa's G-3 would, or should, have known the dispositions of the 1st Air Fleet, so we may assume that his information was not a complete fabrication, but, still...we're dealing with a recollection of a relatively minor statistic more than a year after the event. No matter how good a staff officer Ohmae may have been it's hard to believe he was that good.

So far as I know there has never been a Japanese-language account of the engagement translated into English, either first-hand or as a work of secondary-source history.

Other than Morison - whose history is more for the general reader than the true military historian - the other significant secondary English-language source is William Y'blood's 1981 Red Sun Setting, described as a very good source as well as a very readable account.

Internet resources appear to be taken largely from either Y'blood or Morison. The Wiki entry appears fairly well researched and written, though it appears to rely heavily on two other published sources: E.B Potters' 1990 Admiral Arliegh Burke and H.P. Wilmott's 1984 June, 1944.

The salty blogger-samas at Nihon Kaigun - "Combined Fleet" - have a relatively brief account but one that is profusely illustrated and well worth the visit.

The World War II Database has a page covering the Marianas campaign but the actual section on the naval engagement we're discussing is fairly sparse, the bulk of the information being from the ground actions on Saipan, Tinian, and Guam.

The NavWeaps site has a relatively complete order of battle, though with the same problem with the IJN's land-based air assets I have.

The U.S. Army's official history of the Marianas Campaign is available in PDF form; obviously the focus is on the island defenses, but there's a useful discussion of Japanese preparations for the defense of the Marianas in Chapter IV and a brief summary of the naval engagement as well.
The Campaign: To discuss the place of the Battle of the Philippine Sea in the overall course of the Pacific War we have to back up a bit, to late 1943 and early 1944.

Actually, we could go as far back as early 1943, as we did back in December of 2013 when we talked about Bataan or even earlier, when we talked about the strange little action on the Clatsop Plain near Astoria, Oregon in 2012.

In both we discussed the original objectives and factors that led, or drove, Imperial Japan to war with the United States. But once the war was on, the Empire really had only one strategic plan; to defend as fiercely as possible and hope that the Yankees would get sick of slaughter.

Well, that wasn't the entirety of the plan, but that was the gist. The idea was to push the Imperial perimeter as far out as possible and then make the Allies pay in blood for every kilometer, sea mile, fortification and island they took. It wasn't pretty, but given Japan's economic and logistical limitations it was really the best the Imperial General Headquarters (大本営 or Daihon'ei) could come up with.

The Imperial Navy, though...they had another idea, an old, bad (thought they didn't recognize it as bad) idea that had been kicking around the IJN for forty-odd years.

This was the so-called "Decisive Battle Doctrine" (艦隊決戦, or Kantai Kessen).

The idea was that the naval part of the Pacific War would be decided - in Japan's favor, of course - by one great decisive battle. The IJN had been designed around this monster gunbattle because of a book and a battle.

The book was Mahan's The Influence of Seapower Upon History and the Japanese were just one of the many groups of squids who got all misty-eyed and semi-erect reading Mahan.

His was a seductive tale of naval power, a sort of Fifty Shades of Battleship Gray naval strategy porn for the wannabe admiral, a stirring ode to battleships and their command of the sea that would lead to total victory.
This sort of military arm-waving wasn't unusual for the late 19th and early- to mid-20th Century, by the way. Some joker named Douhet scribbled a similar paean to wing-wiping called The Command of the Air, and a Russian named Seversky cranked out something similar called Victory Through Air Power that was made into a Walt Friggin' Disney cartoon, fercryinoutloud. Compared to the zoomies the sailors were so restrained as to be positively Stoic...
The battle was the one-sided 1905 slaughter of the Imperial Russian Navy at the Tsushima Strait that appeared to have written the epilogue to Mahan in blood and steel.
The problem with this is that the lesson was wrong. Completely, utterly wrong. Tsushima was not the Mahanian decisive victory of one better-prepared, better-commanded battlefleet over another but a back-alley beatdown of a ramshackle shitshow of a floating goat rodeo by a Japanese fleet that was at least capable of pointing its cannon in the right direction.

Most "decisive battles" at sea - Trafalgar in 1805, Manila Bay and Santiago in 1898, Coronel and the Falkland Islands in 1914 - all had one thing in common; one side was completely, utterly fucking hopeless. Hopeless physically, numerically, technically, tactically, morally...or some combination of all of them. The winning side would practically have to have turned their cannon on themselves to have lost.

So the IJN had based their overall strategy on a couple of total goofs, and not surprisingly, that didn't go well.

Of course, the IJN had other issues. COMSEC, for one; most of us know about the breaking of the Imperial diplomatic PURPLE cypher but the Imperial Navy's codes, particularly the primary command and control cypher the Allies termed JN-25, were being read as early as 1942 and the Japanese both refused to believe that or significantly change their codes.
Logistics, always Japan's weak point, was particularly difficult for a Navy dependent on fuel oil. The Indonesian oilfields produced a relatively light "sweet" crude that could be used for bunker fuel at a very rough level of refining...but impurities in the fuel were hard on the boilers.

The Navy also used a tremendous amount of fuel (as always, the guys as Nihon Kaigun are great on this, providing a good discussion of the fuel problem as it affected the Solomons Campaign here...) and given the small number of tanker marus and the aggressive US submarine war against them the Mobile Fleet was largely tied to its bases and those bases both dispersed widely and perforce pushed southwest - closer to the fuel source but further from the outer defenses.

And those defenses had to be reconsidered. The U.S. Army's official history discusses the revisions in Japanese strategic planning that had been forced by the losses of 1942 and 1943;
"The defeat at Midway in June 1942 followed by the loss of Guadalcanal and Papua early in 1943 had been serious but not fatal blows. More serious had been the loss of ships and pilots, and these, it was hoped, would ultimately be replaced. But MacArthur's and Halsey's victories in the Solomons and New Guinea during the summer of 1943 cast a more serious light on the situation. Obviously the Allies were making a determined assault on the Solomons, eastern New Guinea, and the Bismarck Archipelago, which the Japanese called the Southeast Area. Failure to hold the outposts in New Guinea and the Solomons, they recognized, could have disastrous consequences and might well be the prelude to an Allied advance toward Truk and the Philippines." (Chapter XXVII)
The Imperial defenses of the Southeast Area in 1943 and early 1944 looked like this:

The Daihon'ei planned that the 8th Area Army and the Southeast Area Fleet would slow MacArthur's SOWESPAC offensive towards the Philippines by defending north-central New Guinea and the Bismarcks. In the South and Central Pacific the concentration of forces around Truk in the Carolines and around Rabaul and Bougainville in the northern Solomons aided by the Mobile Fleet would slow Nimitz's advance. This would attrit the Yankees and - at least the IJN hoped - would set up the Kantai Kessen somewhere in the central Pacific.

Supposedly the Japanese war councils considered the Marianas too isolated and too far from any large-enough landmass to be militarily vulnerable. They also didn't anticipate the speed with which the US had learned to plan, stage, and launch amphibious attacks, so CENPAC's attacks on Tarawa and Makin in the Gilbert Island chain in late 1943 and Eniwetok in early 1944 unhinged the planned defensive schemes.

Following on the loss of the Marshall and Gilbert line the Fast Carrier Task Force raids hammered Truk and the eastern Carolines as well as the 1st Air Fleet bases in the Marianas and pounded them flat;
"During the operations against Eniwetok, United States carrier task groups struck Truk on 16 and 17 February, Jaluit on the 20th, and the Marianas on the 22d. At Truk the sustained two-day attack by carrier aircraft destroyed 26 merchant vessels, 6 combatant ships, and 270 aircraft, damaged 6 more naval vessels, and inflicted severe damage on fuel and provision storage. It had been planned to strike Ponape and Jaluit next but in view of the success of the Truk operation a carrier strike on the Marianas was substituted. Although detected by reconnaissance aircraft 420 miles east of Saipan, the task force fought its way through a night-long series of attacks (and hit the Marianas bases) the next morning. One hundred twenty aircraft were destroyed. The destruction of the advanced echelon of the First Air Fleet was a severe blow to the air organization upon which Japan was depending heavily for the defense of the Marianas and Western Carolines."
In theory the Japanese should have been able to use "interior lines" to shift forces from the southwest to the central Pacific to counter MacArthur's and Nimitz's moves; in practice both the Allied organizations got "inside the Japanese OODA loop" to put it in modern terms.

Part of this success included outthinking the Daihon'ei on the vulnerability of the Marianas.

The Japanese had counted on bleeding CENPAC through the Carolines, assuming as we've noted that the logistical difficulties precluded a jump to the northwest into the Marianas. And given the methodical way the U.S. had worked up the Solomons and New Guinea that wasn't exactly a stupid assumption.

But the Daihon'ei didn't recognize what a huge strategic advantage the Fast Carrier Force was. The success of the Gilberts/Marshalls operation and the late winter raids were so shocking that even the American commanders missed the implications for half a year; now the U.S. didn't need island bases to rule the skies and, ruling the skies, rule land and sea.
And that, in turn, meant that the U.S. didn't have to grind slowly island-by-island through the Japanese defenses. With USN airpower crushing Japanese air and surface ship movements the island garrisons were reduced to the force projection metric of 1890 - the maximum effective range of their shore batteries. The "island fortresses" were so many military prisons for their garrisons. They couldn't run, they couldn't reach out to fight, all they could do was sit and slowly starve.

So after battering the hell out of Truk, for example, the USN didn't need the Carolines or the Palaus; they could go right through the outer defenses and punch a hole in the inner ring of the Marianas.

This was a knife at the throat of the Home Islands. Aircraft and USN warships sailing from the Marianas could control the air and sea right up to the southeastern coast of Honshu. Worse, the new B-29 bomber could range almost all of Japan from Marianas bases and without the logistic and tactical problems inherent in flying from China bases.

At this point the Imperial Naval General Staff had to make some kind of countermove, and VADM Ozawa's "Operation A-Go" was it.

The idea behind A-Go was your basic Kantai Kessen; the Mobile Fleet would sortie from its ports in the the Philippines and hit the USN as it supported the Marianas landings. The Wiki entry notes that the critical part of the plan was that the Mobile Fleet wouldn't have just carrier airpower but the "unsinkable carriers" of their land bases along with some material and tactical edges:
"In addition, the Japanese aircraft had superior range, which could allow them to engage the American carriers beyond the range of American aircraft. Furthermore, with island bases in the area, the Japanese hoped to launch at distance, have their aircraft attack the U.S. fleet and then land on island airfields. They then could shuttle back and attack again on the return flight. Thus the U.S. fleet would be in the position of receiving punishment without being able to deliver it. Lastly, the area was dominated by the easterly trade winds. Naval aircraft of the era needed a head wind blowing across the flight deck to enable the aircraft to launch. The easterly trade winds that dominated the Central Pacific seas meant that aircraft carriers would necessarily have to be steaming eastward to launch and recover aircraft. This meant that a fleet located to the west of the Marianas would be in position to initiate and break off the battle, placing the initiative in the hands of the Japanese."
But this meant that the success of the plan largely depended on;

1. Fending off the USN airpower from the airbases; unable to run and hide they had to be defended and their runways and aircraft ready to attack,
2. Finding the US carriers first, and
3. Seeing the attacks through; not just striking but sinking American carriers to clear the way to attack the amphibious forces.

The opportunity came sooner than expected; on 12 JUN 44 USN aircraft appeared over the main Marianas bastions of Tinian, Saipan, and Guam and began striking the airbases and what fortifications, weapons systems, and troops they could catch.

Over the next three days the air and naval bombardment pounded Saipan and on 15 JUN the first Marine infantry from the 2nd and 4th divisions crossed the strandline. The U.S. Army's 27th Division landed the following day.

The warning order to the Mobile Fleet to execute Operation A-Go had been issued in early May. Now the execute order went out on 13 JUN. The main body of the Mobile Fleet sortied from Tawi Tawi that day heading north through the Philippine archipelago, while several more heavy cruisers and destroyers were pulled from the actions around Biak and headed northeast around the east side of the islands.

These two main portions of the fleet met east of Leyte on 16 JUN and refueled over the night and into the following day in preparation for what the commanders and their crews hoped, and many, probably believed, would be the victorious decisive battle.

The Engagement: You probably know what happens next; I mean, this thing is called the Great Marianas Turkey Shoot, right?
So what I want to do from here on is to use the events of the next three days to point out why the IJN - for all that it seemed to be more grounded in reality than the Imperial Army - had by 1944 put itself in a position where it could only lose and should have known that. The rational option at that point should have been surrender. A hell of a lot of people on both sides had yet to die in June of 1944, and an Imperial General Staff that recognized how utterly fucked they were might have helped a lot of them, well...not.

But they didn't, and one of the reasons is that the leaders of the IJN - hell, the leaders of Japan - were as blind to their hopelessness as they were to their technical and tactical and logistical failings.

The story of the battle is really pretty simple, and goes like this:

17 JUN 1944 - The IJN attack forces - the main body of the Mobile Fleet from Tawi Tawi and the southern force pulled from the fighting around Biak rendezvous in the western Philippine Sea off Samar. USN Task Force 58 is still close to the fighting on Saipan, although two task groups have been diverted northwest of the island to intercept air attacks from the Volcano and Home islands.

18 JUN 1944 - IJN forces continue north and east, with search aircraft ranging out front to locate the American carriers. At this point the USN deployment looks like this: TF58.7 (the fast battleship escort unit) was spread out in front - west- of the carriers as an antiaircraft screen. TF58.4 was close and north, since it was the smallest and weakest of the carrier units. East of the the three main carrier groups were in a rough line, with TF58.1 north, 58.2 in the center, and 58.3 to the south.

Around midnight on the 18/19 JUN the USN intercepted an IJN transmission it located about 350 miles west-southwest of the TF58 location. This presented two problems; one, that if the two forces continued on convergent courses they chanced a night surface engagement (that the IJN had long been great enthusiasts of) and, two, that the IJN had a long tradition of using decoy forces, and especially sending out single vessels to transmit to spoof radio direction-finding into fooling enemies into mislocating their forces.

Mitscher wanted to move west, risking the night fight, so as to be in position to strike the Japanese carriers at dawn. Spruance, concerned about uncovering the amphibious force, refused permission, and held TF58 eastward during the night.
19 JUN 1944 - Both sides searched for each other. The IJN had the weather gauge - meaning that they didn't have to reverse course to get the wind over the carriers' bows to launch aircraft - and the longer-range aircraft, so an A6M fighter out of Guam spotted the USN first at about 0550hrs and the Japanese launched two strikes from both the Mobile Fleet's carriers and with what land-based aircraft they could.

IJN Raid 1: The first raid was a land-based strike and was spotted on radar about 1000 hours. Fighters of the US carriers were vectored to intercept, and some USN aircraft were sent to attack the airstrips on Guam.

The interception and retaliatory strikes were crucial, and we'll talk about that in depth in just a bit.

IJN Raid 2: Also at about 1000hrs more Japanese aircraft showed up on US radar to the west - this was the first IJN carrier strike. US aircraft over Guam were recalled and, along with the fighters orbiting over the US carriers - the "Combat Air Patrol" or CAP, intercepted this raid at about 70 miles out at 1030hrs.

Of the 68 aircraft in the IJN strike force 41 were shot down before sighting an American warship and the attack formation was badly scattered. Small groups and individual aircraft did manage to reach the advanced US gunline group and attacked TF58.7. A bomb damaged USS South Dakota and killed or injured 50 of her crew, the only significant damage or losses suffered by a US ship that day.

Most of the attackers were shot down. The US antiaircraft batteries had recently begun using a new fuze, the variable time or "VT", that used radar emissions to burst when near a target. Combined with radar fire control and the CIC (which we'll discuss in a bit) the Japanese aircraft were pretty roughly handled.

IJN Raid 3a: At about 1100hrs US radar picked up the largest strike of the day thus far, about 100 carrier aircraft. These were intercepted 60 miles west of TF58's positions and at least 70 were destroyed over the horizon from the US fleet. This time small groups of survivors got through to the US carrier groups.
The Wiki entry notes that:
"Six attacked Rear Admiral Montgomery’s group (TF58.2), nearly hitting two of the carriers and causing casualties on each. Four of the six were shot down. A small group of torpedo aircraft attacked Enterprise, one torpedo exploding in the wake of the ship. Three other torpedo-aircraft attacked the light carrier Princeton but were shot down. In all, 97 of the 107 attacking aircraft were destroyed."
IJN Raid 3b: Concurrent with Raid 3a - 47 aircraft spotted inbound from the north, intercepted 50 miles out and broken up; seven A/C shot down but the majority don't press through and return to the IJN carriers.

IJN Raid 4: - carrier strike launched between 1100 and 1130hrs but given bad target location. The flight leaders searched in two loose gaggles for some time before abandoning the mission and heading for Guam and Rota for fuel. En route one of the groups encounters TF58.2 and attacks; half the Zero fighter escorts are shot down by the CAP but a group of 9 D4Y "Jill" divebombers attack USS Wasp and USS Bunker Hill; no hits and 8 of the 9 are shot down.

The larger group is intercepted by USN F6F fighters over Orote Field on Guam; of the 49 IJN aircraft all are either shot down or damaged so badly on landing that they are unflyable.
The aircraft engagements of the first day are entirely defensive for the USN (aside from the strikes on the land bases on Guam) but are also entirely one-sided; about 350 IJN aircraft are destroyed and the majority of their pilots and crews killed. The USN loses roughly 30 aircraft but probably less than half of the aircrews. No US warships are more than slightly damaged.

What Happened and why? IJN aircraft versus USN aircraft and surface forces

Contrast the events of June 1944 with June 1942, the Battle of Midway. In the engagement two years earlier the two opposing air groups had pretty much traded blow for blow. The USN TBD "Devastator" torpedo bombers were annihilated but the SBD "Dauntless" dive-bombers hammered the Kito Budai, sinking four fleet carriers.

In return IJN D3A "Val" dive bombers and B5N "Kate" torpedo bombers put three bombs and two torpedoes into USS Yorktown, crippling her. Both air groups were pretty badly torn up, although the loss of all her flight decks meant that the IJN effectively lost all her carrier aircraft, as well. The radio intercepts had enabled the USN to achieve strategic surprise, VADM Nagumo's mistake to strike Midway twice helped sink his own carriers...but the USN still came away knowing they'd been in a scrap.

By 1944 the Teikoku Kaigun Kōkū-tai was flat-out butchered, the USN - both on the sea and in the sky - barely nicked.

What the hell had happened?

First, technically the Japanese carrier warplane design and production teams had fallen behind.

Let's look just at the fighters. In 1942 the A6M Zero was a decent carrier fighter; long-ranged, maneuverable, and reliable if lightly armored.

Its opponent, the F4F Wildcat, was sturdy but slower, relatively lightly armed, and with a fairly short range.

By 1944 the IJN fighter pilots were still flying the Zero but their American opponents now had the F6F Hellcat.

Heavier, better armed and armored, faster, and longer-ranged the F6F was a significantly better aircraft than the F4F. It was still less agile than the Zero but much more robust; the A6M paid for its agility by lacking armor and self-sealing fuel tanks, making it both flammable and fragile. And the USN had learned not to try and dogfight with the Zero, at least the Zero with a good pilot. US tactics negated most of the low-speed turning advantage that the Zero still maintained when flown by an experienced pilot.
And that was the second problem.

Many of the Zeros flying east over the Philippine Sea were flown by young men with very little experience.

When the war started the situation was completely reversed; in 1941 the Teikoku Kaigun Kōkū-tai required 700 hours in the air to qualify a pilot for fleet service compared to 305 hours for a USN aviator. At the beginning of the war about three-quarters of the USN pilots had fewer hours than the IJN's most cherry fliers.

This was because the IJN flight school was insane. It was a four to five year program that only produced about 100 deck-landing qualified guys a year. Proposals to shorten and lighten up the training were rejected; the idea was to win a short war with outstanding quality, not slog through a long one with average crap.

Although the Japanese program was shortened - dramatically, by 1945 - it never managed to keep pace with the appalling losses of 1942 and 1943. By 1944 the most of the pilots of the 1st Mobile Fleet had no more than a couple of hundred hours in the air, if that. The 3rd Carrier Squadron pilots had less than 100 hours of flight time. These poor meatsticks were so many targets.

Morison (1947) has a couple of quotes from US naval aviators that give you an idea of what happened when they met these poor sad Japanese bastards. One of the US fighter interceptors met the first, land-based, raid:
"Brewer reported after his return that the enemy planes seemed to have no formulated defense tactics. The bombers mostly scattered, rendering themselves vulnerable, and the fighters did not appear to cover them but went into individual quick, sharp acrobatic maneuvers…to escape the Hellcats. (p.267)"
The carrier aircrews encountered later in the day weren't much better:
“It was the opinion of all the fighter pilots with whom I talked,” wrote Ensign Charles D. Farmer USNR of Enterprise, “that the Japanese attack was not a good one. Many of their fighter pilots were content to stay out of the action entirely. The dive bombers and torpedo planes were loath to keep their defensive formations, and separated, thus losing all chance of a coordinated attack.” (p.285)
Hard to say which was worse; the Japanese pilots' technical flying skills or the Japanese aviation units' tactical training level. But whatever the problems, the combination was a deadly one for the Japanese.

The USN had two other huge advantages that made the aerial combat of the first day so one-sided; radar, and command and control.
The radar problem was one that the Japanese never really solved. The Pacific War Online website says of this that:
"Japanese inferiority in radar technology was the result of Japan's lack of depth in its technical base and of neglect by the military and naval leadership.

One visiting German professor noted in 1938 that "Japanese universities resembled senior high and vocational schools, because what was to be studied in the university was established in advance. Under such circumstances there was little freedom within the university, nor much freedom for academic instruction" (Nakagawa 1993).

The Japanese Navy was aware of the potential of radar...(h)owever, lack of interest and support meant that Japan quickly lost its lead in this (cavity magnetron) crucial technology. (T)he Navy's Electrical Research Department, which was responsible for radar, had grown to just 300 staff by August 1943.

By the end of the war, quality control on Japanese electronics was so poor that often only one tube in 100 actually worked, and even those that passed inspection had a mean time to failure of as little as 100 hours. For a system with 40 tubes, this meant a mean time to failure of just two or three hours."
The US technological advantage kept the USN radars well ahead of the IJN's throughout the war. The USN developed search radars of the sort that identified the Japanese airstrikes over 100 miles out and control radars - the ancestors of today's AWACS - capable of vectoring interceptors in on the attackers tens of miles out.
That vectoring was done in the Combat Information Center. The CIC was the USN's recognition that modern electronic warfare was too complex for the individual ship's captains and air group commanders to manage. Instead, the CIC used a compartmented staff to process the raw radar (and/or sonar) data, develop an assessment of the tactical situation, and then communicate up to the operational commanders and down to the maneuvering aircraft.

Morison (1947) relays the following anecdote:
"A unique role in this great air battle was played by the Japanese air coordinator. Mitscher’s flag communicators were on his circuit early and Lieutenant (JG) Charles A. Sims USNR stood by to translate his order to the Japanese aircraft… Someone proposed to vector out the air patrol to shoot him down, but wiser heads prevailed; and this Japanese officer continued throughout the action to furnish Commander Task Force 58 with specific information of when and where strikes were coming in. After the slaughter was over “Coordinator Joe” signed off and started for home. “Shall we get him now?” asked and eager officer. “No indeed!” said Admiral Mitscher. “He did us too much good!” So little Joe flew home, proud of having done his duty to his Emperor, and followed by the blessings of the American fighter-directors (p.274)"
The CIC was to "Coordinator Joe" what a modern AWACs is to the CIC of WW2; the full next generation of what we call today "command. control, communication, and information" or C3I.

So the Japanese raids on 19 JUN 44 failed because the USN had learned and changed and the IJN had not. The USN had improved its aircraft and radars, had trained thousands of aircrews and technicians, had developed new techniques of defending their carriers and striking out at their enemies, and the Japanese had not.

Meanwhile, the Mobile Fleet was running in to trouble in another place that the IJN hadn't kept up; underneath the sea surface.

What happened and why? USN submarines versus IJN surface and air forces/IJN submarines versus USN surface and air forces

The USN's submarine fleet had been aggressively patrolling and harassing the Japanese merchant navy and warships from the outer edge of the Co-Prosperity Sphere all the way to the Malacca Strait since 1943.

A sensible Japanese admiral would have had aggressive anti-submarine patrols in both the air and on the surface around his force, but IJN ASW tactics and techniques never significantly improved from 1941 while the USN moved from the old S-boats through several and serially improved classes of fleet submarines.

So it should come as a surprise that a USN Gato-class fleet sub patrolling the Japanese Mobile Fleet anchorage at Tawi Tawi (USS Harder, SS-257) caught the fleet preparing to sortie back on 11 JUN. Harder had been a pain in Japanese ass for a week or more in the Celebes Sea; let me give you a little flavor of what she'd been up to on her Fifth War Patrol (from the WW2 Database):

6 JUN 44 - attempted to attack a Japanese convoy in the Sibutu Passage between Tawi-Tawi of Philippine Islands and Borneo, but was in turn targeted by two destroyers. Submerging and sailing away from the convoy, Harder fired three torpedoes out of her stern torpedo tubes, hitting destroyer Minazuki twice, sinking her. She fired another spread of torpedoes at the second destroyer shortly after; all six torpedoes missed.
7 JUN 44 - detected by a Japanese aircraft in the Celebes Sea, which called in destroyer Hayanami to attack. Harder fired three torpedoes at Hayanami, two of which hit, sinking her.
9 JUN 44 - attacked two Japanese destroyers between Tawi-Tawi and Jolo in the Philippine islands with 4 torpedoes, sinking Tanikaze with 2 hits and sinking another with 1 hit; 4 torpedoes were expended in this attack.

10 JUN 44 - detected a Japanese fleet in the Celebes Sea consisted of three battleships, four cruisers, and a number of destroyers, but was in turn spotted by a Japanese aircraft. As destroyers approached her position, she moved forward and fired three torpedoes before diving, hitting one destroyer with 2 hits. She survived a two-hour depth charge attack.
11 JUN 44 - While reconnoitering the Tawi-Tawi anchorage in southern Philippine Islands, she detected Japanese cruisers and destroyers. She did not attack; instead, she sailed out to open sea to report the finding.

This was crucial; aggressive USN submarine patrolling had blown the IJN's operational security before they had even left their anchorage. Spruance now knew that something was up and could begin preparing the tricky job of covering the Saipan invasion and fighting off the Japanese fleet.

The US submarines also acted decisively on 19 JUN.

USS Albacore (Gato-class, SS-218) had made contact with the 1st Carrier Squadron earlier in the morning of 19 JUN. She had been moved south of her original patrol position in hopes of intercepting the IJN carrier force. The Wiki entry gives a neat little description of what happened next:
"At about 0800 the next morning, Albacore raised her periscope and found herself in the midst of Ozawa's main carrier group. Blanchard allowed one Japanese carrier to pass unharmed and selected a second one (HIJIMS Taiho) for his target. Once inside 5,300 yards (4,800 m), the submarine's Torpedo Data Computer started giving false information. To maximize the odds of a hit, Blanchard fired all six bow tubes. The carrier was in the process of launching an air strike, and one of the pilots (Sakio Komatsu) intentionally dove his plane into a torpedo, setting it off early. Three Japanese destroyers immediately charged Albacore. While the submarine was diving to escape, her crew heard one solid torpedo explosion. About that same time, 25 depth charges began raining down on the submarine. Then Blanchard heard "a distant and persistent explosion of great force" followed by another..."
But in case you think that Taiho went down...well, wait. We'll talk about that some more later.

A second submarine, USS Cavella (Gato-class, SS-244) came up on the 1st Carrier Squadron from the south at about 0900hrs. This force had already been attacked by a US submarine (see below), but the Cavella's patrol report indicates that neither the Japanese escort destroyers nor the CAP was aggressively hunting submarines.

At about 1120 the carrier Shōkaku was recovering aircraft and Cavella fired a full spread of six torpedoes, hitting the carrier three times. The Shōkaku was an old veteran of the Kito Budai and her damage control parties had saved her twice, once at the Coral Sea and again at the Battle of Santa Cruz. But this time the damage was too much; she sank in mid-afternoon (Nihon Kaigun has a great analysis of the attack and sinking...).

OK. So that's the USN submarines accounted for; not a bad day at work.

How did the IJN's submarines do?
Not particularly well. We talked about the problems the IJN seemed to have with submarine warfare in the Clatsop Plains post. Despite being fairly competent at submarine design - although typically the Japanese I-boats couldn't submerge as quickly or as deep as German subs - and crew training the IJN's submarine force did remarkably little during the Pacific War.

Part of that might have been the tactical obsession with submarines as a fleet scout and warship-hunter (part of the IJN obsession with the Kantai Kessen meant that they saw their subs as part of the battleline instead of potential commerce raiders). This meant that IJN subs often ran into ASW teams trained to kill them and were often killed.

Another was their frequent use as supply ships for the island garrisons, where they were fairly useless as freighters and often sunk without a fight.

No IJN submarines were covering the Marianas in mid-June; the CINCNAV, ADM Toyoda, was still convinced the MacArthur's drive up through the New Guinea/New Britain axis was the main effort. The first subs dispatched sailed no sooner than 16 JUN when 20 boats were ordered out for the engagement area.

But as former IJN submarine officer Zenji Orita remembered in his 1974 book I-Boat Captain:
"...all Japanese submarines were to remain east of the 145th east meridian of longitude. The first battle of the Philippine Sea took place west of that, so the order kept our submarines well away from where they might have done the most good. The order was rescinded on June 20, but by that time it was too late. The major damage to our fleet had already been done. The enemy no longer feared our aircraft carriers, which were fleeing. His own carrier planes were available for heavy attacks on our island garrisons and for intensified anti-submarine patrol sweeps."
Of these twenty submarines five I-6, I-185, and I-5 were all sunk by ASW vessels and I-184 by ASW aircraft as much as 300 miles east of the battle area. Two more boats, RO-42 and I-10, were sunk in the seas to the west, and R0-36, RO-111, RO-114, and RO-117 from the waters around the Marianas themselves.

The IJN lost 11 of the 26 submarines sortied to the Marianas during June and July of 1944. Submarines RO-115 and I-6 claimed to have fired torpedoes at US carriers; no hits were reported.
What happened and why? The loss of HIJMS Taiho

Meanwhile, let's check in on our pal the HIJMS Taiho. Seemed like she shrugged off Albacore's torpedo which you'd have expected if you'd have been her designers. Taiho was the newest of the Imperial fleet carriers - just commissioned in March - and had been designed unlike her predecessors to absorb considerable damage and remain operational.

She had an armored waterline belt, armorer deck, and armor plating around her magazines, so when Albacore's torpedo hit just forward of the island on the starboard side her crew could be forgiven for thinking that the damage was manageable. The hit cracked some aviation fuel tanks and knocked the forward aircraft elevator off its track, jamming it halfway between the flight deck and upper hangar deck. Flooding drew her down by the bow, but after damage control parties planked over the forward elevator shaft (the elevator itself was too badly damaged to move) the carrier continued aircraft launch and retrieval operations for the rest of the morning.

So it must have been a pretty huge fucking shock to everyone aboard when Taiho suddenly exploded at 1430hrs, over six hours after the torpedo hit.

Here's a kind of odd but intriguing little analysis of what happened. To give you the Clif's Notes version, volatile aviation fuel - "avgas" - leaked from the damaged tanks, pooled in the forward elevator pit and the fumes began to fill the hangar decks. It's not hard to smell avgas; it stinks. The Taiho damage control parties tried to vent the stuff but without much success since both hangars were enclosed and the only way to cycle the air through was using mechanical fans. They tried hard, opening as many vents as they could and even dropping the aft elevator below the lower hangar deck level to increase the draft. It didn't help much.

Remember, the source of the problem was the pool of avgas in the forward elevator well. The simple solution would seem to have been pumping that over the side. Either this was never tried or it didn't work. Taiho also had one of the first foam fire-suppression systems in the IJN, and another solution might have been to foam the well to cover the volatilizing fuel to suppress vaporization; probably because of crew inexperience that wasn't tried, either.

Finally someone - probably Taihō‍ '​s damage control officer - ordered all doors and hatches opened and the ventilation fans turned on full to try and vent the fumes. All this did was fill the interior of the ship with vapor. All this fuel vapor found the inevitable flame or spark, and...

So at 1430hrs "a senior staff officer on the bridge saw the flight deck heave up." Her sides and probably the bottom of her hull blew open and she began to sink.

VADM Ozawa transferred his flag to a destroyer and the carrier sank around 1630hrs. How many died with her is open to question, but somewhere between 600-800 and 1,500 of her crew did not get away alive.

What happened and why? US aircraft vs IJN surface and air forces

Night 19/20 JUN 1944 - Taiho was the last vessel to sink that day; no further attacks were made by either side as darkness approached. Both fleets moved west and northwest throughout the night, the Japanese to regroup and (presumably) clear the USN submarine operating area, TF 58 to close the distance for airstrikes against the Mobile Fleet on the following morning.

VADM Ozawa was working with poor intelligence as well as operations staff work. While he obviously knew of his own losses on 19 JUN he had very little information on the condition of the Japanese land-based aviation forces, which had been well worked-over. Guam, Saipan, and Tinian were militarily worthless at this point and the other land bases, especially Truk, had lost many aircraft and aircrews.

20 JUN 44 - Both sides spent the morning trying to organize an ass-kicking party and failing.

The IJN's major problem was that the vessel that took the Mobile Fleet command party off Taiho was the destroyer Wakatzuki. As good a destroyer as she might have been (and she was; the Akizuki-class were terrific little ships, largely designed as anti-aircraft screening escorts but sturdy, versatile, and powerful light vessels) she was still a destroyer; her commo gear wasn't designed to control an entire fleet plus aviation assets scattered all over thousands of miles of the Pacific.
I cannot find out why, but it took all morning for Ozawa and his staff to finally transfer his flag into Zuikaku at 1300hrs. So it wasn't until then that the admiral learned of the disaster of the previous day, and that he had only about 100 to 150 aircraft left operational with the Mobile Fleet. Still, Ozawa gave his staff orders to organize airstrikes for 21 JUN on the still-mistaken assumption that his land-based air strength was largely intact.

The USN, on the other hand, just couldn't fucking find him.

It is sometimes difficult for us, in our satellite- and digital-information age, to imagine how vast and often unknowable the great oceans were, even as recently as 1944. Having shaken the US submarines late on 19 JUN the Japanese fleet could have headed anywhere.

Consider the geometry. Let's say that the Japanese fleet was steaming at 15 knots, a reasonable speed for a mixed force. Assume the USN lost contact at last light on 19 JUN (which would have been about 1830hrs or so) and didn't resume searching until first light on 20 JUN (about 0600). By the time the USN aircraft left their carrier decks the Japanese could have been anywhere within a circle with an area of about 77,800 square miles.

Not surprisingly the search was long and largely unsuccessful. Finally at 1512hrs an Enterprise aircraft reported sighting something - the message was fragmentary and difficult to confirm. A second message arrived about half an hour later placing the Mobile Fleet 275 miles northwest and moving west.

This was a long flight for the USN aircraft with very little daylight remaining. VADM Mitscher decided on attacking and the US carriers turned east and began launching aircraft. turned out that the actual strike profile was worse than expected. After the USN first strike had been launched the reconnaissance aircraft revised its location 60 miles further northwest, putting them at their fuel limit and making a daylight recovery impossible.

The TF 58 commander cancelled a planned second strike but chose not to recall the 226 planes on their way; 95 F6F fighters (some armed with bombs, as well), 54 TBF torpedo bombers (most armed with bombs, probably less than a third with torpedoes), 51 SB2C, and 26 SBD dive bombers.

The strike force found Mobile Fleet at around 1800; local sunset time.

Mobile Fleet's 35 A6M CAP was swarmed, and the USN aircraft were able to bore in on the capital ships in a relatively well organized attack.

Wasp's strike group attacked three fleet oilers, sinking two and damaging a third. Supposedly the choice of these relatively unimportant targets was driven by the pilots' concern over their fuel levels; they wanted to kill the first thing they found and get the hell back to their hangar deck.

The Belleau Wood Avengers found and attacked Hiyo. At least one bomb and two torpedoes hit her, damaging her propulsion and steering, starting leaks from the avgas storage tanks (again!), and starting fires. Her damage control parties managed to contain the fires for about an hour and a half but the same problems that sank Taiho caught up with Hiyo; at about 1930hrs she was ripped open by a fuel-air explosion and sank quickly with about 250 of her crew.
In addition to sinking Hiyo the USN raiders hit the carriers Zuikaku, Junyō, and Chiyoda, and the battleship Haruna with bombs; all were damaged.
(I wanted to interject something here using the difference in Zuikaku's days of 20 JUN and 25 OCT 1944. On the former she was hit by a bomb that started what sounds like a fairly severe hanger-deck fire. Her damage control parties got it under control and she steamed home under her own power. In October, acting as a decoy at the Battle of Leyte Gulf, she was hit by seven torpedoes, quickly rolled over and sank with half her crew...which brings up the navy fliers' saying; if you want to fill a ship with smoke hit it with a bomb. If you want to fill it with fucking water? Hit it with a torpedo.)
The US strike force lost about 20 aircraft to the CAP and antiaircraft fire (which was reported to be random but heavy).

The return of the US aircraft is a story in itself. Low on fuel and flying in the dark the US aircraft appeared highly likely to suffer disastrous losses as they either failed to find their carriers or crashed on landing. VADM Mitscher made the decision to abandon light discipline and illuminated his carriers, including turning the big searchlights up to make pillars of light visible for miles, and firing starshells to guide the pilots home.

Still, 80 of the 200-some aircraft failed to find a flight deck, ditching (or crashing) in the dark. About 60 of the aircrews were rescued; the 20 or so others were never found.

Night 20/21 JUN 44 - At 2046hrs ADM Toyoda ordered the First Mobile Fleet to withdraw. The IJN general staff had accepted that the Marianas, and the naval engagement, was lost. The USN conducted a brief pursuit before turning back to cover the Marianas operation.
The Outcome: Grand tactical US victory with strategic implications

The Impact: For the USN, another day at the office. The primary value of the naval victory was the protection of the Marianas amphibious force, which completed the seizure of the main islands by August. After all, Operation Forager was just another step towards the presumed invasion and occupation of the Japanese home islands some time in 1945 or 1946.

I don't want to make the operation sound easy; the 31st Army on Saipan, Tinian, and Guam died futilely but died very hard. Some 9,500 US sailors and marines were killed taking the islands or in the naval engagement. More than 65,000 Japanese were killed, including for the first time (at least, for the first time in American eyesight) women and children.
The Marianas had been Japanese before the war and many of the civilian residents hadn't left. The American soldiers, sailors, and marines had never before seen these helpless people kill themselves and each other. Now they had, and would again on Iwo Jima and Okinawa.

The USN didn't recognize the magnitude of its victory at the time. ADM Spruance was severely criticized for his caution, especially from within the naval aviation community. The flyboys wanted carnage and the relatively poor bag - since submarines had sunk two of the three Japanese carriers - was little to their liking.

The effect on the IJN from the appalling human carnage the USN visited on their aircrews was not appreciated at the time; victory meant sinking carriers and in the opinion of the Navy fliers too many had escaped.

It's worth considering, however, the difference between the effect of Spruance's caution with Halsey's recklessness at Leyte in October. Halsey, perhaps the poster-boy for USN aviation, took off after Ozawa's carrier decoy forces and exposed the invasion beaches to destruction. Had Kurita been a harder man and the sailors of the USN's Taffy 3 softer the entire Philippine campaign might have been ripped up badly that day because of Halsey's impatience.

For the IJN the Philippine Sea was the end. By 1944 naval war meant carrier war, carrier war meant pilots and aircraft, and the IJN had no more pilots and very few aircraft.

The pilot losses were the most crippling. By the fall of 1944 many of the Japanese flight schools were training Tokkō (特攻) "Special Attack" pilots - kamikazes because the technical disparity between the Japanese aircraft and the US anti-aircraft measures - aircraft, AAA cannon, radar, command-and-control - was so overwhelming. A suicide aircraft didn't need to be intact to hit its target, and a suicide pilot didn't need to be capable of deflection shooting, or bomb-aiming.

The loss of the Marianas was strategically punishing for Japan; from the islands the USAAF could begin seriously pounding her cities with heavy bomber raids. By the summer of 1945 firebombing had gutted a huge area of Japan's major cities and with them a hell of a lot of Japan's manufacturing. Thousands of people were dead, millions without homes or work.

And, of course, there was one particular raid flown from the Marianas in August of 1945...

Touchline Tattles: I don't have anything small, or whimsical, or funny about the Philippine Sea. It was, really, even more pointless than many of the other times and places that people have found to kill themselves or each other. The IJN was already whipped; it was a dead fleet floating that just didn't know when to sink. Perhaps only Leyte Gulf was more futile outside purely suicide rides like that of the battleship Yamato.

Less so, perhaps, than the Imperial Army, the Imperial Navy seems to have had problems with what looks to me like a peculiar, perhaps peculiarly-Japanese, combination of insularity, aggressive complacency, and irrational optimism - perhaps the shorter term would be 戦勝病, senshoubyou or "victory disease". The IJN - outside of a small handful of individuals - never seemed to realize the desperate hopelessness of the struggle it had started and, towards the end when that desperation had become unavoidable, it responded not by discarding its' illusions but with more desperate illusions.
It would seem to me that the responsibility of a fighting service is to fight well. But, above and beyond that, it is to advise its political masters whether or not that, even fighting as well as it can, it can achieve the political solution its' masters have asked of it. The IJN never had its von Rundstedt "Make peace, you fools!" moment; instead it followed - and led - its' people and nation to that ground zero where only death and ashes remained.

And they weren't the only sufferers. That death-march jerked a hell of a miserably high cost from the people who had to beat them down to that ash and sow the ashes with the salt of fire and steel.