Good Company

Good Company
Good Company

Tuesday, August 09, 2011

Guadalcanal Days 2 and 3: The Battle of Savo Island

 UPDATE: Let's do this like historians - up to this addition, I've given you the result and the lessons learned, but left out an explanation of how things came to pass. Start with the fact that the U.S. Navy and Marines caught the Japanese by surprise and invaded the lightly held (maybe 800 Japanese) but vital island of Guadalcanal on 7 August 1942. See my previous post and the links therein covering how this was accomplished and why here. Now, let's get into RADM Morison's The Two Ocean War as he sets the scene:
Early in the morning of 7 August news of the American landings reached Vice Admiral Mikawa at Rabaul. His decision was prompt and intelligent - to send reinforements to the Tulagi-Guadacanal garrison, and to assemble a task group to attack the American ships unloading there. {Note by E1: The ship carrying the troops is sunk by an American submarine around midnight on 8 August. It's the task force that is the focus of the battle]
Mikawa's naval reaction is a very different story. One hour after receiving the bad news from Tulagi, he began collecting ... a task group to attack the American Expeditionary Force. It was a "scratch team," the ships had never trained together before, but it proved to be good enough for the task in hand. Heavy cruisers Chokai (flagship), Aoba, Kako, Kinugasa and Furutaka, light cruisers Tenryu and Yubari and one destroyer, Yunagi, made rendezvous in St.George Channel around 1900 August 7 and started hell-bent for Guadalcanal. Mikawa's battle plan . . . was to enter Ironbottom Sound in the small hours of the 9th, strike the warships guarding the expeditionary force, shoot up the unloading transports, and retire. An excellent plan; but the chances of detection were great, as the striking force had to steam in full daylight down the "Slot" between the central Solomon Islands before entering the cover of darkness.

Owing to a series of blunders on our side, [5/7/2013 UPDATE NOTE: AN INACCURATE ALLEGATION BY MORISON HAS BEEN TOTALLY REFUTED BY A NUMBER OF SOURCES PLEASE SEE COMMENTS BELOW - THIS POST HAS BEEN EDITED TO REFLECT WHAT IS NOW KNOWN]  . . . one (or more) sighting(s) of Mikawa's force that day (was/were mishandled) . . .  Admiral Turner did not receive it (them) until over eight hours had passed. (Further what was reported evidently was miscontrued) On that basis (i.e. the misunderstanding of what was seen) Turner made the bad guess that the Japanese were not coming through that night, but intended to set up a seaplane bases at Santa Isabel Island, some 150 miles from Savo, and attack later. [As indicated here below and in other places, Admiral Turner and/or his staff got it wrong in many, many ways.]

Dogmatically deciding what the enemy would do, instead of considering what he could or might do, was not Turner's only mistake on that fatal night. He allowed his fighting ships to be divided into three separate forces to guard three possible sea approaches by the enemy. Rear Admiral Norman Scott with two light cruisers and two destroyers patrolled the transport area between Tulugi and Guadalcanal, and never got into the battle.
***[Note by E1: Morison lays out the other groups, commanded by Royal Navy Rear Victor Crutchley, VC - but coordination among his ships was - uh - lacking]

Turner was so certain that the enemy would not attack that night he made the further mistake of summoning Crutchley, in Australia, to a conference on board his flagship . . . some 20 miles away . . . This action of Turner's stemmed from the worst of all blunders that night: Admiral Fletcher's decision to retire his three-carrier task force from covering position, depriving the landing force of air covernext day. **** He commenced this withdrawal at about 1810 August 8 without consulting Turner, who was below him in the chain of command. That was why Turner felt he had to confer with Crutchley and Vandegrift {Marine commander] to decide whether the partially unloaded transports should depart that night, or risk more Japanese air attack without air protection. Consequently, cruiser Australia and the O.T.C. [Officer in Tactical Command] were not on hand when badly needed, and the depleted cruiser group south of Savo Island was commanded by Captain Bode of Chicago, who acted as if dazed.
And then the confusion began.

From HyperWar: The Battle of Savo Island [ONI Combat Narrative], a discussion of what Samuel Eliot Morison called, ". . . [P]robably the worst defeat ever inflicted on the United States Navy in a fair fight":
The disposition of our cruisers and the remaining destroyers was governed by "Special Instructions to the Screening Group," issued by Rear Admiral V. A. C. Crutchley, R. N., commander of the escort groups and second in command of the Amphibious Forces. To protect the disembarkation area from attack from the eastward, the American San Juan and the Australian Hobart, both light cruisers, were assigned to the area east of longitude 160° 04' E., guarding Lengo and Sealark Channels. They were screened by the destroyers Monssen and Buchanan. At 1850 these ships began their patrol at 15 knots on courses 000° and 180° between Guadalcanal and the Tulagi area.

As a precaution against surprise from the northwest, two destroyers were assigned to radar guard and antisubmarine patrol beyond Savo Island. The Ralph Talbot was north of the island, patrolling between positions 08° 59' S., 159° 55' E. and 09° 01' S., 159° 49' E. The Blue was stationed west of the island between positions 09° 05' S., 159° 42' E.4 and 09° 09' S., 159° 37' E., patrolling on courses 051° and 231° at 12 knots.

The area inside Savo Island, between Guadalcanal and Florida, was divided into two patrol districts by a line drawn 125° T. from the center of Savo. It was upon the vessels patrolling these sectors that the Japanese raid was to fall. The area to the north of this line was assigned to the heavy cruisers Vincennes, Astoria, and Quincy, screened by the Helm and Wilson. The last-named replaced the Jarvis, which had been damaged by a torpedo during the day's air attack. This group was patrolling at a speed of 10 knots on a square, the center of which lay approximately midway between Savo and the western end of Florida Island. At midnight it turned onto course 045° T. and was to make a change of 90° to the right approximately every half hour.

The area to the south of the line was covered by the Chicago and H. M. A. S. Canberra, screened by the Patterson and Bagley. H. M. A. S. Australia was the flag and lead ship of this group, but at the time of the action she was absent, having taken Admiral Crutchley to the conference aboard the McCawley. Capt. Howard D. Bode of the Chicago was left in command of the group, although the Canberra ahead of his ship acted as guide. The group was steering various courses in a general northwest-southeast line--the base patrol course was 305°-125° T.--reversing course approximately every hour.

Admiral Crutchley's instructions were that in case of a night attack each cruiser group was to act independently, but was to support the other as required. In addition to the Melbourne warning, a dispatch had been received indicating that enemy submarines were in the area, and night orders placed emphasis on alertness and the necessity for keeping a sharp all-around lookout. The destroyers were to shadow unknown vessels, disseminate information and illuminate targets as needed. It was provided that if they should be ordered to form a striking force, all destroyers of Squadron FOUR except the Blue and Talbot were to concentrate 5 miles northwest of Savo Island. This arrangement was to cause some confusion during the battle.
"The fact must be faced that we had an adequate force placed with the very purpose of repelling surface attack and when that surface attack was made, it destroyed our force," said Admiral Crutchley. After full allowance for the element of surprise and for the fact that the attacker at night enjoys an immense advantage, there remain many questions about the action which cannot be answered.
The redeeming feature of the battle was the splendid performance of our officers and men. They had been on the alert for days and had had about 48 hours of continuous, active operations immediately before the battle. In spite of this, their conduct under the most trying circumstances was beyond praise, and they made it, in the happy phrase of one of our officers, "a night in which heroism was commonplace."
HMAS Canberra sinking

Instead of repeating the ONI Combat Narrative here, please go read the Hyperwar link. It is well worth your time.

You can also find a report in the USNI Solomons Campaign series at Execution at Savo Island by CDR Bill Ballard:
There is no shortage of lessons learned from Savo Island.  To pull out just a few of the big ones:

Communication between supporting/-ed commanders: With the entire operation in its infancy, URR’s lesson of streamlined dissemination and obtaining of INTEL and aerial reconnaissance information is still applicable.  Allied shore-based reconnaissance aircraft were either under the command of COMSOPAC or COMSOWESPAC, with little effective communication or coordination.  COMSOWESPAC aircraft detected (and mis-identified) Mikawa’s force the morning of the 8th; by the time it reached CTF 62 it was too late to send any more aircraft to either further identify or attack them.

Capability vs. Intent: Be ready for what your enemy CAN do to you, not what you think your enemy WILL do to you.  Taking two seaplane tenders out of the mis-identified formation, there were still five ships that were capable of reaching Savo that night.  Taking this into account, had some or all of TG 62.6 been in Condition I through the night, with a real plan or set of PPR’s in place…

Trusting your Technology: Only if you understand its limitations.  On a tactical level this battle was less a case of visually targeted torpedoes defeating RADAR-directed gunnery and more a case of over-reliance on incorrectly employed RADAR leading to the complete surprise of TG 62.6.  Even though in training BLUE and RALPH TALBOT exhibited RADAR detection ranges greater than Japanese visual detection ranges, the long wave SC RADAR’s performance suffered terribly near land.  Additionally, their assigned search tracks at times left 20 mile gaps between the two ships. (Bates, pg. 350)  Interestingly, the most capable Allied RADAR, the SG RADAR, was aboard SAN JUAN and never saw action.  One wonders what the outcome could have been if, between Crutchley, Scott and their staffs, they had decided to put SAN JUAN in one of the picket stations.

Technology vs. Tactics, Training and Procedures (TTP): The action at Savo commenced with both sides well within the effective range of their preferred weapons.  This is the range at which training pays off.  Simply put: the IJN had trained; the USN hadn’t.  Of the Allied ships, only ASTORIA had conducted any recent night target practice, and it had been at least eight months (more than likely indefinite in the case of QUINCY and VINCENNES, recently arrived from the Atlantic) since any other cruiser had conducted either night target exercises or a night battle problem. (Bates, pg. 356)

US Navy losses: From here:
Battle of Savo Island
Navy 936 dead 11 wounded
Marine 33 dead
UPDATE: Added some additional paragraphs from the ONI report to set some background.


  1. ewok40k12:49 AM

    it could be worse - had the IJN task force pressed attack on transport ships it would be bad karma for the Marines...
    fortunately fog of war works both ways and Mikawa was afraid of US carriers that were at the time retired away...

  2. Yes, the Japanese failed to press their attack, and that was the good news - because it allowed the Marines to stay.

    But, it also didn't have to be as bad as it was.

  3. Anonymous12:43 PM

    I would like to draw your attention to this site which contains a participants account of the Savo Island battle and adds accounts of the naval battles of the Guadalcanal campaign.

    I would suggest that the Hudson pilot has been badly misrepresented by Morison and there is a lot more to the neglect of sightings than he makes out.

    Even the IJN records show the RAAF pilot broke radio silence whilst in visual contact with them.

    A number of other sightings were made by US and Australian pilots and for whatever reasons they were not acted upon, it was not the fault of the Allied pilots who made them. It's past time to put that inaccuracy to rest once and for all.

    For a little more detail on the reporting:-

    Gregory's site isn't itself a primary source but he provides guides, links to same.

    I enjoy reading your blog but as an historian I had to comment on this matter, no offence intended.


  4. McFriday: You are correct - one of the points made in Hornfischer's new book on the Guadalcanal sea battles is that the Hudson pilot, William Stutt actually went to extraordinary lengths to make sure his warning went out - only to have it caught up in the rat's nest of communications of the day.

    Both Morison and I owe Mr. Stutt an apology.

    No offense taken, we all learn when someone with more knowledge lets us know when we've wandered off the right path. I appreciate you taking the time to set me right.

    We'll be discussing Mr. Hornfischer's new book on Midrats this Sunday at 5pm U.S. Eastern here - I expect to have a lot more of my ignorance improved... you can download the show there or from iTunes, too, if you can't listen live.

  5. Anonymous4:59 AM


    Thank you for your courteous reply, it was not just Stutt I was defending but the USAAF crews as well.

    I will listen in one way or another, being in Australia I'm not sure if I can catch the interview live.

    I also like Mr. Hornfischer's books and will add this one to my shelf in due course, so listening to him will be a bonus.

    I learn a lot from here, glad I could contribute in some small way to healing an unwanted sore point in our combined histories.


  6. Anonymous8:11 AM


    Re. Midrats #84. [and 4 others so far!]

    Compliments to you and Sal,'Midrats' now has one more regular listener.

    I have an even greater regard for Mr. Hornfischer's work since listening to him.


  7. Anonymous3:38 AM

    The comment about the RAAF Hudson pilot (the one sighting of Mikawa's force that day, by an Australian Hudson pilot at 1026 was so mishandled by him,) is completely erroneous and should be deleted from this site.
    I have been in communication with the radio operator of the Hudson who is still, at the age of 94, attempting to clear his name and the names of his fellow crew from this slur.
    Samuel Morison either invented the story about the crews' delay in reporting the sighting or repeated, without checking, unfounded rumours spread by some survivors.
    Other authors have repeated these fallacies without checking.
    The official United States Government position, as relayed in a communication from their Embassy in Canberra is that they in no way hold the aircrew of the Hudson in any way responsible.
    Your statement is also incorrect in claiming that there was only one sighting as another Hudson sighted the Japanese ships about 30 minutes later and also reported them.
    In reality the fleet was sighted by two Australian aircraft on the day prior to the battle and none from the United States.

    Lance Fishman
    Aviation Historical Society of Australia

    1. Lance:

      If you look through the earlier comments, you will see that there was a discussion of how wrong Morison was.

      Initially I thought that was enough to "correct" Morison's error, but your comment has prompted me to revised the post to reflect that whatever happened was not the result of the RAAF pilots but rather falls directly on ADM Turner and his staff.

      It appears, even to Morison, that ADM Turner got caught in believing the enemy would act as Turner would act and Tuner and his staff did not fully assess what other paths the Japanese might follow.

      There were many "flag level" errors made early in the war. Many such errors were founded in underestimating the Japanese capabilities.

      I don't know where Morison got his bad info, but it has the air of a staff CYA.

      Thanks for taking the time to write and point out the problem. Let me know if you see further issues.