Part I: Prologue to the End
Blank page
Chapter 1: Strategic Background
In a report submitted to Secretary of the Navy James V. Forrestal on 12 March 1945, the Commander in Chief, United States Fleet (CominCh), Fleet Admiral Ernest J. King, stated that:–
The amphibious operations of the spring, summer and autumn of 1944 carried our forces such great distances across the Pacific that in February 1945 they were enabled to begin the assault upon the inner defenses of the Japanese Empire itself.1
Recognizing all that had been accomplished to the date of his report, Admiral King at the same time cautioned against complacency and warned of “a long, tough and laborious road ahead.”2
Among the many factors leading to the favorable Allied posture in the Pacific at the beginning of 1945 was the strategic concept for the prosecution of the Pacific War adopted at the Cairo Conference (SEXTANT) in December 1943. In essence, President Roosevelt and Prime Minister Churchill agreed upon a grand plan that dictated the thrust of two concurrent and mutually supporting series of operations across the Pacific towards the heart of the Japanese Empire. These drives along separate approach axes would establish bases from which a massive effort could be launched against the Formosa–Luzon–China coastal areas in the spring of 1945.
One drive, to be mounted by Allied forces under General Douglas MacArthur, Commander in Chief, Southwest Pacific Area (CinCSWPA),3 was to move along the northern coast of
New Guinea and thence to the Philippines; in the second, forces of Admiral Chester W. Nimitz, Commander in Chief, Pacific Ocean Area (CinCPOA), would push through the Central Pacific to the core of Japanese island defenses guarding the heart of the Empire. During this two-pronged advance, the major components of the Pacific Fleet, under Nimitz as Commander in Chief, Pacific Fleet (CinCPac), would support, as assigned, specific amphibious operations within both strategic command areas, and at the same time contain the Japanese fleet.
Almost immediately after the two heads of state had approved at SEXTANT the revised plan for the defeat of Japan, the Joint Chiefs of Staff (JCS), forwarded the directive to MacArthur and Nimitz, whose staffs and commands took steps to implement it. January 1944 opened with a landing at Saidor in New Guinea. At the end of the month, Central Pacific forces landed in the Marshalls and spent February thrusting deeply into the island group to collapse those outposts of the imperial defenses.4 By the end of March, the Bismarck Archipelago barrier had been permanently breached and airfields and harbors seized in the Admiralties. MacArthur’s forces began the drive up the New Guinea coast in April, with landings at Aitape and Hollandia. With the naval attack on and immobilization of Truk, the capture of Guam, Saipan, and Tinian in the Marianas during the summer of 1944, and the defeat of the Japanese fleet in the First Battle of the Philippine Sea, the Central Pacific drive cut the inner ring of Japanese island defenses in several places and consolidated footholds from which the drive westward was to continue. After the amphibious assaults on Peleliu and Angaur and the unopposed capture of Ulithi for use as a fleet anchorage and an advance base, Admiral Nimitz’ forces stood poised on the threshold of the Japanese defenses ringing the Home Islands.5
By the end of July, Admiral William F. Halsey’s South Pacific troops had advanced up the Solomons, and MacArthur’s forces along hundreds of miles of the northern coast of New Guinea, in a series of leapfrogging operations. Thousands of Japanese soldiers on Bougainvillea, New Britain, New Ireland, and New Guinea itself were neutralized and isolated, and beyond hope of being effectively employed elsewhere. In September, MacArthur’s forces occupied Morotai, southeast of the Philippines, before the planned landing on Mindanao.
In the course of naval covering strikes prior to the landings on Morotai and in the Western Carolines, Admiral Halsey’s
fast carrier forces had discovered surprisingly weak enemy resistance in the central Philippines. In a follow-up to this discovery, the line of advance through the Southwest Pacific was reoriented northwards. Fully aware of “the necessity of being alert for symptoms of enemy weakness and of being ready to exploit them,” Halsey recommended an early return of American troops to the Philippines in the Leyte–Samar area and cancellation of certain operations scheduled elsewhere.6 His recommendation was approved by the JCS.
In the overall planning for the defeat of Japan, the strategists anticipated that the final phase of the Pacific War would involve a massive assault against the industrial heartland of the Empire by means of amphibious landings on the southern coast of Honshu in the area bounded by Shimonoseki in the south and the Kanto Plain near Tokyo in the north. Successful Allied operations in 1944 had brought ultimate victory into sight, and submarine blockade and air bombardment both had the Japanese viewing ultimate defeat, but some American commanders doubted the wisdom of using the Formosa–Luzon–China area as a springboard from which to launch the attack against Japan in 1945. More importantly, they believed that valuable time was being wasted and that a decision had to be made. In view of the SEXTANT Plan, and the advanced state of the operations against Japan, JCS planners were confronted with the problem of whether American forces should: (1) move onto Luzon and the rest of the Philippines, (2) invade only Luzon in the Philippines and also strike at Formosa and the China coast, or (3) attack the Philippines, Formosa, and the China coast. Arising out of the third option was an additional thorny problem—which area to attack first.
While Admiral King and some planners in Washington considered the possibility of entirely bypassing the Philippines, this concept was apparently only a minor aspect of the major effort by many officers to have Luzon, in particular, bypassed. The alternative to this was the seizure of Formosa. On the other hand, ample evidence exists to indicate that those who sought the Formosa objective did not intend this to be an exclusive operation, for they believed that the invasion of Luzon could proceed simultaneously with the Formosa operation or take place at a later date.7
Determined to return to the Philippines, MacArthur doubted the necessity of the Marianas campaign but generally approved the Palaus landings since they would directly support his impending operations. Admiral King took just the opposite view; he concluded that the occupation of the Marianas was essential and that the necessity of recapturing all of the Philippine Islands was questionable. Furthermore, he was firmly convinced that the main American effort should be bent in mounting a drive across the Central Pacific to Formosa
and then on to the China coast.8 Although various subordinate commanders in the Pacific Ocean Area held conflicting views regarding what course should be taken for the final phases of the war, the SEXTANT decision made it imperative that their staffs spend most of 1944 in planning for Operation CAUSEWAY, the invasion of Formosa, projected for the spring of 1945.
On the basis of a JCS directive issued on 12 March 1944,9 the prevailing conflict was partially allayed. Admiral Nimitz was directed to land on 15 June in the Southern Marianas and on 15 September in the Palaus. General MacArthur was instructed to seize Hollandia in April and make plans for a landing on 15 November on Mindanao. Contained in the JCS order was a statement of long-range objectives that required Nimitz as CinCPOA to prepare the plans for an assault early in 1945 on Formosa, and assigned CinCSWPA the responsibility of planning for the recapture of Luzon “should such operations prove necessary prior to the move on Formosa.”10
In view of the March JCS directive, which outlined the general concept of CAUSEWAY, Nimitz reconsidered and revised the troop list for the operation many times, and finally designated the task force commanders. Admiral Raymond A. Spruance, Commander, Fifth Fleet and Central Pacific Task Forces was to be in overall charge. Vice Admiral Richmond K. Turner was to command the expeditionary forces, and Lieutenant General Simon B. Buckner, Jr., USA, was to command the expeditionary troops and the Tenth Army.11
Further discussion regarding what the nature of Pacific strategy was to be following the Marianas operation continued after the JCS had directed the preparation of plans for CAUSEWAY. This topic was the subject of one of the periodic conferences which Admirals King and Nimitz and their key deputies held throughout the war, either at Pearl Harbor or San Francisco. At one such meeting on 6 May 1944, Vice Admiral Charles M. Cooke, Jr., King’s chief of staff, pointed out that, although the JCS directive envisioned a landing on Formosa in February 1945, the best time for this operation—in view of other considerations—would probably be November–December 1944.12 Cooke also noted that once Japan had been cut off from the mainland, her islands could be bombed and perhaps Kyushu even invaded.
During 1944, the Joint War Plans Committee (JWPC) had also considered
what the nature of future Pacific strategy should be. In early June, it issued a comprehensive study which far exceeded in scope and perspective the previous strategic positions taken by the Joint Chiefs and the Combined Chiefs of Staff (CCS), and outlined a series of campaigns that would lead to an assault on the Tokyo Plain by the end of 1945.13 In this study, the JWPC pointed out that in view of the present and anticipated rate of advance of increasingly stronger American forces in the Pacific, it would appear that the “Inner Zone Defense of Japan” would be reached by spring 1945. The study concluded that the overall strategy approved at SEXTANT was inadequate, i.e., future operations as planned extended only to the perimeter of the Formosa–Luzon line.
Instead, the JWPC recommended a new schedule or strategic concept for ending the war in the Pacific. The committee suggested that three phases precede the invasion of Japan: (1) During the period 1 April to 30 June 1945, American forces would seize positions in the Bonins and the Ryukyus from which they would launch an invasion against the central China coast in the Hangchow Bay area; (2) They would spend the time from 30 June to 30 September in consolidating and initially exploiting the China beachhead; and (3) The forces would land in Southern Kyushu 1 October and on the Tokyo Plain on Honshu on 31 December. This planning paper was passed to the Joint Staff Planners, who approved and forwarded it to the Joint Chiefs of Staff, who likewise favored the revised concept. On 11 July, the CCS received the study along with a JCS recommendation that the SEXTANT timetable for operations in the Pacific be changed to reflect the suggested JWPC schedule.
At one of the meetings held during the CominCh-CinCPac conference in the period 13–22 July 1944, Admiral King informed the conferees of the JCS action regarding the JWPC study. He also indicated that he believed Luzon could not be invaded before Formosa or Japan without the Americans first investigating what Saipan and Guam could offer in the way of fleet anchorages and base facilities for the support of the Luzon invasion forces.14 Vice Admiral John H. Tower, Commander, Air Forces, Pacific Fleet, stated that neither the areas in American possession at that time or prospectively available would permit the establishment of naval and supply bases which would be adequate for the support of the future operations contemplated in the JWPC study.15
Along these lines, it was suggested that the feasibility and advisability of
taking San Pedro Bay in Leyte Gulf as a fleet anchorage be investigated. Although considerable discussion of this recommendation resulted, no firm decision was made at this time.
Regarding the invasion of the Bonins, Admiral Tower stated that, because the United States plans for the establishment of VLR (very long range) bomber bases in the Marianas were close to being realized, steps to enhance their effectiveness should be taken at the earliest practicable date. This meant the seizure and development of positions in the Bonins, where fighter and bomber aircraft stationed on fields developed there could supplement and support the planned air raids on Japan. On the other hand, Admiral Tower added that a study of the prospective employment of fleet and assault forces did not indicate the Bonins could be taken until 1945, unless the timing of then currently planned operations could be drastically revised. Because these were of greater importance in the overall scheme for the defeat of Japan, the occupations of Iwo Jima and Chichi Jima would have to be deferred.16
Most of the others present at this meeting generally agreed with Admiral Tower’s conclusions. Rear Admiral Forrest P. Sherman, Nimitz’ chief of staff and head of his War Plans Division, pointed out that for the invasions of Leyte and Formosa—the two major operations of a decisive nature scheduled following the completion of the landings in the Palaus—American forces had been tailored down considerably. Sherman emphasized that if more ships and troops became available, they should be employed to supplement those already assigned to the landings on Leyte and Formosa. In no case, should they be diverted for such “minor operations” as the occupation of Iwo Jima and Chichi Jima.17
Admiral King agreed with this line of thinking, and added that it would be unwise to assault the Bonins until American forces were ready to invade Japan following the Formosa operation. Nonetheless, at this time, he directed CinCPac to prepare plans for the invasion of the Bonins.
As for determining those objectives that were to follow the capture of the Southern Marianas and the Palaus and were to be mounted before the invasion of Japan, even President Roosevelt’s visit to Pearl Harbor late in July to confer with MacArthur and Nimitz failed to resolve the impasse. When the conference began, Admiral Nimitz, the first to speak, presented the Navy position.
Contrary to general belief, no real controversy arose between Nimitz and MacArthur regarding the conduct of future operations against Japan. Nimitz made this quite clear in a letter to Admiral King, summarizing in a few words the discussions at the Pearl Harbor meetings. Nimitz told CominCh that:
... our conferences with the Commander in Chief of the Army and Navy [President Roosevelt] and the Commander in Chief, Southwest Pacific Area [MacArthur] were quite satisfactory. The general trend of the discussion, like our own, was along the line of seeing
MacArthur into the central Philippines, thereafter going direct to the Formosa Strait, and leaving the SWPA forces to work into Luzon under the cover of the Formosa operation. It was made clear that the time has not yet arrived for firm decisions on moves subsequent to Leyte.18
It is possible, however, that Nimitz and some of his staff had some doubts on the feasibility of the Formosa operation and the concept underlying the Formosa-first policy. Partial evidence for this is found in the fact that CinCPac staff members had prepared plans to seize Okinawa as a substitute for Formosa “well before such an operation gained serious consideration among high-level planners in Washington.”19
After listening to the views of both MacArthur and Nimitz, the President returned to Washington without rendering a decision on the courses of action to be followed after the landings on Leyte. Nor does it appear that a firm decision for post-Leyte operations was expected. Although Nimitz may have entertained other opinions concerning future strategy, he was still operating under a JCS directive relative to the Formosa operation. On 23 August 1944, the CinCPac joint staff study of CAUSEWAY was published. In this document Admiral Nimitz indicated that he intended to invade Formosa after SWPA forces had established positions in the south and central Philippines. Following the successful operations on Formosa, the Ryukyus and the Bonins or the China coast were to be invaded as a prelude to the assault on Japan itself. A Luzon operation, as such, was not mentioned in this plan.
The dispute remained unresolved until 9 September, when, at the Quebec Conference (OCTAGON), the Combined Chiefs of Staff formally adopted and incorporated the JWPC concept within the SEXTANT schedule for the defeat of Japan, and in effect revised it. For planning purposes, the CCS then approved a new schedule of operations, which ended the campaigns of 1945 with a landing on Kyushu in October and on the Tokyo Plain in December.20
The Combined Chiefs also agreed that, if the Formosa operation materialized, it would be preceded by invasions of the Bonins in April, the Ryukyus in May, and the China coast in the period March to June 1945. On 15 September 1944, the JCS further clarified impending Pacific operations by canceling the scheduled invasions of Mindanao and Yap and
Admiral Nimitz briefs General MacArthur, President Roosevelt, and Admiral Leahy at the July 1944 Pearl Harbor conference. (USA SC207297)
Combined Chiefs of Staff meet with President Roosevelt and Prime Minister Churchill at the OCTAGON Conference in Quebec, September 1944. (USC SC 194469)
setting 20 October as the date for the invasion of Leyte.21
On 11 September, Lieutenant General Millard F. Harmon, Commanding General, Army Air Forces (AAF) , Pacific Ocean Areas, proposed the abandonment of the Formosa operation in favor of amphibious landings in the Bonins and the Ryukyus. Harmon recommended the capture of Iwo Jima by 1 January 1945 and Luzon by 1 June 1945. Further, he suggested that POA troops seize Okinawa and Amami O Shims after MacArthur’s forces recaptured Luzon; Kyushu was to be invaded in September 1945. Harmon also stated that he believed that the seizure and use of Luzon was an important consideration in the overall strategy of the Pacific War and that the launching of a major operation against Formosa would dilute some of the force being applied against the Japanese in other action areas.22 In order to husband resources and to accelerate the march toward Japan, Harmon believed that the capture of Luzon for its airfields was imperative. Air operations launched from Luzon could neutralize Formosa and effectively cut Japanese communications to South China and Malaya.23
Less than a week later, after a review of the plans contemplated for the CAUSEWAY operation, Admiral Nimitz set forth his thoughts in a letter circulated to his senior commanders. In a key section of this letter, CinCPac recommended to consider the possibility:–
... of a re-orientation of a strategy in the Pacific which will provide for an advance northward with eventual assaults on the Empire itself, rather than intermediate action along the China Coast, thus indicating the probability of occupation of Iwo Jima and Okinawa with target dates as early as practicable after CAUSEWAY.24
He also directed Admirals Spruance and Turner and General Buckner to recommend suitable physical targets in the Formosa–Amoy–Pescadores areas for Operation CAUSEWAY. Criteria for the selections were the number of naval and air bases that would have to be established and the type and total of major troop units required. On 26 September, General Buckner submitted what he considered to be the primary objection to the entire projected operation; he said that the shortage of available supporting and service troops in the
POA for CAUSEWAY made it infeasible.25 In an afterthought, on 4 October Buckner wrote Nimitz that the need for invading Formosa would be diminished greatly if plans for the invasion of Luzon came to fruition.26
The minutes of the CominCh-CinCPac conference held in San Francisco from 29 September to 1 October 1944, indicate that by this time, Admiral King had given serious thought to bypassing Formosa. He told Nimitz that, at his proposal, the Joint Logistics Committee (JLC) had made a survey of the resources available for the Formosa operation, and that the report of this committee was very discouraging. At the time of its survey, the JLC found that resources were not available for CAUSEWAY, and would not be available unless Germany capitulated a long time before it was expected to do so.
CinCPac then told King of General Buckner’s requirements for additional men and equipment and that he, Nimitz, was in no position to dispute these figures. Nimitz then submitted a memorandum recommending changes for future Pacific operations based on the nonavailability of necessary resources and the favorable results of recent carrier operations. Admiral Nimitz recommended that CinCPac forces support the SWPA invasion of Luzon with a target date of 20 December 1944, and the invasions of Iwo Jima on 20 January 1945 and of Okinawa on 1 March by POA Forces.27
CinCPac stated that the proposal for the SWPA forces to work up through the Philippines from Leyte by shore-to-shore operations had been discussed with President Roosevelt and General MacArthur at the Pearl Harbor conference in July. Because MacArthur had stated that he could not undertake these operations and in view of the insufficient resources for Formosa, Nimitz believed that the best way to keep pressure on the Japanese was for him to support the Lingayen Gulf operation proposed by MacArthur and to take the Bonins and the Ryukyus with POA forces.28
Admiral Sherman then told King that Nimitz expected to take Iwo Jima with two divisions and then to send in large numbers of construction personnel to build up the airfields rapidly. Following that, assuming that enemy air power on Formosa had been neutralized by carrier strikes assisted by shore-based air from Luzon, it was expected that Okinawa would be invaded on 1 March. King asked Nimitz why he was going to seize the Bonins if Okinawa was to be taken.
Nimitz replied that fighters based in the Bonins could give protection to the B-29s raiding Japan, and that the AAF wanted this added protection.
King returned to Washington and on 2 October proposed to the JCS a course of action for the Pacific, He stated that in view of the lack of necessary resources in the POA for CAUSEWAY, and because of the inability of the War Department to makeup the deficit before the end of the war in Europe, he believed that operations should be mounted against Luzon, Iwo Jima, and the Ryukyus in succession. He also added that CAUSEWAY might be feasible at a later date if conditions in the Pacific and Europe warranted.29 Concurring with King’s proposal, the next day the JCS ordered MacArthur to invade Luzon on 20 December 1944 and Admiral Nimitz to land Marines on Iwo Jima on 20 January 1945. Following these operations, Operation ICEBERG was to be launched on 1 March 1945.30 This date was flexible, however, since it would be affected by the: (1) Capture of Iwo Jima in time for the prompt release of fire support units and close air support squadrons required at Okinawa; (2) Prompt release of supporting naval forces and assault shipping from the Luzon operation; and (3) Attainment of undisputed control of the sea and air in the target area in preliminary strikes against the Ryukyus, Formosa, and Japan.31
With all attention and efforts now focused on the new objectives, the Formosan venture was reserved as a strategic goal for possible future reconsideration.32 Although the basic command concept and troop list organization that had been set up for CAUSEWAY were retained for employment in ICEBERG, 33 there was much to be done between the time that the JCS ordered the capture of Okinawa and the actual date of the invasion.