The Invasion That Never Was

At the time of the Hiroshima and Nagasaki bombings in August 1945, an invasion of Japan was scheduled to begin on 1 November. While the Joint Chiefs of Staff (JCS) hoped to defeat the Japanese quickly, they did not deny the possibility of a protracted campaign and drafted plans for the possible seizure of a number of key areas in the Home Islands.

The earliest mention of an invasion surfaced in a May 1943 document prepared by the Joint War Plans Committee (JWPC). The plan, JWPC 15, envisioned the seizure of bases in eastern China, where the Allies would launch missions to bomb, blockade, and invade Japan. The planners did not consider an invasion inevitable but stated that “the unconditional surrender of Japan may require the invasion of the Japanese homeland” and that bombing and blockade alone might bring about surrender.¹

During the next year, some discussion did concern an invasion, but only in the most general terms. Meanwhile, U.S. forces were advancing steadily in the Pacific. The Marshall Islands fell in February 1944, and the Mariana Islands were invaded in June 1944- In the Southwest Pacific, General Douglas MacArthur’s forces leapfrogged along the northern coast of New Guinea. Future plans included assaults on the Palaus in September 1944, Mindanao in November 1944, and either Luzon or Formosa in February 1945. These operations conformed to the overall objective of the Pacific war: “To obtain objectives from which we can conduct intensive air bombardment and establish a sea and air blockade against Japan, and from which to invade Japan proper should this prove to be necessary" [emphasis added].² No specific plans were in place, however, following the invasion of Formosa. The JCS ordered the Joint Staff Planners (JPS) to develop a final phase war strategy, which they titled JCS 924.³

Since the overall initial objective would be achieved with the seizure of Formosa, JCS 924 included a new objective: “To force the unconditional surrender of Japan by (1) Lowering Japanese ability and will to resist by establishing sea and air blockades, conducting intensive air bombardment, and destroying Japanese air and naval strength, (2) Invading and seizing objectives in the industrial heart of Japan” [emphasis added].⁴ This represented a major change in policy. Prior to this, invasion was nothing more than a possibility, not a necessity. As long as U.S. forces were far from Japan, any invasion decision could be deferred. But as they drew closer, addressing the problem became a necessity.

The joint planners recommended invasion for two reasons. First, while some officers—among them Fleet Admirals William D. Leahy and Ernest J. King—maintained that bombing and blockade would bring about a Japanese surrender,” the planners considered such a view as “overly optimistic.”⁵ Second, even if such an approach were successful, it would “involve an unacceptable delay in forcing an unconditional surrender.”⁶ An invasion, therefore, seemed the best way to end the war.

The JPS used three criteria for selecting targets: The thrust to the industrial heart of Japan should take the fewest steps possible, assaults should focus on areas where local tactical superiority could be achieved and maintained, and the area should be capable of supporting large numbers of forces. Therefore, the planners rejected a direct assault on Honshu, for they considered tactical superiority practically unachievable. Shikoku proved impracticable for the same reason as well as for its geographical limitations. The JPS considered six launch sites for the eventual invasion of Honshu: Korea, Hokkaido, Kyushu, and the Shanghai, Shantung, and Foochow- Wenchow areas of China.⁷

Shanghai, Shantung, and Korea were unsuitable, because the risk was too high of becoming involved in protracted campaigns against major enemy forces. Foochow-Wenchow was too underdeveloped and would require too much effort to make it a suitable base. Hokkaido was lightly defended, but it was far from the primary U.S. lines of communication in the Central Pacific as well as beyond fighter range of most Japanese industrial centers. Overall, Kyushu was the best choice. Its loss would contribute greatly to the isolation of Japan; deprive the Japanese of vital industry; and provide bases from which to bomb the industrial heartland of Japan. Any assault, however, probably would encounter stiff opposition.

Therefore, supporting operations in southeastern China, the Bonins, and the Ryukyus would precede the seizure of Kyushu on the following timetable: Bonins, 15 April 1945; Ryukyus, 1 May 1945; Kyushu, 1 October 1945; and Honshu, 31 December 1945. An estimated 10 divisions would be required for the Kyushu operation, and 14 for an assault on Honshu. The availability of troops would depend to a large extent on the redeployment of forces from Europe to the Pacific after the defeat of Germany.⁸

Based on JCS 924, the Joint War Plans Committee began work on invasion plans. One of the earliest, JWPC 235/3, outlined an assault on southern Kyushu by 13 divisions on 1 October 1945.⁹ Eight divisions make the initial assault, with the rest arriving by around D+15. The main targets were the port of Kagoshima and the airfield complex near Kanoya, which required landings at three points. The plan called for the occupation of the southern third of the island, with the final beachhead line running from Sendai to either Miyazaki or Aburatsu. Estimates concerning opposition to the landings were up to 10 divisions, 800 aircraft, and surviving units of the Imperial Navy.¹⁰

During spring 1945 a number of events paved the way for an invasion. On 1 April U.S. forces landed on Okinawa to begin a long, bloody battle for control of the strategic island. On 3 April, General MacArthur was designated Commander-in-Chief, U.S. Army Forces, Pacific, giving him operational control of all Army ground and air forces in the Western Pacific and placing him on the same level as Fleet Admiral Chester W. Nimitz, Commander-in-Chief, Pacific Fleet. This move rescued MacArthur from professional oblivion, since he feared that the Navy would oversee the invasion of Japan, while he was relegated to mopping-up operations in the East Indies and the Philippines.¹¹ On 25 May the JCS issued a directive to MacArthur, Nimitz, and General Henry H. Arnold, head of the Army Air Forces and Commanding General, 20th Air Force (the Marianas-based B-29s) to begin planning and preparations for the invasion of southern Kyushu. Code-named Olympic, it was scheduled for 1 November and designed to “increase the blockade and bombardment of Japan, engage and destroy major enemy forces, and to obtain positions from which to support further attacks against the industrial heartland of Japan.”¹²

The JCS could not agree on the appointment of an overall commander for Olympic. The Army pressed “the battle in Kyushu is primary” and that naval operations be “in conformity with . . . MacArthur’s plan for the campaign on land.” In essence, this would place the Navy under MacArthur’s command, something to which the admirals vehemently objected. In response, the Navy emphasized the complex interrelationship between the ground and naval aspects of the operation, stating that, while naval operations “must take account of and support the plans for the land campaign, . . . the land campaign adopted must be one that fits in with the practical aspects and possibilities of the amphibious operations.”¹³ In the end, a split command was adopted: MacArthur was responsible for ground operations, Nimitz for naval operations.

The Olympic directive did not irrevocably commit the United States to an invasion, for President Harry S Truman had final approval authority. The JCS, together with Secretary of War Henry L. Stimson and Secretary of the Navy James V. Forrestal, met with Truman on 18 June to discuss the plans. The JCS were unanimous in seeing the need for an invasion, since it seemed to be the best way to bring about unconditional surrender. Much of the discussion centered around the issue of casualties. King and Army Chief of Staff General George C. Marshall estimated that losses would be from 31,000 to 41,000 men out of a projected invasion force of 766,700. Leahy, however, considered this figure optimistic and estimated losses in the area of 35%. By whichever estimate, it would have been a costly campaign. Truman agreed with the JCS and instructed them to continue preparations for Olympic. But he withheld approval for further operations, pending its result.¹⁴

When the war ended, planning and preparation for Olympic were well advanced. Basically, the plan was a modification of the earlier JWPC 235/3. The invasion called for landings at three locations on Kyushu and on a number of offshore islands by General Walter E. Krueger’s Sixth Army, consisting of nine Army and three Marine divisions. Admiral Richmond K. Turner, veteran of amphibious assaults from Guadalcanal to Okinawa, would command the amphibious assault, supported by Admiral Raymond A. Spruance’s Fifth Fleet. The invasion force totaled 551,580 men, considerably fewer than the 766,700 cited at the June meeting.¹⁵ Heavier than expected casualties on Okinawa and in the Philippines and the difficulty in obtaining sufficient replacements made it impossible to bring the assault forces up to full strength.

The Japanese considered Kyushu to be the most likely invasion area and increased their defensive preparations there. According to Joint Intelligence Committee estimates, by late July 545,000 enemy troops were in Kyushu, most of them deployed near the invasion area. The Japanese had also deployed nearly 9,000 kamikaze aircraft within range of the invasion area, as well as large numbers of suicide boats, submarines, and manned torpedoes.¹⁶

Because of the Japanese buildup in Kyushu, the Joint War Plans Committee considered less-costly alternatives to Olympic. One plan, JWPC 398/1, called for the seizure of northern Honshu by a force of 12 divisions around 15 October.¹⁷ The area boasted a good anchorage and a number of airfields and airfield sites, was lightly defended, and was just 400 miles from Tokyo. Southern Kyushu was 600 miles away. Timing for the operation was crucial, for only 55 days were available to get the airfields operational before the onset of winter halted construction. In addition to establishing a lodgement in Japan, this plan might have also misled the Japanese regarding U.S. intentions. They might have attempted to change their defensive dispositions, possibly creating such confusion and disorder as to make subsequent landings easier.¹⁸ The plan’s greatest disadvantage was that it did not engage and destroy major enemy forces, and that required a major reorientation of effort on short notice, which would have tested the versatility and adaptability of U.S. forces.

Because it was the political and economic center of the empire, Tokyo was the primary target of the invasion. An early plan for the seizure of Tokyo, JWPC 263/4, envisioned two possible scenarios: an assault within three months after the invasion of Kyushu or, if Kyushu was bypassed, in January 1946.¹⁹ The first scenario required the use of troops redeployed from Europe, while the second would use only troops already in the Pacific theater. The objective of the operation (later code-named Coronet) was to inflict a decisive defeat on the Japanese Army in the heart of the empire and thereby force an unconditional surrender. If the Japanese did not surrender, then it would have given the United States a position from which to launch further attacks against Japanese forces elsewhere in the Home Islands.

Coronet called for the employment of 18 infantry and 5 armored divisions. The main landings would take place at Sagami Day and would seize the vital Yokohama-Yokosuka port complex, while secondary landings would hit at Choshi and Katakai on the Pacific coast. The Joint War Plans Committee estimated initial enemy opposition at nine divisions (including an armored division) totalling some 300,000 men, with an additional 27 to 35 divisions available as reinforcement. These would be committed piecemeal until expended. Since the Japanese probably would have exhausted the bulk of their air power against Olympic, air opposition was estimated at some 1,300 aircraft. Naval opposition was expected to be negligible.²⁰

One of the reasons for invading Kyushu was to gain airfields for support of Coronet. The Olympic lodgement was approximately 600 miles from Tokyo, and one question was whether or not fighters based there would have been able to support the invasion effectively. In June 1945, JWPC 359/1 examined the possibility of seizing areas closer to Tokyo in order to provide better air support for the invasion. Eight sites were considered: the Sendai Plain and Hamamatsu Lowlands of Honshu, the Tokushima and Kochi plains of Shikoku, and four islands in the Izu Group, south of Tokyo. With the exception of the two on Shikoku, all of the sites were within 300 miles of Tokyo. The plan concluded, however, that assaults against any of the sites probably would involve heavy losses and thereby delay the launching of Coronet. Further, proper coordination between Kyushu-based and carrier-based fighters would be able to furnish adequate air support.²¹

Planning and preparation for Coronet were in the early stages when the war ended. For the most part, the overall assault plan was based on JWPC 263/4. The assault, tentatively scheduled for 1 March 1946, would have employed two armies. The Eighth, commanded by General Robert L. Eichelberger and consisting of six Army divisions, would make the landings at Sagami Bay. The First Army, under command of a European veteran, General Courtney H. Hodges, and consisting of three Army and three Marine divisions, would land near Choshi and Katakai.²² Many of the follow-up forces would have consisted of troops redeployed from Europe, among them British units. London offered to furnish three to five divisions, which would serve under U.S. command. The British Pacific Fleet, under command of Admiral Sir Bruce Fraser, was slated to participate in both Olympic and Coronet.²³

Although Coronet was scheduled for four months after Olympic, the possibility existed that the timetable could not be met. Olympic might have become a protracted campaign and required additional troops, which may have been drawn from units designated for Coronet. Coronet also called for the employment of troops transferred from Europe, and any delays in the redeployment schedule would have also affected the operation. In order to maintain pressure on the Japanese, the Joint War Plans Committee proposed the seizure of northwestern Kyushu as an alternative to Coronet. Their plan, JWPC 363/1, called for an assault by 11 infantry divisions, an armored division, and an airborne regimental combat team near Fukuoka around 1 March 1946.²⁴ This operation would result in the seizure of vital defense industries, the opening of a passage to the Sea of Japan, the complete occupation of Kyushu, and, together with Olympic, in the destruction of one-third of the enemy troops in Japan.

While U.S. planners hoped that the Japanese would surrender after Coronet (or better still, after Olympic), they did not discount the possibility that the enemy might still hold significant reserves elsewhere in Japan and refuse to capitulate. In case of such a contingency, the committee drafted a plan, JWPC 333/1, which called for assaults against remaining Japanese pockets in the Home Islands in order to destroy major enemy forces and secure key areas.²⁵ The plan listed five potential targets: the Shimonoseki Strait area of Kyushu, the Osaka-Kobe-Kyoto area of central Honshu, the Nagoya area of Honshu, the Hakodate-Aomori-Ominato area of Honshu and Hokkaido, and the Sapporo Plain of Hokkaido. The committee estimated that the Osaka-Kobe-Kyoto, Nagoya, and Hakodate- Aomori-Ominato operations each would require eight divisions, Shimonoseki Strait nine, and the Sapporo Plain five. If possible, these operations would employ troops already in the Western Pacific in order to simplify logistics and avoid the use of the strategic reserve based in the United States. The first attack was tentatively scheduled for 1 July 1946, with the target dependent upon the strategic situation.²⁶

Of the operations planned, Olympic would have been by far the most difficult. Ideally, an attacking force should have at least a three-to-one numerical superiority over the defenders. But in the case of Olympic, the U.S. and Japanese ground forces were of almost equal strength. The greatest advantage the United States had over the Japanese was sheer weight of fire power, but this might not have been employed to the best effect, for a number of reasons.

For instance, pre-invasion air strikes against targets in Kyushu, western Honshu, and Shikoku were the responsibility of the Ryukyus-based Far East Air Force. In mid-October, however, a typhoon devastated the Ryukyus such that the bases there were virtually unusable. Thus, the JCS would have had to decide whether or not to go through with the invasion. If they invaded, the burden of air support would have fallen on the Navy. Naval air power would have had three primary missions: fleet defense, deep- strike interdiction, and close-air support for the invasion troops.

The Navy probably could have handled any one (or two) of these missions easily, but to handle all three simultaneously would have been beyond its capabilities. Something would have had to give, and in light of the experience with the kamikazes at Okinawa, the Navy probably would have concentrated on that threat, at the expense of supporting troops ashore, a policy that would have carried seeds of confrontation between the Army and Navy. The Army would have blamed slow progress ashore on the Navy’s failure to provide adequate close-air support, while the Navy might have criticized the Army for endangering the fleet because of its slowness in getting land-based air power operational. Unless such accusations could have been rectified quickly, interservice relations—never ideal under the best conditions—would have been tested severely.

The Soviet Union would have played a significant, if indirect, role in ensuring the success of the invasion. The JCS were concerned that, in the event of an invasion, Japanese leadership might move to the mainland and form a government-in-exile, or that Japanese troops stationed on the mainland might ignore a surrender order. In either case, the war would continue. But who would destroy the Japanese forces on the mainland? Chinese Nationalists were not up to the task, and while the Chinese Communists might have been able to defeat the Japanese, U.S. support of them was politically impossible. That left the United States and the Soviet Union. A U.S. invasion of the mainland probably would have involved becoming embroiled in the Chinese civil war, which few wanted but which would be extremely difficult to avoid. Soviet intervention in the war against Japan, while not ideal, seemed to present fewer problems.

The dropping of the atomic bombs ended the war and the need for a costly invasion. After the war, many people, among them Admirals Leahy and King, maintained that an invasion would have been unnecessary, for the bombardment and blockade of Japan had already defeated Japan. Such people spoke in hindsight, however. During the war, few thought Japan could be defeated without an invasion. Fortunately, for both the United States and Japan, Olympic, Coronet, and the other planned invasions never materialized. If they had, the world would be vastly different today.