Downfall: Ending The War With Japan

This is a revised and expanded version of a post that first appeared on this date in 2012.

On June 18, 1945 at a White House meeting with the Joint Chiefs of Staff, the Secretary of War and the Secretaries of the Army and Navy, President Harry Truman approved plans for the invasion of Japan.  Along with the President the other key participants were General George C Marshall and Admiral Ernest King.



Richard B. Frank's 1999 book, Downfall: The End of the Imperial Japanese Empire, using information that had only become available in the prior decade recasts our understanding of the events of the last few months of WWII and the endgame with Japan culminating in its surrender on August 14, 1945 (the formal ceremony took place on the USS Missouri in Tokyo Bay on September 2).  These sources include Russian archives which became available after the fall of the Soviet Union; the release, after Emperor Hirohito's death in 1989, of his lengthy account (dictated in early 1946) of those months; the completion of the Japanese War History Series and the release of additional American intelligence information, most importantly, of the Magic Diplomatic Summaries.  The Magic materials were a daily summary of intercepted Japanese diplomatic cables produced by U.S. intelligence analysts.  These summaries, distributed to senior American policy makers, provide us with a new window on the information they were receiving about Japanese intentions and the interpretations they placed on that information.

While in recent decades the end of the war has focused on the American decision to use the newly developed atomic bombs, Frank's book covers much broader ground, opening our eyes to a vision of a surprising counterfactual history in which the U.S. may not have invaded Japan even if the bombs had not been dropped and the war had continued beyond mid-August 1945.

What were Truman and the others thinking about as they entered the meeting room on June 18?

The night before, Truman had written in his diary that the decision whether to "invade Japan [or] bomb and blockade" would be his "hardest decision to date".

The men entering the meeting knew the American public was increasingly war-weary and shocked by the enormous casualties of the past year.  In the first 30 months of WWII, the U.S. suffered 91,000 battle deaths, an average of about 3,000 a month.  With the D-Day landings in France and the American assault on Saipan in the Pacific in June 1944, the toll accelerated.  During the next twelve months 196,000 Americans died in combat, an average of more than 16,000 a month(1).  With the end of the European war in May, public pressure to start bringing the troops home was increasing though a poll that month found the U.S. public still preferring unconditional surrender to a negotiated end to the war by a margin of 9 to 1.

In early 1945 the Pacific war grew even more horrendous as we approached Japan.  On the 8-square miles of Iwo Jima over five weeks in February and March 1945, 7,000 Americans died and 17,000 were wounded fighting 21,000 Japanese soldiers; the desperate nature of the fighting captured in the words of General Graves Erskine at the dedication of the 3rd Marine Division Cemetery on Iwo:

"Victory was never in doubt . .  . What was in doubt, in all our minds, was whether there would be any of us left to dedicate our cemetery at the end, or whether the last Marine would die knocking out the last Japanese gun and gunner."

Iwo was followed on April 1 by the American landing on Okinawa.  In the next ten weeks another 50,000 American soldiers and sailors were killed or wounded in the course of eliminating a Japanese garrison of 92,000 in a struggle that came to resemble the trench warfare of WWI, the grinding and unrelenting nature of which had also resulted in thousands of additional psychiatric casualties.For a better idea of what the awful fighting conditions were like read one Marines account, With The Old Breed: From Peleliu To Okinawa by E.B. Sledge which contains an unforgettable account of combat on the hillsides under continuous shelling amidst the mud and broken bodies. Along with these campaigns significant fighting continued in the Philippines, at sea and in smaller operations on islands across the Pacific as well as by our British, Australian and New Zealand allies engaged across the Pacific and in Burma.

Along with the weariness, the increasing toll from these battles enraged American civilians and soldiers.  Many accounts by American soldier bitterly reflect the senselessness of what the Japanese army was doing - they had clearly lost the war by this point - why sacrifice themselves and why should they cause more Americans to die in the process?  There had been great anger against Japan since the start of the war, triggered by the surprise attack at Pearl Harbor and increasing reports of atrocities against American prisoners (according to polling data public anger against the Germans was less until the discovery of the Nazi death camps at the end of the war).  Now it was being ratcheted up even further as thousands of Americans died needlessly because the Japanese could not recognize they had lost the war and inducing a high degree of fatalism among the U.S. soldiers who were told that summer they would be part of the invasion force.

They knew that the Allied policy was unconditional surrender for Japan as set by FDR and Churchill at the Casablanca Conference in 1943.

They knew that the Magic Summaries showed no Japanese government disposition for peace on these terms.

They knew that Japan still had 2 million military personnel stationed outside Japan, scattered across Pacific islands, New Guinea, the Dutch East Indies, China, Korea, Burma and Indochina and they wanted to force a formal surrender by the Japanese government to avoid years of piecemeal fighting with each of these isolated forces.

At the June 18 meeting the broad strategy for the invasion of Japan was set out.  The first landings would be on the island of Kyushu, the southernmost of the four main Japanese homeland islands, on November 1.  Kyushu's seizure was necessary so that the Air Force could build the airfields needed for the fighter aircraft to provide air cover for the climactic landing on the Honshu plain near Tokyo planned for March 1, 1946. 

Truman was told that the military planners assumed that about 760,000 American troops would face 350,000 Japanese on Kyushu supported by about 2500-3000 aircraft.  Although the Joint Chiefs unanimously supported this decision, the President was not told that the Navy, unlike the Army, did not believe an invasion would ultimately be needed and that in Admiral King's view he was only supporting preparations for the landing on Kyushu.  King believed a blockade and aerial bombing would bring about surrender.  His Pacific commander, Admiral Nimitz, had recently told King he had changed his mind about supporting the invasion "after further experience in fighting against Japanese forces".

Six weeks later, American intelligence had assembled a completely different picture of what awaited us on Kyushu.  The Japanese Army had figured out that the American landing would be on the island and bet everything on a strategy of inflicting maximum casualties in order to achieve a negotiated settlement to the war, involving preservation of the Emperor, no Allied occupation of Japan and retaining at least some portions of Japan's overseas empire. What had changed in those few weeks?

• Instead of 350,000 troops, American intelligence now estimated there would be 650,000 (and, it was discovered after the end of the war that the Japanese had actually packed 900,000 troops onto the island)

• Instead of 2500-3000 aircraft, the Japanese had between 6,000 and 10,000 and were going to employ many of them in waves of kamikaze attacks against vulnerable transport ships packed with thousands of American troops (the Okinawa kamikaze attacks had been on warships)

• The entire civilian population of the island had been mobilized, armed (in some cases just with hoes and spades) and trained to attack the American soldiers when they came ashore, creating a situation where the U.S. military would be unable to distinguish between soldiers and civilians, resulting in enormous casualties on both sides.

• The Japanese military had issued orders to kill all Allied prisoners of war once the American invasion started.

During these weeks the Magic Diplomatic Summaries indicated no improvement in the prospects of a peace offer from Japan on Allied terms.  An enormous literature on this topic has been created over the past half-century.  For a time in the 1960s and 1970s revisionist historians held the high ground with claims that Truman and company ignored Japanese peace overtures because of concerns about the rising power of the Soviet Union which caused them to find a way to use the atom bomb.  As more documents and information have become available, along with revelations of how some revisionist historians distorted and cherry-picked existing data, the tide of revisionism has receded.  Without rehashing the entire sage, suffice it to say that Japan's foreign minister admitted after the war that the Cabinet never agreed on a specific route for terminating the war and the Magic intercepts revealed a series of communications between the government at home and its ambassadors that were confusing in many respects but always clear in one: unconditional surrender was unacceptable and future events (i.e, casualties inflicted on Americans during the anticipated invasion) might lead to termination of the war on more favorable terms. For those interested in knowing more about the rise and fall of revisionism read this scholarly paper.

According to Franks, the new intelligence would probably have led Admiral King to withdraw his support for the Kyushu landing, precipitating a new strategic review by President Truman in the second half of August, particularly in light of the President's concern over American casualties, if the war had not ended on August 14 after the bombing of Hiroshima (August 6) and Nagasaki (August 9) and the entry of the Soviet Union into the war on August 8.

At the same time, the Air Force had come up with a new approach to strategic bombing that it planned to implement in September 1945.  Unlike the massive incendiary attacks which burned down large parts of Japan's biggest cities between March and June, the new campaign focused on a small number of key rail yards, bridges, tunnels and ferries.  The Air Force had finally realized that with Japan's poor road network (mostly still unpaved) the distribution of food supplies could be paralyzed by disrupting fewer than 100 rail and shipping locations.  With Japan's population already on the brink of starvation, the effect of this campaign would have been catastrophic.  It was already so bad that even with the war ending in August, as late as March 1946 the average daily ration for Tokyo civilians, nominally only 1,042 calories was, in reality, closer to 800 calories and starvation only avoided by massive U.S. food supplies.

This strategic review would have provoked intense controversy within the Administration since the U.S. Army was still committed to the invasion strategy.  There is no indication that Truman ever knew of the new intelligence on the Japanese military buildup on Kyushu or of the new Air Force bombing plan and with the end of the war it was not necessary to raise the issue to the Presidential level.

All of this creates a hypothetical future where no American invasion of Japan occurs even if the war went on beyond mid-August.  The likely results:

• Continued American blockade of the Japan home islands and complete disruption of the food supply by the Air Force bombing campaign inducing famine in the civilian population.

• The invasion of lightly defended Hokkaido (the northernmost home island of Japan) by the Soviets in September 1945 - one of the revelations from the opening of the Soviet archives in the 1990s.

• The British proceeding with their planned amphibious landing in Malaya, scheduled for early September, and incurring heavy casualties against Japanese forces anticipating the operation.

• Continued fighting in the Philippines, on smaller islands across the Pacific and in China.

• Huge death tolls of Asian civilians under Japanese occupation (primarily in China and secondarily in Southeast Asia), estimated to be 100,000 to 250,000 a month from famine, disease, imprisonment and execution.

The question is how long could Japan have survived in this scenario and whether the ending would be an organized surrender of all Japanese military forces or a disorganized collapse in which scattered fighting continued across the Pacific and mainland Asia.

The end of the Pacific war, just as that of the European war, would have been grim under any scenario.  

This post only begins to touch on the issues impacting the end of the war and covered in detail in Downfall.  Frank discusses the Soviet attack on Japan in Manchuria and its impact on the Japanese government, the lead up to, and the impact of, the Hiroshima and Nagasaki attacks, Japanese cabinet deliberations and debates over peace terms, the controversy over American casualty estimates for the invasion (for an excellent summary of the complex history and methodology of the casualty estimates read "A Score of Bloody Okinawas and Iwo Jimas" by DM Giangreco in Hiroshima In History, edited by Robert James Maddox (2007)) and the nuts and bolts of U.S. and Japanese military operational planning.

The book is thought provoking, giving the reader a greater appreciation of the information decision makers had available, the different paths that could have been followed and the consequences that would have flowed from them.  It is particularly valuable in conveying what it is like to have to make decisions affecting the lives of millions with only the information you have available at the time and without the advantage of hindsight.

It strikes THC that these events would be a terrific instructional tool for students and others regarding real-life contingencies and decision-making.  A course where students were assigned roles in the American civilian and military hierarchy and then fed information as it became available and asked to make decisions based on the available information would make for a memorable learning experience and would probably be humbling and sobering to those who think everything looks as clear to the participants at the time as it does to others in hindsight.  It could be done in two parts - the first based upon what we know happened through the decisions to drop the atomic bombs and accept the continued role of Emperor Hirohito and a second based on a scenario where the bombs are not dropped and the war continues.  Most importantly, those participating should be challenged along the way by the instructor(s) but not led to any predetermined outcome.



(1)  196,000 is double the total number of American combat fatalities in the 70 years since 1945 including Korea, Vietnam, The Gulf War, Iraq and Afghanistan.