Looking Back: Gen. Marshall and the Atomic Bombing of Japanese Cities

November 2015

By Barton J. Bernstein

The U.S. atomic bombings of Hiroshima with a uranium weapon and Nagasaki with a plutonium weapon in August 1945 caused the deaths of more than 100,000 Japanese, mostly noncombatants, by the end of that year and injured about the same number. Those bombings also killed thousands of Korean noncombatants and some Allied prisoners of war.

Much has been written over the years about the bombings. That literature includes considerable discussion of the process that led to President Harry Truman’s decision to use the two bombs and to target enemy cities, the resulting killing of massive numbers of noncombatants, and the debate over the necessity and ethics of conducting two bombings rather than one or even none.

The still-controversial history of the bombings involves disputes in areas such as sources, the uses of evidence, the standards for assessing and interpreting decision-making, and the basis for defining and evaluating morality and ethics in the context of war.

A key source in this contested history has been the evidence of individuals who were involved in the U.S. government’s development of and decision-making on nuclear weapons. Yet, many of these individuals, be they supporters or critics, somewhat rewrote their history in the aftermath of the August 1945 bombings.

Untangling postwar contentions from actual pre-Hiroshima actions is generally not a simple matter. It requires close attention to pre-Hiroshima archival sources, which account for well more than 100,000 pages in multiple libraries. In using these sources, the analyst should give much greater weight to the pre-Hiroshima sources than to the later, postwar claims in cases of significant discrepancies.

After the war, in memoirs and in related statements, three former wartime members of the U.S. Joint Chiefs of Staff—Admiral William Leahy, the de facto chairman of the Joint Chiefs; General Henry Arnold, head of the Army Air Forces, and Admiral Ernest King, head of the Navy—publicly raised sharp questions about the military-political necessity of those atomic bombings and indicated there were other ways of ending the war. In his own widely reviewed memoir I Was There,¹ Leahy passionately condemned the bombings as unethical and even barbarous. Yet, the available records, including his own diary, give no indication that he expressed this opinion to Truman or to any other government associate before the bombings.

Marshall as the Exception

Of the four retired wartime military chiefs, only General George Marshall, the wartime Army chief of staff, never in public or in any documented private statements in the postwar period joined in such criticisms or even in hints of having ever entertained such doubts about the 1945 use of nuclear weapons. In various interviews and related statements, Marshall strongly defended the atomic bombings of Japanese cities as militarily and politically necessary and as ethically justifiable.

Nevertheless, Marshall was the only one of those four top-level men who, before the atomic bombings, actively sought to avoid the use of the new weapons on Japanese cities or to provide an adequate warning to Japan so that noncombatants could safely flee the cities. Fiercely loyal to Truman, unwilling to write a memoir, and wary of political controversy, Marshall in the aftermath of the war never revealed or even hinted that he had been a partial dissenter before the bombings, nor did other high-level memoirists ever disclose such information about Marshall’s pre-Hiroshima thinking.

On May 29, 1945, however, about seven weeks before the Trinity test on July 16 in New Mexico and about 10 weeks before the August 6 bombing of Hiroshima, the 64-year-old Marshall argued generally for not using the weapon on Japanese cities but on a “straight military [target] such as a naval installation,” in the words of the meeting minutes.² If a city was chosen, it should be a “large manufacturing” area, he urged, and the United States should at least provide a substantial warning about the weapon in order to spare noncombatant lives.

For him, the definition of noncombatants included workers, as well as their families. That conception apparently did not change for Marshall in his thinking about the manufacturing that might contribute directly to Japan’s war effort.

Marshall presented his ethically influenced analysis in a secret meeting with Secretary of War Henry Stimson and Assistant Secretary of War John McCloy, both Republicans in a Democratic administration.

Stimson was serving in his fourth presidential cabinet; his previous posts included secretary of war under President William Taft and then secretary of state under President Herbert Hoover. As part of what was designed as a bipartisan national security system, Stimson had rejoined the government as President Franklin Roosevelt’s war secretary about 18 months before the Japanese bombing of Pearl Harbor in December 1941.

By late May 1945, the 77-year-old Stimson and the 50-year-old McCloy had been working comfortably with Marshall for about five years. It was a close relationship of mutual trust and sincere respect. Their policy differences, when they occurred, were hammered out in private. During the war years, the three men easily maintained such discretion and energetically practiced secrecy. Such behavior flowed from their temperament and the context.

Marshall, McCloy, and Stimson sought to avoid or minimize open controversy about how to conduct the war, including the uses of weaponry. These men did not want fractious publicity about policy decisions, and they sought to limit considerations of the politics of war to high-level, often secret decision-making. That was part of the intended political strategy of creating and maintaining national consensus on many war-related matters.

The three men were generally aided in such arrangements by the Senate’s special investigating committee. Initiated early in the war under then-Senator Truman (D-Mo.), the panel had helped insulate the government’s policymaking on defense issues from public scrutiny. When Truman, elected vice president in 1944, became president in April 1945, that Senate committee, maintaining its ways from the Roosevelt period, continued to help protect the chief executive and his bipartisan national security system from intrusive scrutiny on the conduct of the war.

On the subject of the atomic bomb, only a few congressional leaders drawn from each political party even knew of the top-secret project. The funding, about $1.8 billion by mid-1945, was tucked into appropriations bills without the bulk of senators or representatives knowing about it or about the Manhattan Project itself and the weapons it was designed to build.

The May 29 Meeting

At the May 29 meeting, much of the discussion among Marshall, McCloy, and Stimson focused on military policies to help end the Asian war and the use of significant weapons—atomic bombs and toxic gas—against Japan in future months. Breaking ranks with the other two officials, Marshall hoped to use gas, thereby violating Roosevelt’s public pledge of no first use of a “poisonous or noxious” gas,³ but McCloy and Stimson showed no interest in endorsing such a radical departure from established policy. By their response, they seemed uneasy about such a proposal.

With a nuclear bomb likely to be ready and to be used in about early August 1945, the three men talked about that weapon. Still untested at the time of the meeting, it was expected to be hundreds of times more powerful than even the largest conventional bombs. The uranium bomb was predicted, as of late April 1945, to produce the equivalent of about 8,000 to 20,000 tons of TNT power, and the plutonium weapon about 4,000 to 6,000 tons.⁴ Regardless of which kind of atomic weapon was used, it clearly would cause massive destruction on any target area by blast and heat, as all three men easily recognized.

At the time of the meeting, none of the three men apparently knew anything of significance, and quite likely they understood nothing at all about nuclear radiation and its great perils for humans: death or agonizing injury and even impairments in the gene pool for future generations. None of the three men actually knew anything in depth about the science and technology of the new uranium and plutonium bombs.

The issue that troubled Marshall, as he made clear to Stimson and McCloy at their meeting, was that these new, massively powerful bombs, if used on a city, would kill many noncombatants. To Marshall, that type of killing apparently constituted a dangerous and dramatic crossing of an important moral and ethical threshold by slaying innocent men, women, and children.

In public statements in the 1930s, Roosevelt had emphasized the moral and ethical line that would be crossed by bombing noncombatants. For Marshall, who had entered the Army more than a decade before World War I, and for Stimson and McCloy, such a line had at one time seemed deeply embedded in the informal U.S. code of conducting warfare.

By late May 1945, however, Marshall and his two colleagues at the meeting knew that U.S. conduct of so-called conventional warfare, especially with B-17 bombers and then the new B-29 bombers, had crossed that threshold with bombs, including incendiaries, against German and Japanese cities. U.S. and UK aircraft massively fire-bombed the German city of Dresden in February 1945, killing about 35,000 to 45,000. The U.S. fire-bombing of Tokyo in March killed about 80,000. The Tokyo count, with about 16 square miles burned out, was the highest by any bombing raid through late May 1945. In both enemy cities, the dead and injured, as all three U.S. officials knew, had been mostly noncombatants.

Whether and how such bombings would speed the ending of the wars in Europe and Asia was often rather vague. Among many U.S. government officials, there was a general sense that such bombings would help erode morale in the two enemy states and in multiple ways, often loosely defined, contribute to pushing the two enemy governments toward surrender.

Those attacks had greatly upset Stimson. He had worried on ethical grounds about the killing of noncombatants in such attacks, and he wished that such killing could be avoided or minimized by “precision” bombing.⁵ Before, during, and after the May 29 meeting, Marshall apparently never expressed, even in secret, governmental inner circles, such worries or doubts about massive conventional bombing of enemy cities.

As the head of the Army, Marshall had the moral authority, had he wished to use it, to complain inside the government and to urge different bombing policies. He never had done so or even come close to doing so.

Yet, even in prospect, the atomic bomb in May 1945 seemed markedly different to him, raising unsettling ethical and moral issues. Apparently troubled by the thought of targeting a city with that weapon, Marshall feared that moral opprobrium might be heaped on the United States in the aftermath. Thus, he offered a proposal for alternatives—bombing a military installation or bombing a large manufacturing area with a prior warning—by which he hoped to protect the United States and its new president from moral and ethical accusations of acting improperly.

A Likely Missed Opportunity

Marshall’s pleas on May 29 did not seem to gain support from Stimson or McCloy. Two days later, Stimson helped shape the nuclear targeting policy by outlining with a secret blue-ribbon panel, the Interim Committee, that the weapon should be used without any warning to Japan and on a military-industrial target surrounded by workers’ houses.⁶

As Stimson and the other meeting participants, including Marshall, well understood, enemy workers were usually defined as noncombatants. Furthermore, even if there might be some ambiguity about categorizing workers, there was none regarding their children, nonworking wives, and elderly relatives. Thus, many thousands of enemy innocents could be expected to be slain or injured by the atomic bomb.

Having been quietly defeated in the small meeting on May 29, Marshall raised no objections about that targeting and no-warning policy on May 31 in the larger meeting, which included James Conant, president of Harvard University, and Massachusetts Institute of Technology president Karl Compton. Marshall also did not do so in the next 10 weeks in various discussions with Stimson, McCloy, or Truman as the United States moved to bomb Hiroshima and Nagasaki.

Marshall quite likely missed an opportunity to establish an alliance with the president on redefining the targeting. Marshall did not know that Truman, around the time in late July 1945 that he endorsed the military order to use nuclear weapons on Japanese cities, had chosen to believe that the bombs’ targets would be truly of a military nature. In what was most likely self-deception, Truman wrote in his diary on July 25 that he had instructed Stimson that the new bomb should be used “so that military objectives and soldiers and sailors are the target and not women and children.” The target, Truman falsely contended, would be “purely military.”

Truman wrote of the Japanese being “savages,” calling them “merciless and fanatic.” The United States, he claimed, was markedly different in its conduct of the war and its concern for noncombatants.

Marshall apparently never said to anyone that he had regretted not seeking before the Hiroshima bombing to make common cause with Truman on the targeting policy. Had Marshall risked going outside the chain of command and bypassing Stimson, the general might well have been successful in July 1945 and thereby have helped to redefine the targeting policy for the atomic bomb to avoid killing and injuring noncombatants. Whether Marshall ever regretted not approaching Truman on this matter is uncertain.

Marshall’s Reshaped History

After the war, Marshall apparently never disclosed, even to his official biographer, historian Forrest Pogue, that, at the May 29 meeting with Stimson and McCloy, he had sought to alter the course of nuclear history as it was then unfolding.

McCloy’s minutes of the May 29 meeting, the only available record of Marshall’s secret counsel there, were not declassified until about two decades after the general’s October 1959 death. Dismayingly, that documentary record has generally been of little or no interest to most historians studying nuclear weapons or Marshall.

While concealing his May 1945 effort, Marshall himself, in an interview with Pogue in February 1957, somewhat rewrote the pre-Hiroshima history on the atomic bomb. Marshall contended incorrectly that the Truman administration, before the first atomic bombing of Japan, had provided the Japanese with a vague warning of the planned use by the United States of that new weapon on Japan.

Although the United States had issued broad warnings about large-scale destruction, it never gave Japan an explicit warning about the atomic bomb in the run-up to the Hiroshima bombing. One wonders—admittedly, in the absence of evidence—whether Marshall’s flawed recollection, as captured by Pogue, was an expression of the general’s unacknowledged personal regret. It might also have been an effort to protect his president, government, and country.

At the May 29 meeting, Marshall had unsuccessfully sought to reshape the future, and in 1957, he sought to reshape the past. He largely failed in those two efforts. By remaining silent about his advice of May 29, he helped obscure the possibility of a different history of nuclear weapons.

Stimson never mentioned Marshall’s May 29 advice in his memoir and in an article that drew on it. McCloy did not either in his many postwar commentaries, even though they often were critical of the atomic bombings. Truman, in his ghost-written memoir and elsewhere, never mentioned and probably never knew of Marshall’s urgings on the targeting of nuclear weapons.

Different Targeting?

If the United States had issued a substantial, explicit warning to Japan before using the atomic bomb on Hiroshima; if the U.S. attacks truly had targeted manufacturing areas; or, more significantly, if the targets in Japan had been actually a military installation, the postwar dialogue about the bombings might have been different.

To consider the plausible alternative history in which the United States chose a military target and therefore avoided the deaths of massive numbers of noncombatants is not to seek to justify the atomic bombings. It is intended to force a rethinking of important aspects of the August 1945 bombings and of the moral and ethical underpinnings of those bombings.

Whatever their use, the bombs dropped in August 1945 would have crossed a profoundly important and greatly troubling threshold. They represented exceptional power delivered by a new technology. If they had been used against an enemy military installation with comparatively few deaths and injuries to enemy noncombatants, the important moral and ethical threshold of killing massive numbers of noncombatants, despite the earlier attacks on Dresden and Tokyo and the similar conventional bombings elsewhere, would not have been crossed in the atomic bombings of Japan.

Part of the older moral and ethical code would have been respected in what nevertheless was a brutal, savage war. Fought broadly across three continents and numerous islands, that long war killed many millions of noncombatants among the total of about 55 million people.

By most estimates, noncombatant deaths, including those in German death camps and those from starvation, malnutrition, and disease elsewhere, significantly outnumbered the ones among soldiers from bombing and ground warfare.

The two atomic bombs probably added at least 100,000 to the tally of noncombatants but only about 12,000 to the number of military men killed.

Analysts might contend that limiting the use of the new, devastating weapon to military targets might have made it seem less terrible and thus less effective as a psychological weapon. In a similar vein, the analysts might argue that the large number of deaths and the high percentage that were noncombatants created its own firewall in making future use by any country morally more difficult.

To discuss such matters involving the dead and who was killed or might otherwise have been killed under a different targeting plan may seem ghoulish and cold blooded. In fact, these matters, which involve ethical considerations, are important in critically examining how the atomic bomb has been understood and why.

Marshall’s proposals add an important element to the debate over these matters. His proposals make clear that policies on atomic bomb targeting that were different from the ones actually used were not simply hypothetical possibilities but truly plausible alternatives in 1945. For arms control analysts and others who wish to think critically about such alternatives, the generally neglected document on Marshall’s meeting of May 29, 1945, provides firm evidence that an important member of the U.S. government recommended seeking to use the atomic bomb in a way that would not kill massive numbers of noncombatants.

---

---

ENDNOTES

1.  William D. Leahy, I Was There: The Personal Story of the Chief of Staff of Presidents Roosevelt and Truman Based on His Notes and Diaries Made at the Time (New York: Whittlesey House, 1950), pp. 440-441.

2.  John J. McCloy, “Objectives Toward Japan and Methods of Concluding War With Minimum Casualties,” May 29, 1945, in Marshall Papers, George C. Marshall Library, Lexington, VA (memorandum of conversation with General George Marshall with handwritten note by McCloy or Marshall stating “C/S [Marshall] has noted and has no further suggestions”).

3.  Statement of President Franklin Roosevelt on June 8, 1943, in The Public Papers and Addresses of Franklin D. Roosevelt, Vol. 1943, ed. Samuel Rosenman (New York: Harper, 1950), p. 242.

4.  For the formal estimates in the spring of 1945 of TNT equivalence for the two types of atomic bombs, see Gen. Leslie Groves to Secretary of War, “Atomic Fission Bombs,” April 23, 1945, pp. 5-6, in Top Secret of Special Interest to Groves Files, #25, Manhattan Engineer District (MED) Records, RG 77, U.S. National Archives.

5.  Henry Stimson diary, June 1 and 6, 1945, Sterling Library, Yale University, New Haven, CT. The subject has also been thoughtfully treated by various interpreters. See McGeorge Bundy, Danger and Survival: Choices About the Bomb in the First Fifty Years (New York: Random House, 1988); Sean Malloy, Atomic Tragedy: Henry L. Stimson and the Decision to Use the Bomb Against Japan (Ithaca, NY: Cornell University Press, 2008).

6.  For the Interim Committee minutes of May 31, 1945, see Harrison-Bundy Files, #100, MED Records, RG 77. For President Harry Truman’s Potsdam diary of July 25, 1945, see Truman Papers, Harry S. Truman Library, Independence, MO. For Marshall’s interview with Pogue of February 11, 1957, see Larry and Joellen Bland, eds., George C. Marshall Interviews and Reminiscences for Forrest C. Pogue, rev. ed. (Lexington, VA: George C. Marshall Research Foundation, 1991), pp. 422-425.