In "Half a Million Purple Hearts" (December 2000 / January 2001), the authors, D. M. Giangreco and Kathryn Moore, in effect refute their own thesis. They say the fact that the United States ordered 9,000 new Purple Hearts during the bombing of Kosovo "had nothing to do with imminent combat," that the medals "were ordered for the simplest of bureaucratic reasons": The United States "had to replenish its own inventory." I would simply emphasize that a half million Purple Hearts on hand when World War II ended does not prove that the military or President Truman believed that a half a million American lives would have been lost in an invasion of Japan. The military is notorious for ordering far more equipment than it needs.
Historians have demonstrated that the largest pre-Hiroshima estimates of casualties by military planners for the invasion of Japan (southern Kyushu, November 1945; northwestern Kyushu, highly unlikely; and Honshu, March 1946, probably unlikely) were 46,000 dead, 170,000 wounded, and 4,000 missing. Moreover, at a June 18, 1945, meeting at the White House, General Marshall informed President Truman, "There is reason to believe that the first 30 days in Kyushu should not exceed the price we paid in Luzon [31,000 casualties]."
Merrel Clubb
Missoula, Montana
The authors reply:
Mr. Clubb’s point that “the military is notorious for ordering far more equipment than it needs” is well taken, and everyone is familiar with the tales of waste and inefficiency. Documents held by the U.S. Army and the National Archives, however, present a somewhat different picture for events surrounding the manufacture of Purple Hearts during World War II. Unlike bullets or tent pegs, the Purple Heart was a high-ticket item, requiring the use of restricted wartime commodities, such as precious metals, and scarce, highly specialized workers. Increases m production could not be arranged in a cavalier manner; indeed, the large Purple Heart orders by the Navy in the spring of 1945 could not be met in time for the invasion of Japan, which was why the Navy had to borrow medals from the Army.
The U.S. Mint’s production was neither capricious nor excessive, and the casualty estimates that drove it were based on ratios constructed from the increasingly heavy losses during the recent fighting. Truman’s Secretary of War, Henry L. Stimson, regularly used the rounded figure of 500,000 during briefings for casualties expected in Japan, and the Army planned to supply replacements for 720,000 “dead and evacuated wounded” during the campaign, a figure that does not include Navy and Marine casualties. To meet these demands, monthly Selective Service draft calls in 1945 were raised from 60,000 to 100,000 to feed a training infrastructure that peaked at 400,000 men in June 1945, long after the last shots had been fired in Europe.
The documents cited by Mr. Clubb are frequently portrayed by critics of Truman as proof that the President invented large casualty estimates after the war in an effort to justify the use of nuclear weapons. The figures come from briefing papers prepared for the President ahead of his June 18,1945, meeting with the Joint Chiefs of Staff and the service secretaries. What the critics uniformly overlook, however, is that the first set of figures (totaling some 220,000 battle casualties) were not presented to Truman because they were found to be flawed, and the 31,000 represents just the Army and only the first 30 days of a series of campaigns then expected to last until nearly 1947. The briefing paper statement that “the first 30 days in Kyushu should not exceed the price we paid in Luzon” refers to the cost in terms of a stated ratio of one American casualty for every five of the Japanese. The figure of 31,000 for the earlier, and much smaller, operation in the Philippines was presented in a chart only as a baseline number to establish a ratio that might be applicable to the fighting ahead.
The misrepresentation of these documents is widespread today, but presenting such numbers as all-encompassing final estimates for the much larger and longer operations in Japan necessitates a spurious methodology. In effect, it is as if someone during World War II had come across a casualty estimate for the invasion of Sicily and then declared that the number from this opening operation would represent the total casualties for the entire Italian campaign—and then went on to announce with complete confidence that the original number actually represented likely casualties for the balance of the war with Germany. Of course, back then, such a notion would have been dismissed as absurd; today, speculative scholars doing much the same thing win the plaudits of their peers.