Methodology for Estimating
Casualties or KIA

August 2008



[Disclosure: I have never been in the military myself.  Nearly all the information
below has been gleaned from things I’ve read in connection with arguments
over estimates of American casualties prevented, or lives saved, by President
Harry S. Truman’s decision to use the atomic bomb against
Japan.  None of
what appears here is original.  -- G. J. Desnoyers]


In relation to Truman’s decision to use the atomic bomb against Japan, some people like to speak of numbers of casualties prevented by using the a-bomb.  Other people prefer to speak of the “lives saved” by Truman’s decision.  The numbers refer to different things.  “Lives saved” is a smaller number than “casualties.”  In wartime, “casualties” are the sum of: the killed in action (KIA), the wounded (especially to the point of being removed from battle), and the missing in action (MIA).


Many arguments occur over estimates of American casualties, or lost lives, which would likely have resulted if the a-bomb had not been used against Japan, and the United States had had to invade Japan’s home islands to achieve the unconditional surrender U.S. leaders were demanding.


Certain questions logically arise in relation to these arguments.  How is it possible to estimate casualties from an anticipated battle or proposed course of military action?  How can people estimate the number of lives saved by certain war strategies, for example, by using the atomic bomb to make an invasion of Japan unnecessary?  How can those numbers be of any value when there are so many variables in a war?

Reasonable and useful estimates of casualties are possible

Those are good questions!  However, reasonable estimates of casualties are not impossible when there is a large war going on like that in the Pacific Theater of WW II.  Many battles provide for much experience and data collection.  Statistics are collected for battles of every type.  In making estimates, no-one is expecting, or trying for, precise figures.  Estimates are only the best guess, but they are not worthless.  A military that did not try to make estimates of casualties before deciding on courses of military action could be deemed negligent.

Methodology simplified


What follows is a very simplified outline giving some basic features of methodology for making estimates of casualties.  Similar methodologies could be used to estimate KIA, MIA, troops wounded, troops lost from Battle Stress, and troops lost from disease.  All these types of casualties are important to commanders in the field.

I. Those responsible for estimating casualties before a battle begin by gathering information relative to the obvious factors in play for nearly all battles.  Information is collected both for our side and for the enemy side.  It’s generally easy to get the necessary information relative to our side; it’s harder to get information on the enemy side.  For the enemy side, information might come from aerial surveillance, spies, interrogation of prisoners, intercepted communications, or other sources.


For some specific examples, information will be collected on: (1) the numbers for each type of force (troops, armor pieces, weaponry and vehicles available, etc.), (2) the terrain on which the battle(s) will occur, (3) the requirements for protecting supply routes and communication links, (4) whether the objective is to destroy the enemy force or just to neutralize its ability to contribute the enemy’s goals, (5) the time-frame allowed for victory or achievement of a goal, and (6) the basic nature of the assault(s) to be waged.


II. Then they study statistics and correlation factors (ratios) gathered from previous battles, especially looking at battles similar to the proposed or anticipated one.  They apply correlation factors from previous battle which are thought to be appropriate to the information gathered relative to the proposed or anticipated future battle, interpolating and extrapolating data wherever necessary, and then make their best guess on a number of casualties.


III. They also consider special factors which are relevant for the coming battle but which are not necessarily always in play, e.g., (1) the expectation of an unusually large number of captured prisoners who have to be dealt with, (2) the need to consider a liberated civilian population, (3) the presence of a crucial but exceptionally battle-weary force, (4) favorite and other possible tactics of the enemy force’s commanders, and (5) information obtained from intercepts of enemy communications.

IV. They consult experienced American commanders, sometimes including commanders having responsibility for the coming battle or operation, looking for information from them which might be helpful in adjusting estimates up or down.  While a military operation’s commander may help in adjusting estimates up or down, and a commander anticipates and plans for casualties as a part of overall planning, the time-consuming nature of making casualty estimates precludes the task from being primarily his responsibility.  If he has concerns about an estimate, he has the option of discussing it with those who made it, to better understand their rationale, or with his superiors in the military’s chain of command.



One Million American Lives Saved by Using the A-bomb against Japan?


I’m sure you are now asking, “How do those who say that there would have been a million, or even more, American lives lost in an invasion of Japan justify THEIR numbers?”  They do it in a variety of ways.


For most, they justify their high number by mentioning a source they've taken as authoritative.  Perhaps they learned it from a history teacher back in high school, or a person they know who was in the war.


Others, more knowledgeable, may point to a mysterious memorandum from an "unknown economist."  The memo, written in either late May or early June of 1945, was delivered to President Truman through former president Herbert Hoover.  The unknown author outlined a proposal for bringing the war to a close without requiring an invasion of Japan.  [Multiple plans for ending the war without requiring an invasion of Japan were available to Truman.  He rejected them all except the one to use the a-bomb.]  The memo's author also made the dramatic claim that his approach would save 500,000 to one million American lives.  This point is critical.  Until this memorandum appeared in early June 1945, NO REAL NUMBERS had been discussed formally regarding the human cost of future operations against Japan.  This estimate, cropping up at a propitious moment, captured the fancy of Truman and Secretary of War Stimson.  The memo has been called "the source for the myth of half a million casualties predicted for the invasion of Japan."


For still others, it’s actually funny how they justify their large estimates!  They start out by noting the number of Japanese troops on Okinawa (approximately 200,000 counting soldiers from Okinawa, soldiers from the main Japan islands, and civilians from Okinawa pressed into service) where the Japanese troops were unusually well dug in and possessed the high ground and many well-camouflaged positions.  Then they determine a ratio of American casualties per thousand enemy troops.  They then take that ratio and apply it to ALL OF THE JAPANESE HOME ISLANDS, TAKING THE ENTIRE HEALTHY AND ABLE ADULT POPULATION OF JAPAN (ONE-QUARTER OF THE COUNTRY'S POPULATION OF 72,000,000) TO BE EQUAL IN LETHALITY TO THE TRAINED AND DUG-IN JAPANESE FIGHTERS ON OKINAWA.  Here’s the math: They first multiply the wartime healthy and able adult population of Japan by the number of American casualties (or KIA) on Okinawa.  Then they divide that result by the number of Japanese troops on Okinawa.  If the numbers are still not as high as they want (to justify to themselves the dropping of the two a-bombs) they throw in one or more multipliers (“fudge factors”) above 1.0 to increase the numbers.  Fudge factors used might involve (a) the supposed extreme desperation of the Japanese military and citizenry in defending their homeland, (b) the [alleged] threat from 3000 kamikazes pledged to give up their lives in suicide assaults on American ships involved in an invasion, and (c) the war-weariness affecting troops called on to fight an Okinawa-type battle throughout ALL the Japanese islands from one end of each island to the opposite end.  Consider the following sample math exercise:


Sample Math Exercise for Estimate of KIA

[Showing the kind of figuring performed by many
folk who argue for one million American lives saved by
use of the a-bomb against Hiroshima and Nagasaki]


Okinawa had about 200,000 troops fighting on the Japanese side, including: soldiers from Okinawa, soldiers from the main Japanese islands, and indigenous Okinawa civilians pressed into service.


American KIA in the Battle of Okinawa numbered approximately 12,520.


The population of the main Japanese islands, excluding Okinawa (already under U.S. occupation), was 72,000,000.  Assume that one-quarter (25%) of the 72,000,000 are of age, health, and ability to serve in the military defense of Japan against an American invasion force.  One-quarter of 72,000,000 equals 18,000,000.


(18,000,000  x  12,520) / 200000 = 1,126,800 American KIA


This estimate of 1,126,800 American KIA assumes: (1) that the U.S. would have to invade ALL of the Japanese home islands, (2) that battles would have to be fought from one end of each island to the opposite end, (3) that the lethality of the primarily untrained civilian Japanese fighters to Americans would be equal to the lethality of the mostly military-trained Japanese force on Okinawa, and (4) that the battles for the Japanese islands would, from start to finish, even when at the point where continued fighting by the Japanese would be suicidal, be as ferocious as the fighting over Okinawa.



The facts


The JWPC's estimate, the June 18 White House meeting, and the Leahy estimate


The responsibility of estimating American casualties in an invasion of Japan was given to the Joint [Army & Navy] War Plans Committee (JWPC).  The work of the JWPC required that it continually be given updated information on the condition and placements of Japanese forces.  The JWPC's highest estimate of KIA in an invasion of Japan was 46,000.  That estimate was for a two-front invasion of Kyushu followed by a one-front invasion of Honshu, on which Tokyo was located.  Smaller estimates were made for two other invasion schemes: 40,000 for a single-front invasion of Kyushu followed by an invasion of Honshu, and 25,000 for only a two-front invasion of Kyushu.


The JWPC's estimates had been prepared for a crucial June 18, 1945, meeting at the White House.  Present at the June 18 meeting were nine individuals: President Harry S. Truman, Fleet Admiral William D. Leahy, General of the Army G. C. Marshall, Fleet Admiral E. J. King, Lieut. General I. C. Eaker (Representing General of the Army H. H. Arnold), Secretary of War Henry L. Stimson, Secretary of the Navy James Forrestal, Assistant Secretary of War John J. McCloy, and [as Secretary] Brig. General A. J. McFarland.  According to the meeting's minutes, Truman "stated that he had called the meeting for the purpose of informing himself with respect to the details of the campaign against Japan set out in Admiral Leahy's memorandum to the Joint Chiefs of Staff [JCS] of 14 June."

Admiral Leahy's June 14 memorandum to the JCS had informed them that the President desired to prepare for the forthcoming Potsdam Conference.  Truman sought information on such matters as the numbers of men and ships required for an invasion of Japan, estimated casualties, U.S. intentions for Russia, and contributions expected from America's allies.  Leahy's memorandum to the JCS specified that the decision on the campaign against Japan must economize the loss of American lives.  Time and money in comparison were not important.


The estimates from the JWPC were never presented to the President at the June 18 meeting.  Instead, a table was presented which showed numbers of casualties, and some ratios, from several other invasions, and comments were made about whether or how the numbers for those invasions might be indicative as to what would happen in an invasion of Japan.  No agreement was reached on any numeric estimate of KIA or casualties at the June 18 meeting.  This may have been the only time the subject was discussed at a formal meeting of the President and his top advisors, although it has been alleged that informal discussions took place.  And, since the minutes of the meeting, although good, are not verbatim, it is not easy to know from the minutes the kind of thinking on number of casualties which predominated.  But Fleet Admiral William D. Leahy was at the meeting.  And Leahy gives, both in his diary entry for June 18, and in a biographical book, I Was There, what might be the best indication of the magnitude of casualties those present at the meeting were anticipating from the planned invasion.  In his diary entry for June 18, the same day as the crucial meeting, Admiral Leahy wrote: "It is my opinion at the present time that a surrender of Japan can be arranged with terms that can be accepted by Japan and that will make fully satisfactory provision for America's defense against future trans-Pacific aggression."  He goes on to note that General Marshall believes that an invasion of Kyushu, the southern-most Japanese island, "will not cost us in casualties more than 63,000 of the 190,000 combatant troops estimated as necessary for the operation."


Ongoing controversy


Historians have argued intensely over the Leahy/Marshall estimate of 63,000 maximum casualties.  Some historians think Leahy was far too vague as to what the number represented (other than one-third of the force of 190,000) for the number to be meaningful.  Some have argued that, if it truly came from General Marshall, it must have only been referring to casualties in the first month of fighting as it is thought General Marshall would never have given such a low figure as a "maximum" for such a large operation as Operation Olympic, the invasion of Japan.  Other historians have argued that, as unclear as that number may be, it is the best evidence we have as to the magnitude of casualties expected.  In all the formal discussions the government's leaders held on the subject of casualties from an invasion of Japan, it is almost certain that no estimate of casualties above 100,000 was ever presented to them by an authoritative source and accepted by them.


Between 1981 and 1994, planning and preparations were undertaken to exhibit the Enola Gay airplane, from which the atomic bomb was dropped over Hiroshima, at the Smithsonian's National Air and Space Museum in 1995.  The exhibition was to be in recognition of the 50th anniversary of its use in ending the war with Japan.  A very prolonged and intense argument took place over the issue of what number of American casualties prevented (or lives saved) should appear in the written text accompanying the exhibit.  Parties to the debate were mostly historians, military folk, and veterans' groups.  At the debate's conclusion, the Director of the Museum, Martin Harwit, made a controversial decision that the Leahy figure of 63,000 casualties prevented by use of the bomb should be the one used in the exhibit.  It is thought the controversy surrounding this issue is what caused Harwit to resign as the museum's director in 1995.