By the beginning of the 20th Century, the concept of battle had gotten
extremely vague. The idea that armies would bloodlessly maneuver into position
before engaging in a brief, concentrated clash of arms was a thing of the past.
Twentieth Century wars were continuous in both time and space. They were fought
day and night, winter and summer, over weeks and months at a time. Major
battlefronts often spanned entire countries, and even minor battlefields could
easily be fifty mile wide and fifty miles deep. Obviously any attempt to
isolate individual battles in this continuity is not going to be easy; however,
the events listed here have been labelled battles at one time or another.
(Most of this list would not be allowed under the narrow, 19th Century
definition of "battle". A purist would probably count only a few
naval engagements (Midway, Jutland, etc.) and a few geographically restricted
operations (Berlin, D-Day, etc.) -- especially those confined to single islands
(Okinawa, Iwo Jima, etc.) -- as proper battles. Unfortunately for the purist,
common usage has already expanded the term to include larger events such as the
Battle of Britain, the Battle of the Bulge and the Battle of the Atlantic, all
of which would be called "campaigns" in Napoleonic or American Civil
War usage. My definition is typically loose, and I have counted a "battle"
as any military operation that is usually discussed and analysed by historians
as a single event.)
These battles are ranked by the number of military deaths only. I have not
counted civilian deaths because then we would drift off into the realm of
massacres. If we decide that Okinawa and Leningrad deserve a higher ranking
because of associated civilian losses, then we'd have to consider counting
Hiroshima and Nanjing, and that's just a different category altogether. For the
same reason, I haven't counted POWs that were escorted safely away from the
battlefield, only to die later in captivity.
One of the difficulties in compiling this list is that many writers don't
realize that casualties, deaths and losses are not the
same thing, so they count "losses" as killed, and call the dead "casualties".
I've preferred taking numbers from sources that demonstrate that they know the
difference. If I have found a source where the killed are differentiated from
the wounded on one side but not the other, I generally have applied the same
ratio to all sides. Otherwise, I just make do with the standard 3:1 ratio of
wounded to killed.
One comment: The unprecedented scale of the two world wars becomes apparent
when we realize that even small pieces of them killed more than most of the
other complete wars of the Century. Even more horrific, World War II's Russian
Front produced 13 of the century's 15 bloodiest battles all by itself. (The
siege of Leningrad alone could rank as the 20th Century's 12th worst
atrocity once civilian deaths are added in.) In fact,
part of why this list so long is that I wanted to get far enough down to show
some other wars as well. To help you keep these straight, I've color-coded the
principle wars: World War I,
World War II in Russia and World
War II elsewhere.
Here then is a very debatable and incomplete list of contenders for the
title of the 20th Century's bloodiest battles. My own estimate of total
military deaths in each is in bold following the name and date. All my sources
are listed, so if it's not immediately apparent where I got it, consider it as
just a rough guess.
David Glantz, The Siege of Leningrad 1941-44: 900 Days of Terror
Total number of soldiers and civilians who perished: 1.6M-2.0M
Soviet civilians: 642,000 d. in blockade (est. at Nuremburg trials), plus
another 400,000 d. in evacuations, for a total of about a million.
Soviet soldiers: 1,017,881 k/cap/mis + 2,418,185 wd/sick = 3,437,066
casualties [p.179]. (Based on statistics of Leningrad "Front", i.e.
army group [p.220:
332,059 k. + 24,324 non-combat dead + 111,142 captured &
missing = 467,525 "irrevocable"], it appears that KIA would be
71% of k/c/m, or ca. 725,000)
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA)
Defense (10 July-30 Sept. 1941): 214,078
Operation ISKRA (12-30 Jan. 1943): 33,940
L'grad-Novgorod Offensive (14 Jan.-1 March 1944): 76,686
[Total: 324,704]
Clodfelter
USSR: at minimum 100,000 mil. + 800,000 civ.
Germans: 500,000 casu. k/w/c
Harrison E. Salisbury, The 900 days: the siege of Leningrad (2003) p.516
Overall total of 1,300,000 to 1,500,000 deaths, military and civilian.
Albert Seaton, The Battle for Moscow 1941-1942 (1971) gives these
German losses on entire Eastern Front during Barbarossa:
22 June-26 Nov. 1941: 187,000 Germans K+M
27 Nov. 1941-31 March 1942: 108,000 Germans K+M
[TOTAL: 295,000]
Richard Overy, Russia's War (1997) tells us that the Soviets lost
3.1 million killed by Spring 1942.
Conclusion: During the first chaotic phase of the war, it seems that the
Soviets lost 10 soldiers killed for every German killed. Thus, in each early
battle, we can estimate German losses to be 1/10 Soviet.
Kiev, World War
II (7 July-26 Sept. 1941): 678 000 [make link]
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 616,308
Palmowski, Dictionary of 20th Century World History:
Allied: 400,000 k.
German: 350,000 k
[TOTAL: 750,000 k.]
Wikipedia: >250,000 ("Verdun") or 700,000 ("List of
Battles")
Ellis & Cox, World War I Databook
France: (362,000 casu.)
German: (336,000 casu.)
[TOTAL: 698,000 casu.]
Rzhev-Vyazma,
World War II (8 Jan.-20 Apr. 1942): 272 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 272,320
2nd West Ukraine,
World War II (24 Dec.-17 Apr. 1944): 270 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 270,198
North Caucasus,
World War II (25 July-20 Apr. 1942): 262 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA)
Defensive: 192,791
Offensive Operation DON: 69,627
[Total: 262,418]
Berlin, World
War II (16 April-7 May 1945):
250 000 [make link]
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 78,291
30 May 2004 Washinton Post review of Inside Hitler's Bunker
by Joachim Fest: Fest est. >300,000 Red Army soldiers died, contrasted by
reviewer with Antony Beevor's est. (Fall of Berlin 1945) of 78,000
Soviet dead.
Martin Sorge, The Other Price of Hitler's War (1986): 100,000
civilians d. incl. 20,000 cardiac arrest and 6,000 suicide. Not incl. 52,000
refugees k. caught in air raids. (citing Cornelius Ryan)
"In the Ypres Salient battlefields there are over... 40,000
unidentified graves.... Four memorials list the names of the more than 90,000
soldiers whose bodies have never been found or identified." [Note:
According to this source, these cemeteries and these statistics cover all the
battles in the Ypres area, not just the 3rd.]
Miller [http://www.lib.byu.edu/~rdh/wwi/comment/ypres3.html]
[ESTIMATE: Five of the six estimate of total casualties fall in the range
550-650 thousand. I'm guessing a quarter of the mid-point (600,000) were KIA.
(Estimates of British dead are consistently in the 60,000 to 80,000 range.)]
Okinawa, World
War II (1 April-21 June 1945): 148 000 [make link]
Toland, Rising Sun (1970)
Japanese soldiers: 110,000 lost
Okinawan civilians: 75,000 k.
US: 12,520 marines and sailors K+M
[TOTAL: ca. 122,500 military]
Gilbert, History of the Twentieth Century
Japanese soldiers killed: 107,500 bodies counted + 20,000 burned in caves +
7,800 Japanese aircraft shot down. [= 135,300]
Okinawan civilians: 80,000+ k.
US: 7,613 k. on land + 4,907 k. at sea [=12,520]
[TOTAL: ca. 147,800 military]
Global Security [http://www.globalsecurity.org/military/facility/okinawa-battle.htm]
Japanese soldiers: 107,539 k + 23,764 sealed in caves + 7,830 Japanese
aircraft shot down. [=139,133]
"French suffered 10,000 more than the Australians" [ca. 18,709]
UK: "French dead were less than half those of the British" [UK:
ca. 38,000+]
[TOTAL: 155,000]
[NOTE: I think maybe he meant that the British (instead of French)
suffered 10,000 more than the Australians (i.e. 18,000), which would then make
the French (at half the British) dead 9,000. Then the total would be 125,000.]
Wikipedia:
Allies: 44,072, incl....
UK: 21,255
Aus.: 8,709
NZ: 2,701
Turks: 86,692
[TOTAL: 130,764]
The Canadian Press (CP) January 7, 1990
Allied forces (UK, French, Australian, NZ, India): 46,000 k.
Turks: 86,692 officially. Unofficially, up to 250,000.
[Total: ca. 132,700]
26 April 2005 Guardian
Allies
killed: 55,000
missing: 10,000
disease: 21,000
Turks: 80,000
[TOTAL: 166,000]
22 Nov. 2003, Courier Mail (Queensland) and Herald Sun (Melbourne), AU
UK: 21,000 k.
Aus.: >8,700 k
NZ: 2,700 k
France: one third of 27,000 [=9,000]
Budapest, World War II (late Oct.
1944-mid Feb. 1945): 130 000
John Erickson
Barbarossa: The Axis and the Allies. Table 12.4: 80,026 USSR KIA
(29 Oct-13 Feb)
The Road to Berlin (1983): 50,000 German + Hungarian troops k. (27
Oct.-14 Feb.)
[NOTE: I'm making an exception here and counting civilian deaths (i.e.
merchant seamen) since they were an actual part of the battle, not just innocent
bystanders.
Leyte, World War
II (20 Oct. 1944-12 Jan. 1945)
69 000 [make link]
Toland, Rising Sun (1970)
Japanese: of 70,000, only 5,000 survived
US: 3,500 k.
Donbass, World
War II (13 Aug.-22 Sept. 1943): 66 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 661,166 KIA + 207,356 Wounded = 273,522 casualties [sic.]
661,166 is probably a typo for 66,166
Basra, Iran-Iraq War (1986-87): 65 000+
Clodfelter: During the period 12/86-4/87, mostly around Basra
Iran: 50,000 k
Iraq: 8,000-15,000 k
1 Aug. 1988, U.S. News & World Report, "Lessons of
history's bloodiest battle" by John Keegan: "It is not unrealistic to
estimate that 750,000 Iranian soldiers have died [in the War], most of them
killed in the last three years ... around Basra."
Some guy on Internet
(Elson Boles in 10 Oct. 2002 Counterpunch): "In the last major
battle of the Iran-Iraq war, some 65,000 Iranians were killed, many by gas."
Lvov-Sandomir, World War II (13
July-29 Aug. 1944): 65 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 65,001
3rd Aisne, World War I (May-June
1918):
64 000
Ellis & Cox, World War I Databook ("casualties")
France: 96,200
UK: 28,700
German: 130,400
[TOTAL: 255,300 casualties, one fourth of which is 63,825]
2nd Artois, World War I (May-June
1915):
62 000
Ellis & Cox, World War I Databook
France: 35,000 k. (102,500 casu.)
UK: (28,200 casu.)
German: (49,500 casu.)
[TOTAL: (180,200 casu.) Extending the same ratio as France (34%) gives
61,500 KIA]
2nd Baltic, World War II (14 Sept.-24
Nov. 1944): 61 000+
Erickson, Barbarossa: The Axis and the Allies. Table 12.4 (USSR
KIA): 61,468
Monte Casino,
World War II (1943-18 May 1944): 60 000 [make link]
30 May 2004 Washinton Post review of Monte Casino by
Matthew Parker: 60,000 Allied and German dead.
2nd Arras or
Vimy Ridge, World War I (8 April-16 May 1917): 60 000 [make link]
Tannenburg &
Masurian Lakes, World War I (Aug.-Sept. 1914):
37 000
Clodfelter
Russia: 30,000 k.
Germany: 13,000 k/wd/m
Ellis & Cox, World War I Databook ("casualties")
Russia: 267,000 casualties (137,000 POW)
German: 80,000 casualties
Warfare in World History by Michael Neiberg: 30,000 Russians k. at
Tannenburg + 125,000 casualties at M. Lakes. Germans lost less than 30,000 at
T'burg.
[If Germans "lost" <30T, then maybe 7T k.]
Mukden, Russo-Japanese War (20 Feb.-10 March
1905): 36 000
Clodfelter
Russians: 20,000 k. or missing (not captured)
Japanese: 15,892 k.
Wikipedia
Russians: 26,500 k.
Japanese: 41,000 k+w
Taierhchüang, Sino-Japanese War (24 Mar-8 Apr 1938): 31
000
Iran
Chamber Society: Operation Dawn V: "... within a few kilometers of the
strategic Basra-Baghdad waterway. Between February 29 and March 1 [1984], in
one of the largest battles of the war, the two armies clashed and inflicted more
than 25,000 fatalities on each other."
Warsaw, Russo-Polish War (13-25 Aug. 1920):
25 000
Wikipedia
Soviets: 15,000-25,000
Poles: 4,500
Changsha or Hengyang, Sino-Japanese War (June-Aug. 1944): huh?
Wikipedia: Over a million Japanese lives lost, incl. 360 officers. Five
million Japanese casualties total.
NOTE: A million is utterly ridiculous -- that's how many Japanese were
killed in the entire Pacific War -- but if
we agree that 360 officers were killed, then maybe 15 times as many enlisted men
or 5,400 were killed as well. (That's the ratio from the German Army 1871-72,
the only such statistics I have on hand.)
(Well, one more list for people to yell at me about...)
Here are some of the major episodes in which huge numbers of non-combatants
were killed at more or less a single place over a relatively limited time. The
traditional definition of massacre would require that all the killing be done
deliberately and face-to-face over the course of a day or two, but I've loosened
up that part in order to compare numbers from a variety of concentrated mass
killings.
I'm not saying that all of these events are morally equal. International
law usually allows the destruction of the war-making abilities of the enemy,
including their industrial infrastructure, transportation network and
(unfortunately) their workforce; however, once a civilian population or
collection of prisoners comes under the control of a conquering army, they are
no longer a threat, and they are supposed to be treated with basic human
decency. It all depends on whether the victims constitute an immediate threat,
an eventual threat, or no threat at all. You're allowed to kill sleeping enemy
soldiers in an armed camp in a war zone, but not sleeping enemy soldiers in one
of your POW camps. A merchant ship sailing in an armed enemy convoy may be
torpedoed without warning, but an unarmed merchant ship sailing alone on the
high seas should be challenged and allowed to evacuate passengers and crew
first.
Just to belabor an obvious point, these are individual events at single
points on the map - usually a city, prison or town. Most of the killing at
Treblinka, for example, occurred inside a
.1345 or
.21
square kilometer camp. The Rwanda massacres, on the other hand, were many
events spread out across the
26,338
square kilometers of an entire country, so I don't count those. In borderline
cases such as Katyn (three related massacres) and Kolyma (a large complex of
labor camps operating for many years), I've leaned toward inclusion because
these are smaller parts of a larger whole. Total death tolls for multiple
events considered collectively can be found using the main
index.
For the purists among you, I've starred (*) the events that are usually
considered to fit into the narrowest definition of massacre: the deliberate,
face-to-face and immediate killing of helpless victims. Also, I've used
brown font to label events of the Holocaust.
Adam Hochschild, The Unquiet Ghost: asked four researchers. "One
estimated [the death toll] at 250,000, one at 300,000, one at 800,000, one at
more than 1,000,000."
Raymond Pearson, The Rise and Fall of the Soviet Empire, 2nd ed.
(2002): 30,000
Some guy on
Internet (Marika Henneberg): around 30,000 people buried in 510 mass graves
at Kuropaty Woods, but there may actually be 900 graves [That indicates some
60/grave and maybe 54,000 people total]
Richard Overy, Russia's War (1997): 150,000-200,000 bodies
Some other
guy on Internet: 510 burial pits found in Kuropaty, each containing ca. 150
bodies, or some 75,000 bodies total. Possibly 1,000 pits originally [If 1,000
pits, then ca. 150,000 bodies]
Ian Dear, The Oxford Companion to World War II, "Belorussia":
"several hundred thousand" in mass graves.
2 Feb. 1993, The Age (Melbourne): "[T]wo million [Soviet] men, women
and children ... died." [incl. soldiers]
Moscow Times, Feb. 4, 2003: "More than 400,000 civilians were in
Stalingrad when the battle began.... By the time the battle was over... only
10,000 to 60,000 remained .... Many civilians presumably were killed by the
bombing and shelling, or succumbed to starvation and cold. But others were
evacuated during the battle, sent to toil in Germany as slave laborers or
managed to flee the besieged city on their own."
[ANALYSIS: Purely a guess... The Moscow Times offers 5 possible fates for
the 350,000 missing civilians: killed, succumbed, evacuated, sent, managed. If
we assign equal probability to each of these 5 possibilities, 70,000 civilians
would have ended up in each category. The number who were killed or succumbed
would total 140,000.]
Hiroshima, Japan (nuclear strike by US: 6
Aug. 1945): 122 000
1946 Manhattan Engineer District study: 45,000 first day, 19,000 next few
months, 66,000 by 1946. The author (Ford) suggests adding 20,000 transients not
on official census.
1946 Hiroshima police estimate: 78,150 dead and 13,983 missing. [a total
of 92,133 presumed dead] Also, 129,558 total casualties, including minor and
major injuries.
No sources given:
The Radiation Effects Research Foundation website: 90,000-140,000 in 1945
Hiroshima Peace Site website: 140,000 deaths by December 1945
Guinness Book of Records: 155,200 killed, including deaths from radiation
within one year.
Palmowski, Dictionary of 20th Century World History: 80,000
immediately. 60,000 more within a year.
Swatosh, Wings, wars and life: an autobiography (2007) p.70: 83,793
Howard Zinn, The people's history of the United States, p.422: 100,000
Gilbert, History of the Twentieth Century: 92,233 d in first two
weeks. 138,890 names on 1986 cenotaph.
5 August 1999 Chugoku
Shimbun: "Actual Status Survey of Atomic Bomb Survivors"
Identified victims who died by the end of 1945: ca. 88,800 counted
Official estimate (1988): 140,000 ± 10,000
Johnson, Modern Times: 100,000 that day, 100,000 subsequently
New Scientist, 15 April 1976, "New Japanese Report Registers Grim Atom Bomb Toll": 119,000 died in 1945, plus , acc2 official lost survey.
Bulletin of the Atomic Scientists, Jun 1986, p.37, citing 1946 official city survey:
died in the first year: 118,661
missing: 3,677
TOTAL: 122,338
CBS:
as of 6 Aug 2004, there were 237,062 dead listed on the Hiroshima city cenotaph.
Including 5,142 who had "died from cancer and other long-term ailments over
the past year" [... although IMO counting as fatalities people who
survived 59 years after the event is stretching the definition of massacre a bit
thin. Most people don't even survive 59 years after their own birth.]
Changchun, China (urban siege: May-Sept.
1948) 120 000
Jung Chang, Mao: the unknown story: Civilian population dropped
from 500,000 to 170,000. The official "watered-down" Communist figure
is 120,000 deaths by starvation.
[I'm accepting the official number because Change doesn't take into account the number of refugees who escaped the city. See [www.time.com/time/magazine/article/0,9171,887835,00.html]
Berlin, Germany (urban battle: 16 April-7 May 1945):
100 000
Martin Sorge, The Other Price of Hitler's War (1986): 100,000
civilians d. incl. 20,000 cardiac arrest and 6,000 suicide. Not incl. 52,000
refugees k. caught in air raids.
Davies, Europe: a history: no death toll offered, but it "held
some 300,000 souls in 1953"; "more human beings perished there than at
Auschwitz" [Davies exaggerates these things, IMO.]; "second only to
Kolyma".
Tokyo, Japan (air raid by US: 9 March 1945):
84 000
Factmonster.com:
"Two fire bombing raids on Tokyo [the second was in May 1945] killed
140,000 citizens."
Gilbert, History of the Twentieth Century, v.2, p.650: first official death
toll was 83,793, but eventually set at 130,000
Mark
Selden, Before the Bomb: The "Good War", Air Power and the
Logic of Mass Destruction: Official estimates of roughly 100,000 seem "implausibly
low".
Martin Sorge, The Other Price of Hitler's War (1986): 35,000
Spartacus: "Recent research suggest that 35,000 were killed but some
German sources have argued that it was over 100,000" [http://www.spartacus.schoolnet.co.uk/2WWdresden.htm]
Messenger, The Chronological Atlas of World War Two (1989): 50,000
Gilbert, History of the Twentieth Century: 39,773 bodies IDed. At
least 20,000 missing. All told, as many as 80,000 d.
NOTE: The most commonly cited death toll is
135,000,
but not among scholars.
Wikipedia: 135,000 ("Aerial bombing of cities" 5/04)
Johnson, Modern Times: 135,000
Irving: The Destruction of Dresden (1966): 135,000. See also
Nizkor's
discussion of this.
* Babi Yar, near
Kiev, USSR (massacre of Jews by Germans: Sept. 1941): 33 000
30 Dec. 2003 AP: of 16,000 prisoners at S-21, only 14 survived
Rachel
Hughes: Prisoners from S-21 were taken to Choeung Ek for execution. 89
graves with 8,985 skeletons have been exhumed, out of an estimated 129 graves
total. [Calculation: 129/89*8985=13023]
Generally
it's reported that 17,000 people died here, but I can't find a solid source for
this.
* Pinsk, Belorussia, USSR (massacre
of Jews by Germans: 29 Oct. 1942):
16 000