Are Rüdiger Overmans’ figures of German military fatalities in World War II plausible?

[Richard Anderson]

That's a most serious flaw than criticizing the sample size / selection.

I did not criticize sample size or selection, although there are possible issues with selection. For one, we do not know if it was random and the selection between the two data sets are different.

He might also have missed some...

Or may have counted some as dead who were not actually dead or counted as military casualties non-military casualties or...

Did he say "combat losses"? I thought Overmans was about losses-whatever-their-cause.

Yes, he did. If he was "about losses-whatever-their-cause" he chose a peculiar title for his thesis, Deutsche Militärische Verluste im Zweiten Weltkrieg, nicht wahr?

Mass death other than military are typically pandemic (like at the end of WW1), famine (China 1942, Leningrad 1942-43) or natural catastrophe (volcano eruption etc.) but none of these happened in Germany. So it would be kind of weak to argue for other causes than military to explain the bulk of these deaths, wouldn't it?

What "mass death" do you think I was referring to? Like Overmans I am talking about additional fractions of death that may not have been accounted for by wartime recording.

It's fair to consider they are by and large military casualties, don't you think? Their widows would get pensions without much argument.

Certainly, but then they are already accounted for as wartime missing and known PW.

[Mori]

Did he say "combat losses"? I thought Overmans was about losses-whatever-their-cause.

Yes, he did. If he was "about losses-whatever-their-cause" he chose a peculiar title for his thesis, Deutsche Militärische Verluste im Zweiten Weltkrieg, nicht wahr?

That's the point: "Militärische Verluste" is not the same as "combat losses". A mishandled bomb exploding in a rear area could kill soldiers nearby, a paratrooper can kill himself on training, etc. But that's not "combat".

So did Overmans limit himself to "combat losses"?

[thorwald77]

The Overmans studies have posed a dilemma for historians because his analysis cannot explain about 1 million missing based on the actual casualties at the battlefront. Total losses of the Third Reich total about 7.5 million including an unexplained difference of 2.2 million

• Millions
  KIA 2,350
  Non Combat & DOW . 500
  MIA(my estimate) . 250
  POW deaths 1,100
  Subtotal Military Losses 4,200
  
  Unexplained Difference 1,100
  Military Losses 5,300

• Total Demographic Losses 1939-50
  Germany 6,500
  Austria .300
  Ethnic Germans . 700
  Total Losses 7,500

• Summary 1939-50
  Military Losses 4,200
  Allied Bombing .500
  Expulsions .600
  Unexplained Difference Military 1,100
  Unexplained Difference Civilians 1,100
  Total Losses 7,500

Germans I spoke to 40-50 years ago mentioned post war losses of POWs and civilians expelled from the east. They lived on turnips and horse meat. The urban centers were rubble. They were bullied by the American occupation forces to sign statements denouncing the Nazis. The overall natural death rate was elevated from mid 1945 until 1948. Many women turned to prostitution. The streets were full of limbless men begging.

[Mori]

I did not criticize sample size or selection, although there are possible issues with selection. For one, we do not know if it was random and the selection between the two data sets are different.

On randomness of the sample - by the way, happy we agree that meaningful statistics require a random sample -, it's rather a bad faith argument to suggest Overmans's wasn't. His discussion on sample size (p. 183-185) only makes sense for a random sample.

I like some of your other arguments better.

[thorwald77]

I believe that the key to our dilemma is the timing of the category Verschollen which includes Vermißt and Letzte Nachtricht On page 285 his study overmans lists the combined total as Verschollen.

On page 176 of of Verluste Overmans writes

Vermißt im Sinne der vorliegenden Untersuchung bedeutet, das für den Betroffenen eine Vermißtenmeldung seiner Einheit vorliegt und das dies die letzte Angabe über seinen Aufenthalt darstellt. In ähnlicher Weise besagt Letzte Nachtricht, daß nur bekannt ist, von wann der letzte Feldpostbrief oder ein anderes Lebenszeichen stammt.

I believe that this means that the 701,385 Verschollen listed as Letzte Nachtricht are in reality unexplained losses. They are in fact card corpses. They could have died during the war or were never found after the war. For example surrendered military personnel who are unaccounted for may have never been sent to a POW camp or died in transit. They could have survived the war and died as civilians in Poland. The task of Dientstelle is to trace those missing that lacked reports issued by the military. Tracing the fate of Germans after the war, especially in the East bloc was a difficult task. From the point of view of Dientstelle they could tell relatives that the family member was reported missing during the war or that the person in question was not able to be found.

[Roberto Muehlenkamp]

Whether or not you're right about Operations Veritable and Grenade, your lecturing is quite uncalled for.

Correct: I do not need your permission.

Apparently you didn't realize what the subject of my article is. It is not meant to be an in-depth study about any of the military operations mentioned, but a reality check on Overmans' figures. For the operations I considered in order to do this reality check I picked sources available on the internet and made calculations as explained where these sources do not provide information about the number of KIA and MIA on either side. In each case I assumed the casualty ratio most unfavorable to the German side, whether I considered it realistic or not, in order to show that Overmans' figures are way too high even under such assumptions. I also identified my sources in every case.

You also stated you could not believe a case when German casualties were way higher than Allied, and I fixed that. Here are your words - and what are the statements in bold if not "preconceived notions"?

Operation Grenade (23 February to 11 March 1945): (...) Assuming that 25.52% or 1,863 of Ninth Army’s casualties were fatalities, 6,000 dead on the German side would mean an Allies versus German fatalities ratio of 1:3.22. This is highly improbable considering both the Western European campaign’s overall ratio (assuming Marshall’s figure for German and MacDonald’s figure for Allied battle deaths) and the ratio in Operation Veritable at about the same time, so the German figure must be an overestimate. It will nevertheless be used considering the aforementioned purpose of this exercise. Stacey states ... implying an also improbable Allies versus German fatalities ratio of 1:2.54 in Operation Grenade, so this estimate is probably also too high.

If you have a problem with the figures provided by my sources, I suggest you address these.

I published a book on these campaigns, by the way.

So here we have someone who is full of himself and cannot tell substantiated arguments (Marshall's figures must count for something, and MacDonald, by the way, even assumed that the number of dead and wounded on both sides was fairly equal in the 1944/45 campaign) from preconceived notions.

Not sure if I'm interested in whatever this fellow writes. He seems to have some axe to grind.

[Roberto Muehlenkamp]

Ah shucks Roberto my father was an American. His unit did not take prisoners. Orders were kill the cock suckers.

From https://en.wikipedia.org/wiki/Operation_Grenade:

Ninth Army captured 29,739 prisoners during the operation, and estimated to have inflicted 16,000 other casualties on the German army. In conjunction with Operation Veritable/Blockbuster, the combined allied effort inflicted approximately 90,000 casualties on the German army.[2]
Stacey, p. 522

Stacey = Stacey, Colonel Charles Perry; Bond, Major C.C.J. (1960). Official History of the Canadian Army in the Second World War: Volume III. The Victory Campaign: The operations in North-West Europe 1944–1945. The Queen's Printer and Controller of Stationery Ottawa.

So 9th Army must have taken some prisoners. Perhaps not that many given all those self-inflicted wounds in their ranks, but then these would also hinder killing.

[thorwald77]

Ho Hum Roberto
I wish my father was live to tell us about the real war. He hated the Nabelwerfer

[youtube]https://youtu.be/FqHuOMFZKCo[/youtube]

[Roberto Muehlenkamp]

Note that as of 2012, the WASt Totenkartei had grown to 4.6 million known dead and unaccounted for missing in action, from the 3.1 million cards that Overmans worked with prior to 1996. It would be fascinating to see a rework of his study utilizing the updated data and taking into account the issues identified in his original study. I suspect that the "correct" figure is closer to the 4.6 million rather than the 5.2 million.

The 2005 figures were the following:

Deutsche Verluste insgesamt 7 375 800
Gefallene deutsche Soldaten (einschl. Österreich) 3 100 000
Vermisste deutsche Soldaten (einschl. Österreich) 1 200 000
Deutsche Zivilbevölkerung 500 000
Deutsche Opfer von Vertreibung und Verschleppung 2 251 500
Bevölkerung Österreichs 24 300
Deutsche, durch politische, rassische, religiöse Verfolgung 300 000

Willi Kammerer; Anja Kammerer- Narben bleiben die Arbeit der Suchdienste - 60 Jahre nach dem Zweiten Weltkrieg, Berlin Dienststelle 2005, p. 12.

Is there a similar breakdown after the 2012 update?

[Richard Anderson]

Willi Kammerer; Anja Kammerer- Narben bleiben die Arbeit der Suchdienste - 60 Jahre nach dem Zweiten Weltkrieg, Berlin Dienststelle 2005, p. 12.

Thanks, that's actually informative, which cannot be said of many other posts in this thread.

Is there a similar breakdown after the 2012 update?

No, it was a passing remark in Der Spiegel that also noted the closures on the missing runs about 40,000 a year, mostly from discoveries of formerly lost cemetery plots in Eastern Europe.

[Mori]

So here we have someone who is full of himself and cannot tell substantiated arguments (Marshall's figures must count for something, and MacDonald, by the way, even assumed that the number of dead and wounded on both sides was fairly equal in the 1944/45 campaign) from preconceived notions.

Do you refer to SLA Marhsall? Since his data for the (infamous) "ratio of fire" were made up, it's a source one should refer to with care.

MacDonald and Stacey are both excellent and extremely reliable authors. I found the source documents they used to report German casualties in Veritable/Grenade: it's an estimate prepared by G-2 intelligence and dated March, 18, 1945. That is 8 days after the end of the operations.

And, frankly, it's a good source because it give quite some detail, and I understand why official historians relied on it. But, as Zetterling and others have underlined, it's a methodological error to use Allied sources to assess German losses. Except in this case an off-the-shelf German report assessing their own losses didn't exist.

However, I could gather enough information from German sources to build up an estimate of losses from their point of view.

For the operations I considered in order to do this reality check I picked sources available on the internet

I went quite deeper than mere "picking sources available on the internet",which in your case just meant opening Green Books and Canadian Official history. So you may as well keep quiet here.

[Mori]

Stacey = Stacey, Colonel Charles Perry; Bond, Major C.C.J. (1960). Official History of the Canadian Army in the Second World War: Volume III. The Victory Campaign: The operations in North-West Europe 1944–1945. The Queen's Printer and Controller of Stationery Ottawa.

Whom are you trying to impress by copy/pasting such a common and universally known reference?

[Roberto Muehlenkamp]

So here we have someone who is full of himself and cannot tell substantiated arguments (Marshall's figures must count for something, and MacDonald, by the way, even assumed that the number of dead and wounded on both sides was fairly equal in the 1944/45 campaign) from preconceived notions.

Do you refer to SLA Marhsall? Since his data for the (infamous) "ratio of fire" were made up, it's a source one should refer to with care.

MacDonald and Stacey are both excellent and extremely reliable authors. I found the source documents they used to report German casualties in Veritable/Grenade: it's an estimate prepared by G-2 intelligence and dated March, 18, 1945. That is 8 days after the end of the operations.

And, frankly, it's a good source because it give quite some detail, and I understand why official historians relied on it. But, as Zetterling and others have underlined, it's a methodological error to use Allied sources to assess German losses. Except in this case an off-the-shelf German report assessing their own losses didn't exist.

However, I could gather enough information from German sources to build up an estimate of losses from their point of view.

For the operations I considered in order to do this reality check I picked sources available on the internet

I went quite deeper than mere "picking sources available on the internet",which in your case just meant opening Green Books and Canadian Official history. So you may as well keep quiet here.

That's an improvement over your previous post except for the last sentence. As I said the purpose of my article was quite different from that of your book, so I didn't go into much detail regarding any of the listed operations. Even if I had wanted to, it's hard where I live to get hold of US or British publications on military history other than what you find on the internet. And besides, my point regarding Overmans is made not only with the figures I considered improbable (and used nevertheless), but also with your figures for Grenade and Veritable.

Glad to know that you consider MacDonald a reliable source. On page 478 of The Last Offensive he wrote the following:

How many of the three million Germans that were killed during the entire war died on the Western Front is impossible to determine, but
exclusive of prisoners of war, all German casualties in the west from D-day to V–E Day probably equaled or slightly exceeded Allied losses.

In a related footnote he wrote:

The only specific figures available are from OB WEST for the period 2 June 1941 – 10 April 1945 as follows: Dead, 80,819; wounded, 265,526; missing, 490,624; total, 836,969. (Of the total, 4,548 casualties were incurred prior to D-day.) See Rpts, Der Heeresarzt im Oberkommando des Heeres Gen St d H/Gen Qu, Az.: 1335 c/d (IIb) Nr.: H.A./263/45 g. Kdos. of 14 Apr 45 and 1335 c/d (Ilb) (no date, but before 1945). The former is in OCMH X 313, a photostat of a document contained in German armament folder H 17/207; the latter in folder 0KW/1561 (OKW Wehrmacht Verluste). These figures are for the field army only, and do not include the Luftwaffe and Waffen-SS. Since the Germans seldom remained in control of the battlefield in a position to verify the status of those missing, a considerable percentage of the missing probably were killed. Time lag in reporting probably precludes these figures' reflecting the heavy losses during the Allied drive to the Rhine in March, and the cut-off date precludes inclusion of the losses in the Ruhr Pocket and in other stages of the fight in central Germany.

Assuming MacDonald's assessment is correct, Marshall's figure of 263,000 German dead in the 1944/45 campaign (which is still much lower than the ca. 655,000 suggested by Overmans) would be too high.

[Roberto Muehlenkamp]

Stacey = Stacey, Colonel Charles Perry; Bond, Major C.C.J. (1960). Official History of the Canadian Army in the Second World War: Volume III. The Victory Campaign: The operations in North-West Europe 1944–1945. The Queen's Printer and Controller of Stationery Ottawa.

Whom are you trying to impress by copy/pasting such a common and universally known reference?

No one, I'm just mentioning the source featuring the information I provided. I don't see this thread as a contest of who is cleverer or better informed.

[Roberto Muehlenkamp]

Willi Kammerer; Anja Kammerer- Narben bleiben die Arbeit der Suchdienste - 60 Jahre nach dem Zweiten Weltkrieg, Berlin Dienststelle 2005, p. 12.

Thanks, that's actually informative, which cannot be said of many other posts in this thread.

Is there a similar breakdown after the 2012 update?

No, it was a passing remark in Der Spiegel that also noted the closures on the missing runs about 40,000 a year, mostly from discoveries of formerly lost cemetery plots in Eastern Europe.

I guess you mean the article https://www.spiegel.de/international/ge ... 32063.html. Reading it revived my hope that the remains of my uncle Obergefreiter Ernst August Schmidt (http://www.geocities.ws/orion47.geo/WEH ... kamp.html ) might be found and I would be able to take his twin sister, who passed away in 2017, to a proper burial of those remains in the region where he went missing during the Red Army's Moravia-Ostrava Offensive (https://en.wikipedia.org/wiki/Moravia%E ... _Offensive). Unfortunately that hope didn't materialize.