French Military Losses of WWII

Discussions on all aspects of France during the Inter-War era and Second World War.
Post Reply
takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

French Military Losses of WWII

#1

Post by takata_1940 » 24 Oct 2010, 07:16

French Military Losses of WWII

It was published at the colloque, held in 2000, "La Campagne de 1940", sous la direction de Christine Levisse-Touzé, éditions Tallandier, an article from Dr. Jean-Jacques Arzalier p.427-446 intitled : La campagne de mai-juin 1940, Les pertes?

- The Campaign of May-June 1940, the losses?
Very good question indeed!

This doctor, reservist from the Service de Santé des Armées (Military Health Service), pointed that no statistics existed at the begining of the XXIth century about French losses during this campaign. Archival material was scarce at best, contradictory and confused about detailed figures in several reports. No study was ever made about those figures because several public services, civil and military, were in charge of different parts of the pension system. He asked this question to all of them but nobody could answer, being unable to know precisely how many French soldiers died between 10 May and 25 June 1940.

The best you can get about it was an estimation and it varied from one source to another between 92,304 to 129,466 killed and 120,000 to 250,000 wounded. The official SHAT figure mentioned 94,203 killed in 2000.

Now, J-J Arzallier pointed that the figure about evacuation of wounded taken in charge by the Service de Santé during the campaign was recorded at 122,695 men which statistically would imply between one third and one half this number as killed, hence something around 40,000-60,000 deads.

He estimated also that the number of those who died later from their wounds would be around 10%, adding 12,000 deads to the above figure. This percentage of DOW was 12% during WWI and 4,5% in the US Army 1941-1945. Nonetheless, the French Military Health Service in 1940 was not equiped for this war of movement and many casualties had to be moved to the rear without recieving more care than first aid. Delays for evacuations reached 4-5 days instead of the few hours planned. hence, an appalling rate of dead caused in field hospitals and sanitary trains.

His estimation for the end figure for Killed + DOW (as the official figures should include them) was in the range 52,000 to maximum 72,000 -believing it closer to the low estimate. Still many people are still publishing today the old figures of 90,000-120,000 French soldiers killed in May-June 1940 and some are still claiming that it was higher than early 1914 loss rate (which is not).

Now, before the 70th birthday of the 1940 campaign (2010), the French Ministry of Defense (at least) had put online a database of the French military personnel which died during World War II: http://www.memoiredeshommes.sga.defense ... rubrique54

This is a huge progress... but it is still a complete mess for some people (like me) which would like above all a clean dataset to work with. This is far from being the case.
I don't know if the page "Mémoire des hommes" in English (about WW2) is understandable for you but I barely do understand what they say.

1. The database:
It is filled with all the people who have died/missing during the period 1939-1945, either belonging to the French Army (Regular, Territorial, Colonial, Foreign Legion, Gendarmerie, including Vichy-period Army as well as the Free-French forces) or having at some point a military status: paramilitary organisations of the Resistance (FFI, FTP, and various underground réseaux, men and women).

2. Fields are:

Name,
First-name(s),
Surname,
Born date,
Born place,
Department/Country,

Rank,
Unit,

Date of death,
Death place,
Department/Country of death,
Cause of death,
Mention ("Mort pour la France" or not),
Status (Army, Airforce, Navy),
No. of File (War victims archives).

Now, it would be great if most of the fields were not "optionnaly" filled and had been seriously verified before typing. In fact, they are certainly copied from the initial death/disparition reports (like those of WWI database in the same site) and are full of errors, nor are they updated with coherent infos about the case. Moreover, they have added plenty of typos (or OCRs errors) to the original ones comming from handwritting reading or incomplete data.

The most completed fields are "Name" and "No. of File". One may find out more than one entry for one single individual (name variations) with different data. One may find two different individuals referred by the same File number (typo).

"Rank" field is 99.9% empty.

Places and locations are not reliable (homonymie... the guy dying accidently in Tripoli-Libye in Dec. 1940 certainly died in Tripoli-Lebanon as his unit was deployed in Levant). There is almost as many spelling for one location as it is possible to imagine them. Don't ever hope to find all the guys who died at "Yoncq" or "Stonne" without browsing the whole dataset including many crosschecks.

Colonial troops' data are very poor, many time worse than others.

Too many dates are approximate or incorrect. This guy "killed in combat" in dec. 1940 in Belgium did not fall here or then.

Regiments numbers/name/sigle are many time worse than "Locations" concerning accuracy/typo, also "3rd Battalion", "2nd Company", "11th Battery" do not provide that much info by themselves...

Navy losses are missing at this point but most Airforce looks to be included even if the relevant field (status) is usually incorrect.

The most depressing fact about this base is that less than 50% of the causes of death are filled and it is often not usefull at all or contradictory with the delivered mention "Mort pour la France" (died on Duty) or not delivered. This is the only way to find out if someone was KIA, MIA, Died of Wounds, or committed suicide, died in his bed during leave, from sickness, was murdered, or had had an accident completely unrelated to service.

On the other hand, the base seems as complete as possible, even if somewhat unreliable, as there is certainly slightly more entries than people which died or went missing due to double entries by various services.

(to be continued... with tables of French losses)

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#2

Post by takata_1940 » 24 Oct 2010, 13:58

Lets start with the overall figure.

This first table is showing the number of Yearly entries from the dataset 1939-1945 conflict. The total of 202,991 stand for any military personnel belonging to the Army or Air Force which died worldwide-whatever the cause- during the period, or never returned home.
F_Died_Yearly1.jpg
F_Died_Yearly1.jpg (34.32 KiB) Viewed 16522 times
- Columns:
"T": is the total
"M": mentioned "Mort pour la France" (died on duty: KIA, MIA, DOW, in captivity, executed, from accident or sickness which are related with duty)
"P": not "Mort pour la France";
"Q": not mentioned.
- Rows:
"1939", from 4 September to 31 December.
"1945", 1 January to 31 December.
"1946+", files from 1 January 1946 and after... those are errors or belonging to Indochina and other colonial operations.
"xxxx", files for which the field "date of death" is missing.


takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#3

Post by takata_1940 » 24 Oct 2010, 14:42

Monthly figures (total died) in two parts:
1. 1939-1942
F_Died_Monthly1.jpg
F_Died_Monthly1.jpg (68.79 KiB) Viewed 16520 times
2. 1943-1946+
F_Died_Monthly2.jpg
F_Died_Monthly2.jpg (67.92 KiB) Viewed 16520 times
Yellow rows month "xx": month is not available for this Year file: ie. "1939-xx-xx".

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#4

Post by takata_1940 » 24 Oct 2010, 16:15

As my main period of interest is 1939-1940, I'm posting a more detailed Monthly table 1939-1942 showing the ratio of "Morts pour la France" (M) vs other classes "P" and "Q":
F_Died_Monthly3.jpg
F_Died_Monthly3.jpg (138.18 KiB) Viewed 16513 times
Months bolded are (relative) peak military activity.
- September 1939: Saar
- May-June 1940: Belgium, Netherlands, France
- June-July 1941; Lebanon-Syria (Operation Exporter with French forces on both sides, sometime belonging to the same units)
- June 1942: Madagascar (Operation Ironclad)
- November 1942: North African landings.

It seems obvious that "P" (non duty related losses) is a function of the population size (men mobilized) rather than military activity, while "Q" (no mention of category) is a mix of both as it will have many unsettled cases belonging to either category. There is much more "Q" for the period before the armistice than after.

"P" is quite heavy in September 1939. I guess it is due to many cases of death from sickness which were rejected as "duty related" during mobilization while after, more would be accepted. Sickness is the main cause for documented death in both "duty" and "non-duty" classes when the military activity is low, followed by accidents to a lower level. Suicide cases are strikingly high at home and overseas.

From July 1940, there is a transfer from "P" to "M" because deaths in captivity would not be classified the same as before. All deaths by sickness, accidents, suicide, etc. while in captivity would be considered "M" (on Duty) but many such cases would not be at home. Then "P" classe is shrinking with the Army demobilization as the remaining population is now based mostly overseas (about 500,000) or is POW (about 1,300,000). Previous figures are in the range of 5.5 million with 1,800,000 regulars in Metropole, 4-500,000 overseas, and the remaining affected to various home duty. So by 1942, the population is reduced to below 2,000,000 including POWs.

I'll post later daily figures.
S~
Olivier

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#5

Post by takata_1940 » 24 Oct 2010, 17:28

Here is a big table showing the uncorrected May-June 1940 daily deaths, including all causes and worldwilde.
Still by M, P, Q classes.
Remember that this table is built from the raw dataset, which need many corrections -that I will discuss later, in order to show more accurate numbers by adding incomplete field "date of death" data.
F_Died_Daily_4005.jpg
F_Died_Daily_4005.jpg (173.56 KiB) Viewed 16503 times

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#6

Post by takata_1940 » 24 Oct 2010, 21:07

Getting rid of "Q" column.

A close data examination of "Q" column (duty/off duty unsettled cases) will show that most files do belong to missing soldiers and comparisons with other series will suggest that their number is mostly related to military activity, contrary to "P" which is related to overall Army size. So "Q" may be added to "M" as we are not the French administration which needed some facts to do it.
On 171 "Q" files for June 1940, I've found that:
- 13 killed in combat
- 1 died of wounds
- 3 lost at sea
- 1 died in captivity
- 75 missing
- 78 nothing
So I wonder which informations is right and why the mention "Mort pour la France" wasn't issued for those 18 documented cases. This would need a case by case verifications of each file.

But for the purpose of making an estimation, it is not necessary to keep them aside as those listed "Mort pour la France" could have been KIA, MIA, DOW, died from sickness, died in captivity, killed in accident, but all those deaths are supposed to be related with military activity.

Now, we can check all excess numbers which are not ventilated by years, months and day, all affecting this May-June campaign estimation.
We've got:
- 7,051 deaths, no year => 6,757 (duty); 294 (off duty)
- 644 deaths in 1940, no month => 634 (duty); 10 (off duty)
- 424 deaths in May 1940, no day => 420 (duty); 4 (off duty)
- 779 deaths in June 1940, no day => 775 (duty); 4 (off duty)

It is necessary to dispatch them where they had the greater chance to belongs statistically: in the first case where the Army suffered the heaviest losses (=more military activity), and in the second case, where the Army was the biggest and its daily "P" losses were the smallest (it should be a function of number of men mobiliized*day). Hopefully, there is very few of them to dispatch and it won't change that much an estimation if I'm using a more simple ratio.

The next tables would take into account those changes from the original dataset in order to refine this estimate.

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#7

Post by takata_1940 » 27 Oct 2010, 18:24

Part II.

After having presented above the main dataset and all its limitations, it is nonetheless possible to use it in order to provide a closer estimate of the real combat losses (due to enemy action) during the 1939-1940 campaign which was my main point of interest.

Assuming that the information lacking in files with incomplete dates is barely random over all the period, I have dispatched those deaths (weighted by category) amongst yearly, then monthly figures. The same was applied to other remaining yearly figures without month mentioned.

Here are the figures corrected:
F_Died_Monthly_Corr1.jpg
F_Died_Monthly_Corr1.jpg (85.41 KiB) Viewed 16440 times
F_Died_Monthly_Corr2.jpg
F_Died_Monthly_Corr2.jpg (69.71 KiB) Viewed 16440 times
Also, I stopped the table at the end of August 1945, discarding the post war months but the whole figure of 202,991 is the same including them.

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#8

Post by takata_1940 » 27 Oct 2010, 20:49

Now, focusing on the period Sep.1939-June 1940.

from October to July 1940, when the demobilization of the remnants of the metropolitan army started, the effective of the Army remained at about the same figure of 5,500,000 men (without the Navy, as the dataset is not including Navy personnel - see next table for details).

From samples taken during the period Sept 1939-April 1940, it is obvious that the Non-combat Casualties were much higher than the "Non-Duty" numbers. They would include all death causes: sickness, accidents, suicides, missing... whatever classification was applied for issuing the mention "Mort pour la France".

"Non-Duty" cases are remaining very stable up to the end of June: an average of 20 deaths per day. The high point is during September 1939 - in fact during the first decade of mobilization. But this peak is compensated during the following months: The average for 1939 last four months is exactly the same as for the 1940 first six months: 20 deaths per day (+/- 1).

Then, from the samples, it may be estimated that the average non-combat attrition rate was about 40 deaths per day, or very close to this number during the whole period before the armistice and demobilization. This is close to the estimated rate of 2 to 3 per thousand men dying yearly in the whole Army (affected by various factors as seasons and localizations).

Next table is showing the figures for Army size - considered constant for the period- and estimated attrition rates:
F_Attrition.jpg
F_Attrition.jpg (34.01 KiB) Viewed 16426 times

takata_1940
Member
Posts: 469
Joined: 01 Jun 2007, 06:48
Location: France

Re: French Military Losses of WWII

#9

Post by takata_1940 » 27 Oct 2010, 23:26

Conclusion.

This attrition rate of 40 Non-combat deaths per day seems to fit well with Sep-Dec agregated figures (not so well monthly though) but it seems to be a bit low for the begining of 1940's year where the level of military activity was actually very low. I then raised it to 45 for 1940.

Two possibilities behind that:
1. "Natural" winter deaths might have been much higher. This year winter was particularly harsh - hardly remembered in history, killing scores of horses. Then, natural over-mortality for men also.
2. Bad repartition (handwritting, typos, OCR errors...) of May-June files ending mixed with early 1940 ones. This need a closer look at each case... not possible.

So, lets take this estimation with a possible margin of error of +/- 5 percent due the raw dataset:
F_Estimated_39-40.jpg
F_Estimated_39-40.jpg (20.54 KiB) Viewed 16419 times
The number of soldiers dying from combat will now ammount to a little bit over 61,000 for May-June, +/- 3,000 = 58,000-64,000 killed from combat causes.

It is to be remembered that this total is including "Killed in action" (KIA) and "died of wounds" (DOW).
1. Assuming that the number of KIA is 1/3 the number of WIA and that 1/10 of WIA will die later (usually the next or following next days).
2. Assuming that the "natural attrition rate" in May-June was about 3 per thousands yearly.

Then:
- 61,000 killed = 1.3 *KIA
=> 47,000 KIA, 140,000 WIA, followed by 14,000 DOW, estimated with an error margin of +/- 5%.

Which put the intervall at 45,000-50,000 KIA during the May-June 1940 Campaign. Soldiers dying later of their wounds will compensate for the non-campaign losses accounted during the first 9 days in May and last 5 days in June.

The number of 140,000 WIA will fit pretty well also with the recorded 123,000 evacuated by the French Health Service which I quoted above. Dr Azalier's estimation quoted seems to be supported fairly well by the real data, but the real number of wounded is of course higher due to those which were taken in charge by the Germans. A look at the "Mémoire des hommes" MOD site will show that the official number was also officially revised from 90,000 down to 55,000 "killed" between 10 May and 20 June.

I was fairly long for describing this estimation as my point was not only to provide this end number, but to explain the process. I will welcome any comment about it. The next step, for a closer estimate, will need a full verification of this dataset... not for tomorrow then.

S~
Olivier

Piotr Kapuscinski
Host - Allied sections
Posts: 3724
Joined: 12 Jul 2006, 20:17
Location: Poland
Contact:

Re: French Military Losses of WWII

#10

Post by Piotr Kapuscinski » 23 Feb 2013, 04:20

So all these numbers generally include casualties of French partisans?

Regarding period January 1941 - August 1945:

I counted 105,206 French military (and paramilitary?) deaths on duty (this includes: killed, missing, died of wounds, died as POWs, in executions, in accidents and of sickness - all of these related to duty - as you wrote).

1941 - 11,725
1942 - 7,669
1943 - 11,454
1944 - 48,106 (including as many as 9,715 in August)
1945 - 26,252

But all these numbers include also deaths of partisans, right?

Is it possible to isolate casualties of partisans and Vichy France from casualties of Free France?

And also - is it possible to isolate French deaths on duty in the Far East (against Japan and its allies)?
There are words which carry the presage of defeat. Defence is such a word. What is the result of an even victorious defence? The next attempt of imposing it to that weaker, defender. The attacker, despite temporary setback, feels the master of situation.

User avatar
tigre
Member
Posts: 10549
Joined: 20 Mar 2005, 12:48
Location: Argentina

Re: French Military Losses of WWII

#11

Post by tigre » 11 Jul 2016, 00:48

Hello to all :D; a little complement...............................

The heroic conduct of our Generals in May-June, 1940.

As a result of the speed of operations in modern warfare general officers may find themselves in the thick of the fighting rather than in the traditional sheltered headquarters somewhere behind the lines. Many French generals, during the course of the operations in Belgium and in France, found themselves organizing personally strong points and centers of resistance. Lieutenant General Giraud was captured while at the last center of resistance of his army. Major General Bouffet was killed by fire from hostile planes while returnnig from a tour of inspection of the critical points of his corps' defensive zone, Major General Janssen was killed at his post while attempting, with the remnants of his division, to hold back the enemy during the evacuation from Dunkerque. Major Generals Berniquet and Barbe were killed while fighting at the front wIth their "poilus," Generals du Picq and d'Humieres were killed while inspecting the disposition of troops in their brigade zones. General Moulin was shot and killed while organizing strong points with his troops, despite the danger of being enveloped Brigadier Generals Corbe and de Prevaux were surrounded by the enemy, but they succeeded after several harrowing experiences and several days of marching in getting back to unoccupied France.

Source: Catalog of Selected Periodical Articles. RML Vol XX. Nº 79. Dec 1940.

Anyone else in that list? TIA. Cheers. Raúl M 8-).

User avatar
Loïc
Member
Posts: 1227
Joined: 14 Jun 2003, 04:38
Location: Riom Auvergne & Bourbonnais France
Contact:

Re: French Military Losses of WWII

#12

Post by Loïc » 12 Jul 2016, 18:55

hello
the ranks are anglosaxonized, Lieutenant General (for an Army Corps General) ceasing to exist in the French Army since 1848 and was a Divisional general or more sometimes, Major General (for an angloamerican Divisional) was not a rank but a high function in the French Army Staff only to one general or marshall (Marshall Berthier under Napoléon, General d'Armée Bineau in 1939 then Doumenc in 1940)

in both cases this poor Giraud was certainly not lieutenant general (french 2nd rank anglosaxon 3rd rank) but Général d'Armée (top 4th rank of the generals officers) and Bouffet was not major general (divisional) but an Army Corps General highest rank general killed after Billotte

Generals killed in may-june 1940 :
http://forum.axishistory.com/viewtopic. ... 2&t=100909

Air Brigadier General Augereau killed while defending 9th Army HQ
Brigadier General Caille with the 21e DI in Boulogne encircled
Brigadier General de Courson de la Villeneuve while resisting to his capture
Brigadier General Thierry d'Argenlieu encountering a panzer column
Brigadier General Deslaurens with the troops sent to Flessingue, Netherlands

User avatar
tigre
Member
Posts: 10549
Joined: 20 Mar 2005, 12:48
Location: Argentina

Re: French Military Losses of WWII

#13

Post by tigre » 12 Jul 2016, 23:38

Thanks for your reply Loïc :wink:. Cheers. Raúl M 8-).

Post Reply

Return to “France 1919-1945”