German Casualty stats

[wiking85]

I've come across the debate about German casualty reports not including lightly wounded soldiers in their numbers, especially in reference to the Sanitätsbericht and the Somme. However, in a scholarly article "Blood test revisited" by James McRandle and Jame Quirk, this argument is dismissed as fatuous. Yet I continually see this argument in reference to the Somme and German casualties there. In fact, the recently released book "Bloody Victory" by William Philpott, furthers this argument, stating that the Sanitätsbericht should have 30% greater numbers than is actually stated. Does anyone have any idea why this idea continues to float around and if it holds any water?

[bob lembke]

I can offer some comments and anecdotes, but not hard data. Might shed a bit of light.

My father was wounded four times. I have considerable information on each, plus his Militaer=Pass, his decorations, etc. From this, on his wounds:

#1 - Shot on a raid with a small caliber handgun, muzzle blast burn obscured the entry wound, he was refused treatment, went back to first aid post when the bullet worked its way to some skin, was removed locally. Verdict - a non-wound. Not entered in Militaer=Pass.

#2 - Hit by shell fragment in arm, had to be left (entire Trupp wounded by shell burst) in no-man's-land, found three days later. Continual re-infection, in and out of hospitals for a year, classified not fit for combat and put to training duties. Wound spit bone for over 10 years. Verdict: Officially classified "light wound", entered as such in his Pass, which p****d him off 50 years later.

rated not fit for combat, toward end tricked his way back to the front to fight with his unit

#3 - Wound to face, much bleeding, but then almost drowned in mud by stupid medics. I don't think he was hospitalized, 99% certain. Verdict - a non-wound.

#4 - Gassed and blinded, hospitalized for at least a week, evacuated. Regained sight later. Verdict: A wound, entered as blinded by gas in Pass.

He was given the black Wound Badge, for one or two wounds. Lighter wounds officially ignored.

I can also remember, Ernst Juenger, sitting in his hospital bed, and commenting/writing: "My papers say I have been wounded seven times, but I can count 14 holes in me." ("Storm of Steel")

Seems like really light wounds not counted. If hospitalized, probably counted. If you can walk, even with a serious wound, more likely to be a "light wound".

German casualty figures suggest that officers were much less likely to be declared "wounded". Many German regimental hisories give reports like, for an engagement: "Enlisted men, 85 killed, 290 wounded. Officers, 6 killed, 7 wounded." Not sure of the reason for this, but I have seen it a number of times.

Hope that helps a bit.

Bob Lembke

[Ralph J. Whitehead]

'German casualty figures suggest that officers were much less likely to be declared "wounded". Many German regimental hisories give reports like, for an engagement: "Enlisted men, 85 killed, 290 wounded. Officers, 6 killed, 7 wounded." Not sure of the reason for this, but I have seen it a number of times.'

Bob,

I am not sure about this last bit, from what I have seen if a soldier, regardless of rank, was wounded, missing, etc. then they were listed in the casualty reports. In the instance you mention it is quite possible all of the officers involved in the units were casualties, it all depended upon the total numbers and the ratio between killed and wounded was never the same for both categories. Many companies had as few as two officers at any given time, others more, it depended upon losses already suffered, a shortage ot trained officers, etc.

Ralph

[bob lembke]

I am not sure about this last bit, from what I have seen if a soldier, regardless of rank, was wounded, missing, etc. then they were listed in the casualty reports. In the instance you mention it is quite possible all of the officers involved in the units were casualties, it all depended upon the total numbers and the ratio between killed and wounded was never the same for both categories. Many companies had as few as two officers at any given time, others more, it depended upon losses already suffered, a shortage ot trained officers, etc.

Ralph

Hi, Ralph!

I have seen this discrepancy so many times that it has made a strong impression. Unfortunately, as it is not one of the 83 things that I have decided I am especially interested in, and build elaborate time lines for, I have not collected examples of it in an organized (or disorganized) fashion. Sometimes I see accounts of casualties, and the wounded / killed ratio seems "normal"; e.g., about three wounded per each KIA, but sometimes I have noticed a sharp difference in the "ratio" between the two classes of men. Having spent mose of my life as an analyst, computer programmer, and economist (or a student of same), numbers, ratios, etc. tend to stick in my mind. At the same time, I have never actually pored thru official tables of casualty statistics, presented as same.

I had wondered about the possible causalities for this phenomenon, if true, but my thoughts were so fragmentary that I didn't add them to the account above. The analyst in me sits on my shoulder and whispers in my ear that due to the laws of probability, while the numbers of officers killed and wounded might be higher or lower than the proportionate share of the OR/EM killed and wounded, adjusted by the ratio of the two classes of men in the unit, due to factors like "officers leading from the front", the randomness of the nature of wounds inflicted would suggest that the ratio of killed to wounded within the two classes of men should be roughly equal.

However, behaviorial differences between the two classes might account for such a difference. Possibly most ORs, once wounded, would feel that they had done their duty, drop to the ground, take cover if possible, and await aid; while most officers might struggle forward, wave their real or figurative swords, plow ahead, and perhaps take another round or two before falling dead. Perhaps most officers might lead from the front, act officer-like, and receive multiple aimed shots by an alert defender, and die in higher numbers. Perhaps in the society of officers it was not considered "officer-like" to report or present minor injuries, while many ORs would make a point of presenting the minor wound. There are three possible causalities, based on assumed behavioral differences between officers and ORs. I really don't know.

But numbers tend to stick in my mind, and usually accurately, I think. When traveling I drive my wife crazy with accounts of the prices of meals or accomodations I encountered 40 years before, like the roast wild boar dinner at the restarant of the Slovene Hunting and Fishing Association in February 1967 for $1.23 (with Maja Smole, the daughter of the then President of Slovenija), or the filet mignon dinner (two filets on the plate) at the Black Cat Cafe with the office tri-lingual secretary (and police spy) Zora in March 1967, both in Ljubljana, Jugoslavija, both at the then official exchange rate of 12.50 dinars to the dollar.

Bob

[Robert Dunlop]

William Philpott, furthers this argument, stating that the Sanitätsbericht should have 30% greater numbers than is actually stated.

Philpott provides a brief review of the different numbers but his conclusion is "The complex, often incomplete and contradictory nature of German statistical returns is not in dispute. An accurate figure for German casualties on the Somme will never be estabished... Whatever the precise number, there is much evidence to suggest that German units fighting on the Somme were decimated".

Jack Sheldon has posted on this issue here .

Ralph Whitehead has described the difficulties of analysing the numbers here and his efforts to do this for the Somme are outlined here.

Robert

[Ralph J. Whitehead]

Since this subject has obviously come up before and I am confident will come up again it seems the more we look at the issue the better we understand it.

We will never know the true numbers of men lost at any given point for any side with some possible exceptions here and there. Simply put, in my opinion alone, the losses recorded on day 1 of an action would be altered as additional facts come to light. Men missing could be found dead, wounded or simply back with the company and reported in error. Let's not forget that people created the lists from information supplied by others that was often accumulated under very stressful times. To consider that 100% accuracy was a goal or was achieved in most instances is not realistic, again my opinion alone.

I am still accumulating the loss details for the Somme regiments and will have 99% of them done by this summer if all goes to plan (as we know this never happens). I have just finished up the 4th Coy, 8th Bavarian Reserve Regiment, one of the units that occupied the Schwaben Redoubt on 1 July 1916. As expected the losses were higher in this company than many of the others and most of the fatal losses and MIA can be attributed to events on 1 July. When looking at the overall numbers it can be easily be determined which units saw the heaviest fighting, which were in reserve or used at a later date by the numbers killed, the dates of their deaths, etc. When comparing to historic records, and in this instance the records from the regiment still exist, we can get some idea of the overall losses associated with 1 July as well as the Somme in general.

I will post the breakdown results along with a brief historic record to illustrate this process shortly. It is a nice day and we are off for a day out.

Ralph

[Ralph J. Whitehead]

I've come across the debate about German casualty reports not including lightly wounded soldiers in their numbers, especially in reference to the Sanitätsbericht and the Somme. However, in a scholarly article "Blood test revisited" by James McRandle and Jame Quirk, this argument is dismissed as fatuous.

Just a brief response to this aspect of the post before we leave for the day. In looking over the casualty returns for the different units for the Somme fighting everything is included in the returns from killed, missing, severely wounded, slightly wounded, slightly wounded but remained with the regiment, injured (slightly or severely) and in cases where wounds resulted in death at a later time, either the same day or possibly days or weeks afterward and the lists had not been published yet, these men also had the dates of their death posted as well.

From looking at literally hundreds of different lists from every period of the war I doubt that if a man was injured in some manner he was not placed on the report. This is easily shown by the men listed as slightly wounded but not even having to leave the regiment. If anyone would be discounted it would be them.

Ralph

[Ralph J. Whitehead]

Here is a look at one small section of Bavarian RIR 8, the 4th Coy, I Battalion. This company was sent into the Schwaben Redoubt (Feste Schwaben) on 27 June and was inside the redoubt along with a company from RIR 99 at the time of the attack.

The 36th Ulster Division men managed to breach several German trench lines on 1 July and captured the redoubt, holding it until later in the day when finally forced out by German counterattacks. There was a small section of German front line still in British hands at the end of the day by the redoubt and it was not until days later when additional counter attacks drove out this remnant of British troops.

In regard to the 4th Coy 8th Bavarian RIR the fighting on and around 1 July 1916 was devastating. The total losses suffered by this company were probably in the 100% range of the total numbers of men engaged. From the Verlustlisten published well after the fighting from this period, 12 September 1916 to be exact, the losses amounted to:

2 officers

54 NCO's

200 Privates

Total 256 officers and men. This would be the company at full strength for the most part and somewhat unusual for this period in the war when most companies were smaller in size.

The breakdown of the losses is as follows:

Officers: 1 Hauptmann KIA, 1 Leutnant der Reserve as a PoW

Vizefeldwebels: 1 KIA, 3 PoW

Unteroffiziers: 5 KIA, 1 slightly wounded, 1 MIA, 9 PoW

Sanitäts Unteroffizier: 1 slightly wounded

Gefreiter: 11 KIA, 3 slightly wounded, 7 MIA, 12 PoW

Privates: 55 KIA, 2 severely wounded and died, 3 severely wounded, 50 slightly wounded, 1 slightly wounded who remained with the regiment, 45 MIA, 44 PoW (1 PoW died in captivity). I should also point out that the original reports indicated 81 MIA and 19 KIA for the Privates. This was later changed to 45 MIA and 55 KIA as new information was gathered and a correction could be made to the original reports.

In looking at the dates of known casualties 3 of the Unteroffiziers and 15 Privates were killed after 1 July (from 7 to 16 July). Also, Hauptmann Schorer as well as a number of men who were captured in the fighting at the Schwaben Redoubt were killed when their party was hit by a German shell on their way back to captivity behind the British lines so even after being captured they were not safe for as long as they were in the zone of fire.

If 3 NCO's and 15 Privates died after 1 July then it is safe to assume that some of the wounded and possibly some of the prisoners were also casualties from this period. We can identify the dates the men were killed by looking at the Ehrentafel in the regimental book. The men wounded, etc. cannot be determined from this source, instead it would take a look at the existing archive records in Munich that show each man and what happened to them in the war.

This last aspect would be difficult for most people as it would take a visit to the archives and time to look each man up. It is unfortunate these records are not in digital form and available to the public.

The original number of men listed as MIA (including Hauptmann Schorer) on the Verlustlisten was 81 but later details narrowed this to 45 men. The remainder were most likely captured or in a few cases just separated from the company, possibly wounded, etc. I have not been able to review the subsequent lists far enough into the months after September where additional details were posted. Some of the known PoW's were found out from private letters sent home from captivity whereby the names of some men were reported to their families and then to the Verlustlisten.

The details, while sketchy in some matters such as dates, do provide a window into the events of 1 July and beyond. The dates of the men killed do indicate there were survivors and the records indicate they were involved in the fighting for several weeks after the initial attack. It was one of the companies that had a high number of 1 July fatalities and gives credence to the level of fighting that has been accepted in current accounts of the battle as they relate to the Ulster Division accomplishments.

Other companies had few 1 July losses and the majority posted indicate involvement in the fighting after 1 July. This indicates to me that there was a substantial number of units available after the 1 July fighting to act as reinforcements, etc. and hold the line in the event of additional attacks, etc.

If it was possible to review all of the records for this regiment and the others (if the Prussian records had survived) then we might have a chance to accurately determine the German losses for 1 July. Even with the limitations we work under it could be assumed to some reasonable degree that if almost 30% of the men reported killed happened after 1 July then a similar number of the other losses, most likely the slightly wounded, also happened after that date. I would say based upon reviewing the records of the regiment that the PoW's were probably all from 1 July. Only additional research will narrow the margin of error.

Ralph

[Ralph J. Whitehead]

the randomness of the nature of wounds inflicted would suggest that the ratio of killed to wounded within the two classes of men should be roughly equal.

Bob

Hello Bob, I had missed this response and just went back through the thread, sorry about that. In regard to the ratios being the same do not forget that the ratio of officers to men was lower in the German army than the British. I recall reading the numbers of officers per men and it seems that as the war progressed the number of serving field officers in the German companies dropped considerably and some companies were lucky to have 2 while others had more.

Given that there were 250 or so privates and NCO's and a few officers the ratios could never truly be equal. Some companies lost 2 officers, that could represent 100% of the effective strength while 50 or so men were lost, about 20-25% of the EM strength depending upon the original numbers.

In looking at the casualty lists for each of the major German states for a period of 3 years so far I have not come across anything to suggest that the actual losses were not reported accurately and that any wound or injury most likely appeared on the returns. Don't forget that over time many of the duties performed by officers at the start of the war eventually became handled by senior NCO's in many instances.

Ralph

[wiking85]

So Ralph, you're stating that the Sanitaetsbericht is accurate and includes wounds of all types? Meaning that the British scholars claiming that lightly wounded are not included are misinformed, correct?

[bob lembke]

Hello Bob, I had missed this response and just went back through the thread, sorry about that. In regard to the ratios being the same do not forget that the ratio of officers to men was lower in the German army than the British. I recall reading the numbers of officers per men and it seems that as the war progressed the number of serving field officers in the German companies dropped considerably and some companies were lucky to have 2 while others had more.

Ralph, I understand the lower number of "real" officers. After a year or two of the war a typical company whould have half of the platoon leaders be a Feldwebel=Leutnant, Offizier=Stellvertrieter, oder Vize=Feldwebel, not a Leutnant. I am talking about a different ratio than the one that I think you are considering.

Given that there were 250 or so privates and NCO's and a few officers the ratios could never truly be equal. Some companies lost 2 officers, that could represent 100% of the effective strength while 50 or so men were lost, about 20-25% of the EM strength depending upon the original numbers.

Of course the Germans had less officers. What I meant was that, in many reports that I have seen, within the German forces, sometimes the ratio of killed to wounded was sharply different among the officers, as compared to the OR/EM. In a number of reports, say for a regiment in the attack, there might be eight officers killed and 10 wounded, for a ratio of 1.25 wounded to killed, while among the "other ranks" there might be 100 killed and 300 wounded, for a ratio of 3 to one. I did not mean that there were say 40 killed and/or wounded OR per officer in the German Army, while it was 20 to one in the British Army. That could be due to the different numbers of officers in equivilant sized units in the two armies. While the ratios of killed and or wounded between officers and men could be due to behavior, like officers leading from the front and getting hit more often, or them being recognisable, and therefore being shot with greater frequency. However, you would think that, due to the random nature of the severity of woundings, that ratio of killed to wounded should be roughly similar between officers and between ORs.

I have noticed this in several regimental histories, the sort written after the war by one or two officer veterans, not a really official history. So I don't know where the casualty figures, usually included, actually came from. Other histories seem to have similar ratios between ORs and officers. I have worked with statistics most of my life and this sort of thing sticks out to me.

I think you and others have referred to Verlusten. Was that a series or source of official casualty figures? If so, where or how were these published, and how can they be found today?

In looking at the casualty lists for each of the major German states for a period of 3 years so far I have not come across anything to suggest that the actual losses were not reported accurately and that any wound or injury most likely appeared on the returns. Don't forget that over time many of the duties performed by officers at the start of the war eventually became handled by senior NCO's in many instances.

Of course there had to be some point where a "wound" would be ignored, like a scratch on barbed wire. I know the details of all of my father's wounds, and it seems that two were ignored. Being shot by a French officer at a distance of 3" with a pocket pistol is a wound, to my mind. The other lesser wound was also not trivial, and nearly led to his death. However, my father hated the command structure of the company, and was not shy about it, and perhaps the Feldwebel returned the favor. But in reading dozens or hundreds of memoires I get the impression that minor wounds were shrugged off. (I recall Ernst Juenger's comment.) Was there a formal threshold?

Ralph

Bob

[Ralph J. Whitehead]

So Ralph, you're stating that the Sanitaetsbericht is accurate and includes wounds of all types? Meaning that the British scholars claiming that lightly wounded are not included are misinformed, correct?

From what I have been able to determine to date the information used in the Sanitätsbericht was taken directly from military records accumulated during the war and shortly afterward. These details were reported according to the practices and regulations of the period. In this matter the details and statistical findings were simply based upon current records and in my opinion do reflect an accurate picture of the losses that occurred during the war. I have not seen anything to suggest that a conspiracy to manipulate loss details was considered or completed. I have not looked at the book for some time and will have to refresh my memory on their use of statistics, etc.

From a personal standpoint I have read through thousands of lists reported during the war from all of the major German states. If Slightly/lightly wounded men were ignored in the later statistical reviews it would have been quite a feat to eliminate one of the major categories of wounded reported during the war.

I need to look at the allegations of the British authors to see just what they were trying to portray. I have not really looked at their allegations or reasoning but as this has come up on the forum before I would believe I could find out more from earlier threads. Let me take a look and see if I can find out more details. The biggest issue I see at this point is how the post-war authors arrived at their conclusions and what they used to illustrate and legitimize their methods.

Ralph

[Ralph J. Whitehead]

‘However, you would think that, due to the random nature of the severity of woundings, that ratio of killed to wounded should be roughly similar between officers and between ORs.’

Bob, I suspect that due to the random nature of enemy fire the ratios we are looking at, from whatever point of view, cannot be made separate and distinct. The simple issue of being inside a trench where a random shell falls or a rifle grenade, sniper, etc. the officers and OR should be lumped into one group and one statistic. The artificial separation of officers and OR has always struck me as odd. I understand, or believe I do, why there was a distinction between the two categories. I suppose in part it was due to the different levels of society the two occupied, the levels of esteem or should I say importance given to the officers who most likely came from the upper classes on both sides at first only later to be watered down as members of middle class families were allowed the rank of officer due to the circumstances of the time. I have not seen any true correlation of ratios between different units as to officer vs OR losses.

‘I have noticed this in several regimental histories, the sort written after the war by one or two officer veterans, not a really official history. So I don't know where the casualty figures, usually included, actually came from. Other histories seem to have similar ratios between ORs and officers. I have worked with statistics most of my life and this sort of thing sticks out to me.

I think you and others have referred to Verlusten. Was that a series or source of official casualty figures? If so, where or how were these published, and how can they be found today?’

The casualty information found in the rear of many German histories, or in the case of numerous Württemberg histories published in a separate volume were taken from the records accumulated during the war. The casualty returns were sent to a central point where they were published in a large newspaper style format six days a week excluding Sunday and major holidays.

In some cases the Ehrentafel found in the books actually mention the Verlustlisten edition, the number assigned to each list as it was printed. This was a simple system whereby the lists started at 1 and each subsequent list received the next number in order. At later times in the war the lists grew so large that in some cases the same number assigned to a particular list could be found on two editions, one just being the continuation of the earlier printing.

These lists had several major categories; the main ones consisted of the various regional aspects of the German army. The largest number and size of all was the Prussian list that included the bulk of the German units including those from Baden. Bavaria, Saxony, Württemberg all had separate lists as did Marine losses. There were other periodic lists of returning prisoners of war, attempts to identify men through personal details and effects or in some cases a photo of their possessions, a sketch of their tattoo or the initials found inside a ring, etc.

These were published by the German government throughout the war and well into 1919 before all of the losses were in print. I will try to provide a sample of an entire page if at all possible but I am not sure how it will come out at low resolution. I will try one of the photo display sites I used in the past for such purposes.

The categories of the men named included those killed, died from wounds, wounded (severely, slightly, slightly and remained with the regiment), slightly injured, severely injured (by an accident), sick, captured, missing and suicides. I believe this covers the majority of the categories I can recall from memory.

Today it is difficult to locate them. Some are held in private hands but very few I know of. Their condition is fragile as the paper is not the best quality. Others can be found at the national Archives in Washington on microfilm as these lists were sent to the U.S. on a regular basis up to the time the U.S. entered the war, then the lists stop.

By far the best source would be the major German libraries that hold the originals (or did until the microfilm copies were made when some libraries destroyed the originals) and a full set of microfilm can be obtained. There are several dozen or more full reels of microfilm that hold the accumulated lists from 1914 through 1919.

‘Of course there had to be some point where a "wound" would be ignored, like a scratch on barbed wire. I know the details of all of my father's wounds, and it seems that two were ignored. Being shot by a French officer at a distance of 3" with a pocket pistol is a wound, to my mind. The other lesser wound was also not trivial, and nearly led to his death. However, my father hated the command structure of the company, and was not shy about it, and perhaps the Feldwebel returned the favor. But in reading dozens or hundreds of memoires I get the impression that minor wounds were shrugged off. (I recall Ernst Juenger's comment.) Was there a formal threshold?’

I cannot comment on whether the wounds/injuries were counted on the lists or not without actually having looked them over. The wound from a pistol should have been reported, especially as it most likely resulted in treatment by the medical staff who would have made a record that would be added to the overall list being turned in on a 10 day cycle (in most cases as there were times the existing conditions made it difficult to complete the 10 day reporting period as required). I have not looked into the issue of exactly what triggered a casualty record and what did not. I have often wondered about this aspect of the system before but have not had the time to research it.

As a guess I would say that if a man received medical treatment from the medical staff a report was made of the incident, the nature of the wound/injury, the type of treatment, dates, etc. This would then translate to being mentioned on the 10 day casualty returns.

In looking at the records of one man who was recently found and identified near Loos his records clearly indicate periods of medical treatment, the dates and location of treatment and the ultimate account of his fatal wound. I will look into this aspect further and see if there are any sources concerning this question.

Ralph

P.S. In regard to your father if you could give me the dates of his injuries/wounds and his regimental details I will check the lists to see if he was listed or not.

[Ralph J. Whitehead]

I've come across the debate about German casualty reports not including lightly wounded soldiers in their numbers, especially in reference to the Sanitätsbericht and the Somme. However, in a scholarly article "Blood test revisited" by James McRandle and Jame Quirk, this argument is dismissed as fatuous. Yet I continually see this argument in reference to the Somme and German casualties there. In fact, the recently released book "Bloody Victory" by William Philpott, furthers this argument, stating that the Sanitätsbericht should have 30% greater numbers than is actually stated. Does anyone have any idea why this idea continues to float around and if it holds any water?

I am looking thought the Sanitätsbericht for any clues on the methods used, the numbers and sources, etc. to make sure I fully understand how they arrived at the conclusions for their various charts, etc.

Can someone give me a capsule version of the argument that the losses reported by the Germans in the book should be up to 30% higher? Thanks.

Ralph

[wiking85]

Essentially, all that I've read on the subject, light as it has been, points to an origin with official historians of the British army who state that the German Sanits does not include lightly wounded, and as a result should have 30% add to the total of wounded. I guess it is to make the Allied losses seem less in comparison.

[Jack Sheldon]

Ralph

It is all linked to the question of the lightly wounded: those who appear in the San-Bericht under the column Dienstfaehig geworden bei der Truppe; i.e all those with minor cuts, scratches and wounds, who had their wounds dressed and, therefore, were carefully recorded by the German system, but were never struck off the strength of their parent units and did not disappear down the casevac chain which ultimately led to hospitals back in Germany. This, naturally, becomes a large figure during the major battles and, as I have argued elsewhere - for example in my German Army at Passchendaele book - probably provided the raw material for the casualty figure manipulation in BOH 1917 Vol II, concerning that battle. Despite the fact that Vols I & II of the San-Bericht appear in the bibliography with some attendant waffle, the existence of Vol III which contains, inter alia, the figures for Third Ypres is not acknowledged. Nevertheless there is a suspicious similarity between the BOH assertion of 400,000 German casualties for the battle and the figure obtained by adding together the figures for the period between 1 June and 9 November; viz. KIA and Missing (killed and captured) 67,272, wounded and evacuated 149,922 and 182,396 Dienstfaehig geworden bei der Truppe, which comes to 399,590. You can make up your own mind, but I know what I think and nothing will persuade me that a casualty with a minor splinter wound dealt with on the spot, so that he remained on the strength of, say 12th Coy IR 172 and continued to do his duty with a bandage on his arm should be lumped together with his mate, standing next to him when the shell exploded, who was evacuated ultimately to Germany with his leg blown off.

Jack

[wiking85]

Ralph

It is all linked to the question of the lightly wounded: those who appear in the San-Bericht under the column Dienstfaehig geworden bei der Truppe; i.e all those with minor cuts, scratches and wounds, who had their wounds dressed and, therefore, were carefully recorded by the German system, but were never struck off the strength of their parent units and did not disappear down the casevac chain which ultimately led to hospitals back in Germany. This, naturally, becomes a large figure during the major battles and, as I have argued elsewhere - for example in my German Army at Passchendaele book - probably provided the raw material for the casualty figure manipulation in BOH 1917 Vol II, concerning that battle. Despite the fact that Vols I & II of the San-Bericht appear in the bibliography with some attendant waffle, the existence of Vol III which contains, inter alia, the figures for Third Ypres is not acknowledged. Nevertheless there is a suspicious similarity between the BOH assertion of 400,000 German casualties for the battle and the figure obtained by adding together the figures for the period between 1 June and 9 November; viz. KIA and Missing (killed and captured) 67,272, wounded and evacuated 149,922 and 182,396 Dienstfaehig geworden bei der Truppe, which comes to 399,590. You can make up your own mind, but I know what I think and nothing will persuade me that a casualty with a minor splinter wound dealt with on the spot, so that he remained on the strength of, say 12th Coy IR 172 and continued to do his duty with a bandage on his arm should be lumped together with his mate, standing next to him when the shell exploded, who was evacuated ultimately to Germany with his leg blown off.

Jack

But what about the British stats? Do their loss numbers include only those evacuated or also wounded but remained with unit? Comparing stats, though bloody minded, still gives one an idea of the course of the campaign. My understanding is that 250,000 British soldiers were killed or wounded badly enough to be evacuated. Is this correct, or does that number need to be revised?

[Ralph J. Whitehead]

Jack, Wiking, It seems that like minds......I wrote this while at work but was unable to post until now.

Thank you for the insights into the issues raised in some books. I have had a chance to review much of the material that relates to this question and unfortunately it raises more questions that should be looked at before any possible consensus can be reached.

In regard to the manner in which loss figures were used I can say that both sides are correct. In that I mean that looking at purely statistical details the numbers used by the British authors in question are correct. These were the numbers of men treated at some level from wounds, sickness and injuries as well as those who were killed. If a man was treated then it would trigger a record that would become part of the ten day loss reporting requirement. This information would pass through channels to a place of central processing where the records were kept and a Verlustlisten was published for the public based upon the documents being sent in.

On the other hand the realistic look at the men shown as casualties also needs to be addressed. While it is convenient to use the numbers as a total look at the losses the reality is that a large portion of the men shown as suffering wounds or injuries or illness were not true casualties and were either quickly returned to duty within the 10 day reporting time or simply never left the regiment. As Jack mentioned, there is a great difference between a man with a scratch requiring iodine and a bandage and a man sent to Germany minus a leg.

If you want to accurately portray the combat strength of an opposing German unit it would be dangerous to assume that every man listed as a casualty was out of action. It might look good to the effectiveness of the fighting but it was not a realistic look at the man with a bandaged hand, arm, leg, etc. who was still holding a rifle in the opposing trench.

Now, if you actually look at the wounds and injuries being reported (I am omitting those killed and known prisoners as these were not in dispute as far as I can tell) the largest category of men reported as casualties that had the least concrete detail were men listed as missing. Even the Sanitätsbericht admits that while the numbers are known the exact resolution of their status is not a fixed number and in many cases a formula was used to estimate those killed, prisoners, etc.

In real numbers however the men shown as being fit for duty after being listed for treatment of some kind (generally those taken from the wounded and sick) were either quickly returned to active service with their units or in many cases never left their unit.

Then you have different levels of medical care. Some were seen in local Field Hospitals and from what I can tell many of these were quickly returned to duty after receiving treatment and possibly a short stay out of the front line while others who had been listed as sick either recovered or were not considered seriously ill and therefore could immediately return to duty as well.

Wounds and injuries that were considered more serious ended up in hospitals further behind the front and in some cases were sent to hospitals at home in Germany. Like many countries the treatment and location depended a great deal on the nature and extent of the wound or illness. Of course there were exceptions in that some very seriously wounded were kept close to the front only because moving them might prove fatal.

In an example of a man listed as being slightly wounded but remained with the regiment I have the Feldpost letters of an NCO in RIR 119. Gefreiter Karl Losch was inside a trench on 16 June 1915 near Serre just after the French attacks against this village a short time before.

The men were under sporadic shell fire as was usual when a distant shell burst caused Gefreiter Losch to become wounded. He had his hand on his head, so picture someone resting his cheek against his hand as he looked out over No Man’s Land. A small splinter from the shell struck his pinky finger resulting in a flesh wound. The splinter traveled quite a distance it seems and had lost most of its force otherwise it might have gone through the finger, through his helmet or Feldmütze and resulted in death or a serious wound.

Gefreiter Losch sought the attention of the regimental doctor who brushed iodine on the wound and placed a plaster around it. He would not have to travel far as medical dugouts were placed throughout the lines.

On 29 June Losch wrote to his family and advised them that his wound was almost healed. On 4 July the splinter fell out and he was completely healed. It almost sounds the same as getting a wood splinter in a finger that slowly works itself out to the surface until it either becomes exposed or simply falls out as the skin heals behind it. During this period he was still on active duty with his company.

The attached Verlustlisten image indicates he was one of several men shown as being wounded with three being designated as slightly wounded, remained with the regiment. (Württemberg Verlustlisten No. 215, 8 July 1915, Page 7482).

Some of the questions being raised on this subject that I alluded to in my opening paragraph relate to the British method of reporting casualties. If we are to look at ratios of killed, wounded, missing, etc. then we need to determine if both sides reported losses in the same manner or treated losses in the same manner.

I have no idea how the British army reported casualties. How they became part of the statistics used in books since the war, etc. What do their numbers show? Were there men who required minimal treatment for wounds or slight illnesses that remained with their units or was everyone evacuated to a field hospital, base hospital, etc.? How were men reported who stayed with their unit or were returned to duty after a brief spell of treatment? I believe we need to look at the methods employed by both sides to see if they match in any way so that we can talk oranges to oranges as they say and not apples to oranges.

As it stands now I see why there is a controversy regarding German losses. You could either use the statistics as simple numbers without looking deeper into the categories and actual status of the men as it seems many British authors have in the past or as Jack and I look at the losses, in their true nature. While it is comforting to see larger numbers of enemy killed and wounded it is not as comforting to the man in the trenches facing their opponent who should be out of action and being treated some distance from the front instead of shooting back at him.

In any future issue regarding losses the methods being used must be explained in detail and every aspect of these numbers, from simple figures to the story behind the categories must be explained so that the reader can make a determination of the facts for themselves.

I would be very interested in seeing the British method and how the numbers were produced to determine if there is any correlation with the German methods and to see if the number of British losses rises or falls as the categories and actual disposition of the men is known.

Ralph

[Ralph J. Whitehead]

Here is a prime example of a group of men after a successful raid against a sap. All I can say at present is that the regimental number worn by the men is '15' and that could include a number of regiments.

Their clothing is torn from the action and probably barbed wire, some are bandaged and a number sport recently awarded Iron Cross awards. What was their status: Slightly wounded, slightly wounded but remained with the regiment, etc. While injured it appears that they are still in the active regimental zone and not evacuated to the rear. Most look ready to defend themselves if needed.

The biggest thing brought up in the last threads was that we actually know the systems on both sides of the wire and that when comparing details that they match in their presentation and calculations. We need to know more of the British method, anyone care to help?

Ralph

[phil andrade]

There can be no doubt that the assertions of James Edmonds in the BOH regarding German casualty statistics are preposterous and pernicious.

They are preposterous because they are without merit : there is abundant evidence that German statisitcs, whether in the Sanits ( forgive my abbreviation), or in the Verluuste, make ample allowance for all categories of wounded.

They are pernicious, because they are wilful distortions, and, sad to say, have been readilly accepted by many historians.

It's all very well me making these claims, so I will offer evidence to substantiate them.

For Verdun, between February 21st and September 10th 1916, the Sanits tabulate German casualties :

Killed 41,632; Missing 26,739; Wounded 241,860. This indicates that nearly six were reported as wounded for every one killed. Of course, many of the missing - indeed, most of them - were also killed, which must be taken into account. Compare this with the official French casualties for the battle ( in this case up until mid December 1916):

Killed 61,289; Missing 101,151; Wounded 216,337. In this case, the French record only three and a half men wounded for every one killed ( albeit with the same proviso that a significant proportion of the missing were also killed).

In the light of these tabulations, how can we believe that the German casualty reports in the Sanits were not allowing for lightly wounded ? Indeed, there is a suspicion in my mind that the reverse might have been the case, with the French excluding slight wounds from their records, while the German statistics make full acknowledgement.

To elaborate on this, consider the German Sanits tabulation for the entire Western Front for the whole of 1916:

Killed 142,223

Missing 153,049

Wounded 896,876

Note that the wounded outnumber those posted as killed by 6.3 to 1. Are we to accept that another three or four hundred thousand should be added to the wounded ? If so, Allied weapons were very ineffective at killing Germans !

British casualties for the Western Front in 1916, according to the Medical Statistics:

Killed 107,411 Missing 43,675 Wounded 500,576

In this case, then, the British report 4.66 wounded for every one posted as killed.

It's clear that the Germans reported a far higher ratio of wounded to killed than did either the British or the French; a fact that absolutely contradicts the contention that German casualty records in the Sanits are understated due to the exclusion of slight wounds in the reports.

The Verluuste ( again, apologies for mispelling) also endorse the comprehensive reporting of all wound categories in the German casualty figures. Here is the figure published as at Ocober 31st 1917:

Killed in action/Died from wounds 1,138,768

Died from sickness/accident 85,088

POWs 387,979

Missing 263,043

Severely Wounded 652,021

Wounded 328,421

Lightly Wounded 1,829,820

Wounded remaining with units 315,263

In the light of these German records, how can we trust Edmonds when he asks us to believe that German casualty statistics need to be increased by a third, or even a half, in order to allow for categories of wounded that were excluded in their records but included in Allied reports ? For the Somme, we are asked to inflate the German total from 437,000 to 650,000. For Passchendaele, 200,000 becomes 400,000. This is just absurd : wishful thinking at best, outrageous distortion being the reality. The same sleight of hand is not applied to the battles fought before Haig assumed command.

The BOH account of casualties for Loos, for example, is quite content to accept the German record.

All in all, I find it astonishing that so many historians have accepted this nonsense that Edmonds has bequeathed us.

Phil.

[salesie]

I had intended to keep out of this debate, Phil, because, as you already know, I find this kind of statistical analysis (in isolation) to be completely irrelevant, in that it reduces the study of the Great War to nothing more than an exercise akin to football scores. But after your latest post I feel the need to make a few points.

What is it those of you who persist in this search for the statistical "Holy Grail" of German casualty stats wish to prove? Is it that the German army consistently inflicted greater casualties on the allies than they took? If so, it has to be said that if you do indeed prove this case then the point will arise that it didn't do the Germans much good. Or do you wish to prove that Edmonds was influenced by Haig when compiling the BOH - or wish to prove that the BOH is inherently flawed, full of myth backed by statistical sleight of hand? If so, it has to be said that if you do indeed prove this to be so, how will you account for the fact that, despite scoring more "goals" than the allies, Germany actually lost?

That said, your own figures in the previous post open up speculation. Now, I have no intention of getting into a detailed statistical argument with you, but it seems to me that you need to re-calculate your assessments for the ratios of killed to wounded - for example, you say that the vast majority of missing were in fact killed, so if you add killed and missing together then divide that figure into the number you give for wounded then you will see a different set of ratios to the ones you give.

You will also notice that the numbers of missing vary considerably between the sets of figures you give. The British missing figures for 1916 are considerably less than the number killed, for the French at Verdun the missing are actually more than those killed. But the most important variations, in the context of this debate, are the German figures you provide i.e. for the Germans at Verdun (Sanits) the missing are considerably less than those killed, yet for the whole of 1916 on the Western Front (Sanits) they're considerably more, but the starkest oddity is in the Verluuste figures, here we see that the missing figure when compared to the killed figure is less in proportion to any of the other sets.

I'm no statistician, Phil - but for the life of me I can't see how can you draw any firm conclusions from such sets of wildly varying figures, figures which open up more questions than they answer? However, as I said earlier, if you add the missing figures to the killed figures and divide that into the wounded returns then different ratios to yours emerge (I'll leave you to do your own calculations and draw conclusions).

As I've said in previous threads in this vein, I have no desire to defend Edmonds, Haig or anyone else - but, as per those other threads, I will make three relevant points:

1) British Military Intelligence, throughout the war, had pretty accurate assessments of German manpower wastage without the Germans knowing it i.e. the paybook carried by every German soldier being the main source of this information, but there were several others.

2) Edmonds may well have been privy to this "secret" information when writing the BOH, and would certainly not have divulged his true source when doubting the official German casualty figures. If so, his statistical argument would have simply been a "smoke screen" at the time to cover the secret intelligence operations - but now that these secret papers have been released into the public domain then the actual British Intelligence assessments can be examined.

3) The Sanits figures were published under the Nazi regime, and, by definition, need to be treated with lots of suspicion.

Where is this statistical search leading, Phil, what do you ultimately hope to prove?

Cheers-salesie.

[Ralph J. Whitehead]

As I've said in previous threads in this vein, I have no desire to defend Edmonds, Haig or anyone else - but, as per those other threads, I will make three relevant points:

1) British Military Intelligence, throughout the war, had pretty accurate assessments of German manpower wastage without the Germans knowing it i.e. the paybook carried by every German soldier being the main source of this information, but there were several others.

2) Edmonds may well have been privy to this "secret" information when writing the BOH, and would certainly not have divulged his true source when doubting the official German casualty figures. If so, his statistical argument would have simply been a "smoke screen" at the time to cover the secret intelligence operations - but now that these secret papers have been released into the public domain then the actual British Intelligence assessments can be examined.

3) The Sanits figures were published under the Nazi regime, and, by definition, need to be treated with lots of suspicion.

Where is this statistical search leading, Phil, what do you ultimately hope to prove?

Cheers-salesie.

I am glad the discussion has expanded as more eyes looking at the details should provide a greater understanding of the similarities or differences between the different parties.

I understand how the German Soldbuch was a source of intelligence but a flawed source. Pure numbers provided to recruits coming into a unit might be interpreted as a measure of the numbers of men who were casualties. However, there were often times when men were transferred, whole companies were used to form new regiments, etc. so that while a particular unit might have men assigned high numbers in reality the men they replaced were simply serving in another regiment or arm of the service. This was very prevalent during the expansion periods of the army when new regiments and divisions were being formed as the structure of divisions was changed from the standard 4 regiment set up to one of 3, the surplus units being combined to form new divisions. There was one instance I read where an entire company of veteran troops was assigned to form a new regiment only to be transferred a short time later to a third regiment. I could not imagine the paperwork involved in tracking the company rolls for these men.

You mentioned other sources used by British Intelligence. Could you expand in some detail on these sources? It would be interesting to see their techniques and evaluations and compare them with other details from this period. I look forward to hearing more on this part.

I have never heard of any reference to Edmonds and secret sources. Is there a recent article, book, etc. that could provide additional details on these methods. While the British side of the war is not my main research area it is always good to examine all facts and information before providing an opinion. The more we know the better we understand. Please, let me know where I could look for this, thanks.

The final bit is somewhat suspect. I agree that in books published under the Nazi Regime there is a need to look at the author, the facts, etc. for some were obviously written toward a specific goal. I suspect however that this work, the 3 volumes of the Sanitätsbericht might not fall into this category. From what I can tell so far the work on the statistical volume 3, published in 1934, the year Hitler became the leader of Germany, the year after he was Chancellor under Hindenburg the work on this book and the others had already taken more than 2 years, possibly more so it appears it was started and primarily completed under the Weimar Republic. I have seen no indication that there was a manipulation of facts by the authors as suggested. If you have further details that could point to a different conclusion please let me know as I would like to know if my details are tainted in any way.

I did find it odd that while there are authors who question the German losses, but not in all cases as the Australian Medical Services, the British War Office and Winston Churchill all found the German details credible. There is also a book apparently published in 2002 I believe by an author named Mosier where he contends that the statisticians who wrote Statistics of the Military Effort of the British Effort during the Great War 1914-1920, published in 1922 has been basically ignored by later writers and historians and apparently the Official History went to great lengths to contradict the data provided by their own government.

I also understand that while this sort of discussion is not something you would find interesting others do and as such the more factual details we can all provide would help all who do find this interesting understand all of the influences that apparently interfere with the records from all major armies.

Still, we can argue who fudged the loss details (the U.S. Government apparently did from what I have read so far) but I am more interested in the whole picture of the war. As such any new insights into the other intelligence sources, the secret details, etc. would be great to read about in more detail. Any references to articles or other sources regarding this information would be great to hear about, thanks.

Ralph

[Jack Sheldon]

Sitting here, almost 100 years on, we can afford the luxury of being, in varying degrees, completely dispassionate about the cost of the Somme and the Great War more generally. Salesie often reminds us of the outcome and implies, thereby, that the end justified the means. Certainly that is the viewpoint of William Philpott, whose new book Bloody Victory should be compulsory reading for all with an interest in the Somme. Much as it sticks in my throat to say so, (because the human cost continues to trouble me) he undoubtedly makes the case for attrition on the Somme as an essential precursor to ultimate victory; a view I am sure would chime with Salesie's thoughts.

However, let us not forget that work on the 1916 volumes of the BOH was conducted between 1929 and 1938, when the situation was quite different. To anyone with an interest in this question, I recommend you obtain a copy of Writing the Great War: Sir James Edmonds and the Official Histories 1915-1948 by Andrew Green (ISBN 0-7146-5495-7, or 07146-8430-9). This is a book conversion of his doctoral thesis and is extremely readable, in noted distinction to most such work. The following points are taken from his Chapter 4.

At the time of preparation, there was a spectrum of views concerning the Somme offensive. 'These variously viewed the Somme as a wasteful and misguided campaign in which the flower of British manhood was lost for nothing more than a few yards of ground, or as part of a calculated and predetermined plan to wear down the German forces by a deliberate war of attrition which would take years to reach a successful fulfilment. A vital consideration in any assessment was whether, as a number of authors have claimed, the cost in terms of lives and casualties was far heavier on the part of the aggressor than it was on the part of the defender.' p 61

'The issue of casualty figures was one which genuinely troubled Edmods... the diligence with which he investigated the matter belies the suggestion that he manipulated them for ulterior reasons...' p 75

'The British returns were a record of absentees from a roll call carried out the day after an engagement. Only those definitely known to be killed or wounded were reported as such and a number of the 'missing' subsequently appeared with their units. This method naturally produced an excessive amount of 'missing' personnel and relatively few dead. In order to obtain more accurate figures it was decided to examine the battalion Part II orders which 'sooner or later' accounted for every man. To do this for July 1 alone took a member of the staff of the Historical Section six months, whereupon it was stopped for 'reasons of economy'. The results of this investigation were that the original returns for 1-7 July were too high. The German methods, on the other hand, were based either on returns made in the field three times a month or on casualty lists published for the entire army during the war... The former method excluded those wounded dealt with in hospitals in the corps area and who would, under the British method, have been listed as missing. The latter by contrast, included men so lightly wounded that they had not left their units. Edmonds concluded that to arrive at a sensible comparison with the British treatment of casualty figures one would need to add something to to the first set of figures or deduct it from the second. This resulted in an indeterminate figure somewhere between 437,322 and 582,919. In the end Edmonds concluded that at the Somme 'the Allies appeared to have lost rather more than 600,000, the Germans rather less than that total.' pp 75-76

I have quoted this in full because it is the most charitable view of Edmonds' work on casualties I have ever read. Chapter 4 actually deals with 1916 Vol I and therefore skirts round dubious alterations to the figures generated sometime over the next few years - which were highly controversial and challenged publicly at the time. By the time Vol II appeared, 'revised' figures of German casualties appeared, claiming that they were now 660,000-680,000. Here I can only recommend and urge that all interested contact the Royal United Services Institute in London and request copies of two articles by MJ Williams. The first appeared in February 1964 (I have not read it, but its threads are repeated in the later article). It destroys the basis of the claim that 30% needs to be added to German casualty figures. From the 1966 Journal, you should read The Treatment of the German Losses on the Somme in the British Official History: Military Operations France and Belgium 1916 Volume II.

I am not going to attempt to summarise Williams' work but, in a scholarly, densely written article, he dismantles, takes apart, demolishes Edmonds' work on casualties: 'The method by which 680,000 was obtained is demonstrably fallacious...' 'The revised statistics then are completely invalid...' 'General Edmonds' methods in using German casualty figures here and throughout his preface to 1916 Volume II have a strong resemblance to Humpty Dumpty's' use of words in 'Through the Looking Glass...' 'The British Official History of the Western Front, 1914-18, was in some respects superior to the French and German Official Histories. Unfortunately its treatment of German losses on the Somme is unscholarly and unreliable...'

In the 1930s the debate was emotionally charged; there was clearly a perceived need to get the reading public to accept the counterintuitive view that the defence suffered more than the attack - despite all that was known about the vulnerability of troops in the open to artillery fire and direct fire from men sheltering in even rudimentary trenches or shell scrapes. It was also attended by many other subsidiary issues raised by critics, such as the quality of generalship, a cavalier attitude to human life etc. Now it remains valid and interesting because it assists us in the attempt to make objective judgements, for example, about the true place of the Somme and Verdun in determining the tipping point of the war: hence this and other discussions.

Jack

[phil andrade]

You will also notice that the numbers of missing vary considerably between the sets of figures you give. The British missing figures for 1916 are considerably less than the number killed, for the French at Verdun the missing are actually more than those killed. But the most important variations, in the context of this debate, are the German figures you provide i.e. for the Germans at Verdun (Sanits) the missing are considerably less than those killed, yet for the whole of 1916 on the Western Front (Sanits) they're considerably more, but the starkest oddity is in the Verluuste figures, here we see that the missing figure when compared to the killed figure is less in proportion to any of the other sets.

Cheers-salesie.

What a comprehensive rebuff of my efforts, salesie !

You make pertinent and cogent observations, though, and it's apparent that you've read my post carefully, and I must thank you for that.

On no account imagine that I'm intent on seeking any formula that extolls the qualitative superiority of the Germans. The same sources that I've cited give ample evidence of the enormous casualties inflicted on the Germans by the Anglo-French armies. Here are Sanits tabulations for German losses in the Michael Offensive, for example:

Killed 35,163

Missing 22,701

Wounded 181,694

These are for the period March 21st to April 10th 1918 only, and allude specifically to the forces deployed in the principal attacks. They are horrific, and, in killed and wounded especially, tower above those suffered by the Allies. No attempt at concealment here, surely...just a stark indication of the cost of taking the offensive in a war that gave lethal advantage to defensive firepower.

You raise a significant point regarding the varying proportion of the missing. This is accounted for by the fact that prisoners are included in the number reported as missing. At Verdun, for example, most of the 100,000 plus French recorded as missing were prisoners ( the Germans claimed to have captured 66,000 by mid July), whereas, as of September 10th 1916, fewer than 10,000 German prisoners were claimed by the French. Look at the proportion of missing in those German figures for their spring 1918 attacks that I've posted above....relatively small numbers of prisoners were lost to the Allies in that period, and it's probable that two thirds of the German missing were killed. At the same time, the British reported "only" about 14,000 killed in the same fighting, but admitted to vast numbers of missing, of whom the majority were prisoners. Forgive me if I post another Sanits tabulation - it's hard to digest these statistics - but here is its compilation for the Somme from June 21st to November 26th 1916:

Killed 57,987

Missing 85,683

Wounded 273,132

Again, the high proportion of missing compared with killed reflects the heavy loss of prisoners, who probably accounted for nearly three quarters of the missing.

All in all, the Allies claim to have captured 120,000 German prisoners on the Western Front in 1916. This might be an exagerration, but if we allow for 100,000, it implies that roughly one third of the 153,059 missing reported by the Sanits for that year in the West were killed, and two thirds prisoners. The British figures for France and Flanders for 1916 post 43,675 missing; this is then broken down into 15,516 prisoners, and 28,159 who were subsequently presumed killed....a reversal of the German proportion, and an indication that the majority of the attackers' missing are killed, while the majority of the defenders' are prisoners. When the Germans attacked in 1918, then the majority of their missing were killed: at Verdun in 1916, most of the defending French missing had been captured. I hope that this suffices to answer your point about the varying proportions of the missing.

It distresses me that you might imagine that I seek to deploy these statistics in an attempt to argue that the Germans were always more adept than their Entente counterparts. I seek no such vindication of the German methods : as you say, they lost.

What intrigues me is the way in which statistics are used, and, more significantly, abused, by historians. The ciphering of the Great War is notorious, and, to be fair, there are examples - Mosier and Laffin come to mind - when the abuse is blatantly distorted to give an account in Germany's favour. I do not think that the Sanits, despite the Nazi rule, ever sought to distort figures. It is, I reckon, a didactic work which attempts to provide a meticulous record of how the armies were confronted by death, wounds, sickness and injury in all their guises. There are certainly disparities between the figures compiled therein and other official German records : the number posted as killed in the Sanits does indeed seem too low, especially when compared with huge numbers of wounded. But from what I've seen of volume III it is apparent that the authors acknowledge this disparity, and seek to explian it. One thing is very clear : the number of wound cases reported is very high, and certainly refutes Edmonds's contention that a notional increase in published German statistics needs to be added in order to allow for lightly wounded and make them comparable with British or French figures.

A final word - I do not see the record of millions of human beings suffering death or mutilation as something like a football score.

Phil.

[salesie]

Ralph, details of British Military Intelligence's assessments of German manpower wastage can be found in Armour Against Fate, British Military Intelligence in the First World War, 1989 Michael Occleshaw, ISBN 0-86287-407-6. In this book, about a much neglected aspect of the war, you can find the author's own summary of this intelligence work along with a detailed list of his sources, sources which should give a dedicated researcher such as yourself plenty to look at.

Occleshaw makes the point that,"They, (BMI German manpower wastage assessments) should provide a stock answer to those who glory in the persistent argument that German losses were nearly always far lighter than the British." And please note that he is critical of Charteris' GHQ assessments being too high because of its reliance on German paybooks alone, but he also points out that once these were brought into line in 1917 with the War Office's methods, which used other information as well as paybooks (though still a valuable source), then GHQ's assessments became much more accurate. But, read the book, look at Occleshaw's sources and judge for yourself.

I have no idea whether Edmonds was privy to this secret (at the time) information or not, and therefore put up a statistical "smoke screen" in the BOH to cover his main source in casting doubt on the accuracy of German casualty returns - I am simply pointing out that there is information available in the public domain these days which makes this a possibility, and that this "new" information from British intelligence sources does point to German official casualty stats being inaccurate. As for you never hearing reference to Edmonds and secret sources, that doesn't surprise me, nor should it surprise you - after all, they wouldn't have been very secret if everyone knew about them.

Also, I admire your faith that the Nazi regime never "interfered" with the Sanits publication in 1934 - but, I'm afraid, you'll need more than blind faith to convince me. It seems to me, stating that something published under the Nazis can be taken at face value is an oxymoron - just a cursory glance at the figures posted by PJA throws up suspicion.

Anyway, Ralph, as I said earlier, I have no intention of getting deeply involved in this debate (been there before), I'm just adding a bit of balance by pointing out that when it comes to casualty stats nothing is certain (for either side).

Cheers-salesie.