OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

H A IG’S EN EMY
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Crown Prince Rupprecht of Bavaria

OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

1
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

3
Great Clarendon Street, Oxford, ox2 6dp,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Jonathan Boff 2018
The moral rights of the author have been asserted
First Edition published in 2018
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2017953093
ISBN 978–0–19–967046–8
Printed in Great Britain by
Clays Ltd, St Ives plc
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Acknowledgements

I could not have completed this book without the help and support of too
many people to list here. I am immensely grateful, first, for the encourage-
ment and support, moral and practical, that I have received from friends and
colleagues both at the University of Birmingham and within the history
community more broadly. The generosity I have encountered has been
inspiring. Nick Crowson, Elaine Fulton, Sabine Lee, and Corey Ross helped
create the space for me to research this book, and my fellow foot-soldiers in
the War Studies trenches, past and present, kindly took up the teaching slack.
Pete Gray has been a good friend and a sage mentor to whose wisdom
I owe much. Members of the Centre for War Studies, most notably John
Bourne, have offered inspiration and education in equal measure at our
seminars. I am grateful to the participants at a range of seminars and confer-
ences to whom I have presented aspects of this work over the years, for their
comments and suggestions; and to those who took part in the 2014 and
2016 British Army staff rides to the Western Front, from whom I learnt
much. I am especially grateful to Holger Afflerbach, Tony Cowan, Robert
Foley, Markus Pöhlmann, Christian Stachelbeck, and Dave Zabecki, all of
whom shared their expertise in the history of the German military with a
generosity and patience which was humbling to this novice. My students,
inevitably, have taught me the most: I thank them for that and for turning a
job into a joy.
Special thanks to Professors David French, David Stevenson, and Sir
Hew Strachan, who have been especially supportive in multiple ways, most
of them beyond the call of duty.
Without the help of archivists, not much history would get written. I am
grateful to the staffs of the following archives and libraries: in Britain, the
British Library, Imperial War Museum, London Library, National Archives, and
National Army Museum; in Germany, the Bundesarchiv-Militärarchiv in
Freiburg im Breisgau, and, in Munich, the Bayerische Staatsbibliothek
and Bayerisches Hauptstaatsarchiv, both Abteilung IV (Kriegsarchiv) and
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

vi Ack now le dge m e n ts

Abteilung III (Geheimes Hausarchiv). Special thanks at the last of these to

Dr Gerhard Immler, a prince among archivists, and the very patient Herr
Leipnitz.
Thanks to Barbara Taylor once again for her excellent maps. Others who
helped me in specific ways are thanked at the appropriate place in the Notes.
Several friends were kind enough to read drafts: thank you to Holger
Afflerbach,Tony Cowan,Aimée Fox, Sabine Lee, Corey Ross, Gary Sheffield,
Andy Simpson, and to the anonymous readers for Oxford University Press.
Thank you for the keen attention to detail you brought to the task and for
saving me from howlers. This is a better book for your insights and sugges-
tions: the shortcomings that remain are my fault.
Luciana O’Flaherty first approached me about publishing with Oxford
University Press, and I am grateful to her, Matthew Cotton, Martha
Cunneen, Carrie Hickman, Kizzy Taylor-Richelieu, and the whole team
there for their encouragement, patience, tact, and efficiency. Thanks also to
Sally Evans-Darby for the speed and accuracy of her copy-editing.
When it briefly looked as though this project might be derailed, the pro-
fessionalism and skill of Andy Chukwuemuka, Diana Holdright, Susan
Horsewood-Lee, and the staff of the Cromwell Hospital kept it on track.
I owe them more than I can say.
Friends and family have tolerated the absences, both physical and mental,
of a distracted author with good humour and patience which I didn’t always
deserve. This applies especially to my wife,Yuko. This book is hers as much
as mine: I might never have started it, and certainly couldn’t have finished it,
without her love and encouragement. It is dedicated to her, and to my god-
children, Phoebe, Jack, Tom, James, and Luke, with love.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Contents

List of Illustrations ix
List of Maps xi
Maps xii

Introduction 1

PA RT I.  TO WA R 1914
1. Rupprecht’s Road to War 11
2. The Battle of the Frontiers 17
3. The End of the Campaign in Lorraine 27
4. The First Battle of the Somme 35
5. To Ypres 41

PA RT II.  T H E A N V I L 1915–16
6. A Difficult Winter 57
7. A Successful Spring 68
8. Further Victories 83
9. Verdun and the Road to the Somme 92
10. Early Days on the Somme 100
11. Rupprecht the General 108

PA RT III.  HOLDI NG T H E LI N E 1916 –17

12. Rupprecht Takes Command 121
13. Autumn on the Somme 134
14. Scorched Earth 144
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

viii Con t e n ts

15. The Battle of Arras 156

16. The Battle for Flanders: Summer 1917 168
17. The Battle for Flanders: To Passchendaele 178
18. Cambrai 190

PA RT I V.  Y E A R OF DE F E ATS 1918

19. Planning the Spring Offensives 201
20. Operation MICHAEL 210
21. Operation GEORGETTE and Summer 1918 220
22. The Hundred Days 233
23. Rupprecht on the Run 247

PA RT V.  CONCLUSIONS
24. Rupprecht the Field Marshal 257
25. Rupprecht and Politics 269
26. Last Words 277

Appendix: Note on Military Terminology 285

Notes 289
Bibliography 329
Picture Acknowledgements 343
Index 345
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

List of Illustrations

2.1. Rupprecht with his two sons before the war 20

5.1. Rupprecht setting out for a morning ride 48
8.1. Kaiser Wilhelm II visiting Rupprecht’s headquarters 84
11.1. Rupprecht and his son Albrecht  112
12.1. Rupprecht and his staff 122
13.1. Hindenburg with Ludendorff 135
14.1. Rupprecht inspecting his troops 148
15.1. The front-line trenches of the Hindenburg Line (Siegfried Stellung)
near Bullecourt 159
20.1. German troops in the streets of Bapaume, March 1918 213
22.1. The demonstration of 7 November 1918 on the
Theresienwiese, Munich 245
23.1. Crown Prince Rupprecht and Princess Antonie with their
family between the wars 249
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

List of Maps

1. General Key for All Maps xii

2. Europe and the Contending Alliances, 1914 xiii
3. The Western Front, 1914–18 xiv
4. Lorraine 1914 xv
5. Somme 1914 xvi
6. Flanders 1914 xvii
7. Artois 1915 xviii
8. Somme 1916 xix
9. Arras 1917 xx
10. Flanders 1917 xxi
11. Cambrai 1917 xxii
12. MICHAEL 1918 xxiii
13. GEORGETTE 1918 xxiv
14. The Hundred Days xxv
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Maps

XXXXX
BR British Army Group

XXXX
Bav Bavarian
Army

XXX
FR French
Corps
Australian and New Zealand
ANZAC Army Corps XX
Division

AUS Australian
German
CAN Canadian

Infantry
Gds Guards

Cavalry
NZ New Zealand

Res Reserve (-) Elements of unit

Ter Territorial

Railways

Canals

1. General Key for All Maps

 Key
500 miles
500 km
Triple Alliance
N
Central Powers
NORWAY
Neutral
SWEDEN
Bavaria (within
Germany)

OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

North (1915) Date of joining war
© BAT 2016

Sea DENMARK

GREAT
BRITAIN
RUSSIA
Atlantic NETH.
GERMANY
Ocean BEL.

FRANCE AUSTRIA–
SWITZ. HUNGARY

ROMANIA
(1916) Black Sea
AL

SERBIA
UG

ITALY BULGARIA
SPAIN
RT

(1915) MONT.
(1915)
PO

ALBANIA
OTTOMAN EMPIRE
Mediterranean Sea (1914)
GREECE
(1917)
MOROCCO ALGERIA

2. Europe and the Contending Alliances, 1914

 Zeebrugge HOLLAND Key
Antwerp N
Ostend Bruges Limit of German advance
er BELGIUM in September 1914
Ghent
ov
Nieuport e
eld Approximate front line
D Sch Aachen
of
Dunkirk Diksmuide Louvain from end of 1914
r
s Yse
r ait Flanders Lys Courtrai
BRUSSELS Liège Allied gains during 1916
St Calais Ypres and 1917

OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Hazebrouck use
Me Spa German gains during 1918
Boulogne Lille Tournai
re Namur
Loos
Lens mb Charleroi Armistice line on 11
Mons Sa November 1918
Artois Douai Maubeuge
Arras
Cambrai 30 miles
Le Avesnes 30 km © BAT 2016
Cateau
So LUXEMBOURG e
mm Albert Oise sell
e Peronne Mo
St Quentin Trier
Amiens Mézières Sedan
Dieppe Chemin GERMANY
Picardy Laon des Dames Aisne

Me
use
Compiègne
Aisne
Rouen Soissons
Reims Verdun Metz Lorraine
Ves

ine
ise
le Champagne Morhange

Rh
Se O St
ine Château- Épernay M
Meaux ar Mihiel
Thierry ne Strasbourg
e Nancy
rn
Ma Petit Morin

Meu
Grand Mor
PARIS in Alsace

se

M
ne

os
Sei

ell
e
Epinal

ges
Troyes Freiburg

Vo s
F R A N C E
Mulhouse
Basel
Belfort

3. The Western Front, 1914–18

 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

LUXEMBOURG 30 kms
30 miles N
e
ell
os
M

Sa
Thionville

rre
XXXX GERMANY
6
Rupprecht
LORRAINE
Metz
St Avold Sarreguemines
XXX
Seille

II
XXX Bav
III Morhange XXX
Bav
XXXX
XXI XXX
2 I
Castelnau Bav
Vic
BASSE
ALSACE
Dieuze Sarrebourg XXX

ine
Toul Lagarde

Rh
Nancy XXX
XV

er
Riv
(-)
XIX
Cirey Strasbourg
Lunéville
Manonviller Vezouze

Badonviller

Charmes
Mo

XXXX
sell

Me
s
e

ntain

urt

FRANCE Épinal
7
he

Heeringen
Mou

XXXX
ges

1
Vos

Dubail Colmar

Thann

Key Mulhouse
International boundary Belfort
Approximate limit of ine
HAUTE r Rh
French advance,
ALSACE Rive
evening 19 August 1914
Montbéliard Basle
Approximate line of
maximum German
advance, SWITZERLAND
s
ub

25 August 1914
Do

© BAT 2016

4. Lorraine 1914
 Gommecourt brai

Arra
N Cam

s
Bapaume
Miraumont

Nord
Beaumont-
Hamel

al du
Thiepval

OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Can
Pozières
Longueval
la-Boiselle Morval
Fricourt
Albert Combles Moislains
Mametz
Maricourt
e
ncr Curlu
er A
Riv Bray-sur-
Somme
Péronne
River S
omme
Flaucourt
Corbie
Santerre Plateau
Warfusée-
Abancourt
Amiens
Villers-
Bretonneux
St Quentin
Lihons
Luce
River

Rive
r So

l
na
mm
Riv

Ca
er

e
ise

tin
Avr

rO

en
e

ve

Qu
Ham Ri

St
Nesle

Roye

10 miles
Montdidier 10 km © BAT 2016

5. Somme 1914
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

N XXX Forest of
Houthulst XXXX
XXIII Res
4
XX Württemberg
Ypre Canal XXX
89 Ter FR
s—C

XXVI Res ROULERS

om

XX Bixshcoote Poelcapelle
ines

87 Ter FR Langemarck
XX
Passchendaele
1
Boesinghe Pilckem St
XXX Julien XX Gravenstafel
XXX
Elverdinghe I 2
Zonnebeke Broodseinde
St
Jean
XXVII Res
Vlamertinghe XX XXX
XX
Poperinghe 17 FR
YPRES IV 7 Terhand
Zillebeke
Gheluvelt
XX
Dickebusch
3
Zandvoorde Gheluwe Lys
St Eloi River
MENIN
XX
XXX 2
Wytschaete
Kemmel
Cav Mount XX
Kemmel Wervicq XXX
Messines
1 Comines
V XXXX
XXX
Wulverghem
Neuve
Warneton
XXX IV 6
XX
Eglise Rupprecht
Bailleul 4 I
Ploegsteert
XXX XXX
TOURCOING
Frelinghien
III XIX
Nieppe
ROUBAIX
Armentières
Lys
er
Riv

Bois
Estaires Grenier

Laventie
LILLE

Fromelles

Aubers

Neuve
Chapelle

Festubert Seclin

Givenchy
LA Phalempin
BASSÉE
Key
3 miles British positions on 21 October 1914
3 km British positions on 23 October 1914
© BAT 2016

6. Flanders 1914
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Béthune
La
Bas
sée LA BASSÉE
Can Givenchy-lès-
al
la-Bassée

nal
Cuinchy

eule Ca
Haisnes Douvrin
Annequin

Haute D
XXXX Sailly

1 BR
N Haig
Vermelles Hulluch

Noeux-
les-Mines

Mazingarbe
Loos
Hill 70
Grenay

Bully-
les-Mines
ez
Souch
River

Aix
LENS
Lor
ett
e Angres
Spu Notre-Dame
r XXXX de Lorette XXXX
165 m
10 FR 6
d’ Urbal Rupprecht
Ablain- Givenchy-
St Nazaire en-Gohelle
Souchez
V 140 m

Carency
im
y
Vimy
La
Folie
Neuville- Farm
Farbus
Mont- St-Vaast R
River St-Eloi id
Scarp g
e Thélus
e
The
Labyrinth
Bailleul

Écurie Roclincourt
Maroeuil

Anzin
St St
Nicolas Laurent
River
Scarp
e

ARRAS
Feuchy
Danville

Tilloy
Key
Allied front lines at start of
Beaurains
Second Battle of Artois
Allied front lines by 5 miles
13 October 1915 5 km © BAT 2016

7. Artois 1915
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Gommecourt 5 miles
XXXX 5 km
Hébuterne
3
Allenby
3
Serre
N
XXXX XXX
Res XIV BAPAUME
Res XXX XXXX
Stein
Beaumont-Hamel Gd Res
Warlencourt 1
Butte de Marschall
St-Pierre-
Warlencourt Below
XXXX Divion
Le Sars
Courcelette Le Transloy
Res Gueudecourt XXX
Gough Thiepval
Martinpuich XXII
High Wood
Lesboeufs Res
Pozières Kirchbach
Res Bazentin
XXXX Wood
Delville Morval Sailly-
Ovillers Wood
Sallisel
4 Longueval Leuze
Wood St-Pierre-Vaast
La-Boisselle Trones
Wood Wood
Mametz Wood
ALBERT Montauban
Fricourt Combles
Bernafay
Mametz Wood Rancourt
Bouchavesnes XXX
Maricourt XXVII
XXXX Res
cre BEF X Fas-
An
XX bender
4 XX FR Curlu Cléry

le
To La
Rawlinson

rtil
Mt St
Quentin
Bray-sur- Frise
Feuilleres Halles
Somme
PÉRONNE

Cappy
Somme Flaucourt
XXXX Fontaine- Becquincourt
Lès-Cappy 6 FR Barleux
6 XXXX XXX
Fayolle 10 FR Belloy
Proyart XVII
Foucaucourt Estrées Fleck

Berny Villers-
XXXX Soyécourt Carbonnel
Deniécourt
10 Genermont
Micheler Vermandovillers
Som

Ablaincourt
me

Key XXXX
Pressoir
Front Line on:
1 July 1916 Lihons
Chaulnes 2
10 July 1916 Gallwitz
28 August 1916 XXX
30 September 1916 IX
Chilly Quast
31 October 1916
15 November 1916
XXXX Army boundary
XXXXX BEF/French army
boundary Nesle
© BAT 2016

8. Somme 1916
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

XXX XX
N Key
6 LENS Front line: 9 April 1917
I Liéven
XXX Front line: 27 May 1917
XX I Bav
Res XXXX Army boundary
Lor XX
ette 79 Res
Méricourt
Spu 24
r The Pimple
Souchez Hill 119
XXXX XX Givenchy Drocourt
4 CAN
Hill Vimy Acheville
1 XX
145 XX
DOUAI
Horne 3 CAN
Fresnoy
XX 1 Bav Res Quiéry
XXX Farbus Arleux
2 CAN
Vi
m

XX
CAN Oppy
yR
idg

1 CAN
Bailleul pe
e

1
car
XX
X rS
XXX 3
XX
e
Marœuil
51 XX Gavrelle Riv
14 Bav
XXX 34
XX
Vitry XXXX
XVII 9
XX Athies Fampoux 6
Rœux
15 Falkenhausen
XXX XX XXX
ARRAS XX
Sailly
12 Feuchy
VI IX Res sée
XX
11
Monchy-le- en
3
erS
XXXX XX Tilloy XX Preux Riv
14
17 Res
3 XX
Guémappe
Allenby 56
Wancourt Dury
Neuville Vis
Wailly Héninel
XX
St Martin CAM
Ficheux 30 XX Fontaine BRA
XX I
l

Hendecourt
jeu

18 Res
Co

XX
XXX 220 Cagnicourt
Boyelles
er

Riencourt
Riv

21
Croisilles
VII Bullecourt
XX
Adinfer 62 elle
XX
Quéant nd
3 X iro
rH
XX XXX Ecoust 4 e
AUS Riv
Ayette X 5 Noreuil
V
Ervillers
Mory Lagnicourt
Vaulx- Boursies
Ablainzeville Vrauqcourt XXX
Morchies
XXXX I ANZAC
Beugnâtre Doignies
5
Beaumetz
Gough BAPAUME
5 miles
5 km © BAT 2016

9. Arras 1917
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Key
N St
Forest of Houthulst
Allied Lines:
ee
nb XX Morning 31 July 1917
ee
k Evening 31 July 1917
2 Gd Res
20 September 1917
7 December 1917
Yes-

Formations shown at time of initial assault.

ER DE
Ypre

E X MU
Koekuit

N
PP X X IX
XX
s Ca

Bro

D
YP
enb

E
Bixschoote eek

PP
50 Res
nal

RU
RU
G
G
XXXX Steenstraat
Lizerne XX Poelcappelle
1 FR 111 Langemarck
Anthoine Het Sas

1 FR XX Stroom

St
X beek
XXX Passchendaele

ee
Pilckem

nb
5 Gds Kerselaere XX
XX

ee
Boesinghe Kitchener’s

k
Wood Gravenstafel 221
3 Gd St
XXX XX Julien
XX Tyne Cot
XIV 38
Elverdinghe 23 XX
XX

51 XX 235
Broodseinde
XXX 39
Wieltje XXXX
XVII Zonnebeke
Frezenberg
Brielen
St
XX 4
Jean 55 von Arnim
XXXX XXX
XX
XX
Potijze
Vlamertinghe 15
5 XIX 38
Westhoek Polygon
Gough XX Wood
8 Becelaere
YPRES Hooge
Zillebeke XX Sanctuary
Lake Wood XX Veldhoek
30
GRU
Zillebeke GRPPE
52 Res UPP X X YPER
X
GheluveltEW X N
XXX YTS
53 18 CH
XX AET
II Klein Herenthage XX E
Zillebeke Woods Me
24 nin
119 Roa
d
Dickebusch Shrewsbury
XX Kruiseecke
Vormezeele Forest
Dickebusch
Lake XX 22 Res
5 XXX
XXXX Zandvoorde
2 St Eloi 41
X
XX
XX
Vierstraat 19 Hollebeke
XXX 12
XX
IX
Oosttaverne 10 Bav

XX Houthem
Wytschaete XX
37
Kemmel Yp
XX re 207
s-
Mount XXXX 18 Res Ca Com
Kemmel na in
l e s
2 Garde-Dieu
Messines Gapaard COMINES
Plumer
Wulverghem XX
ys
3 AUS er L
River Douve XXX Riv

I I ANZAC GRUPPE
XX
Warneton WYTSCHAETE
Neuve Eglise Hill 63 NZ XXX
GRUPPE LILLE

2500 yds
Ploegsteert Wood © BAT 2016
2500 m

10. Flanders 1917
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Inchy

Bourlon Bourlon Wood

CAMBRAI

Mœuvres Fontaine-
Notre-Dame XXXX

XX Cantaing 2
XX
XXX
56 20 LW Graincourt
um e XIII
Bapa

XX
36 Flesquières Rumilly
Marcoing

Hermies Masniéres
Havrincourt Ribécourt
XX Crèvecœur
l
na

54
d
Ca

or

XX
N
du

l
62 Trescault na
XX Ca
tin
XX en
Havrincourt 9 Res Qu
St
Wood 51 XX
XXX 6 La
IV Vacquerie
Metz-en- III XX
IV
Couture 20
XX Banteux
Gonnelieu
29
Gouzeaucourt XX
12
XXXX
XXX III Villers-
3 Gauche VII Guislain
Wood
Byng III

Epéhy Vendhuille

Key
British front 20 November 1917
British gains 20 November 1917
Lines of Hindenburg/
Siegfriedstellung position
Limit of British advance 5 miles
5 km
XXX Corps boundary © BAT 2016

11. Cambrai 1917
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

XXXX 6
XX N
6 XX17
Lens Quast R Scarpe
XXXX
Vimy Douai
1
Horne 1 XXXX
XXXX
3 17
Below

e
eld
ARRAS

ch
Sensee

RS
Canal

17 XXX
X 2 XXXX
Croisilles CAMBRAI
2
Marwitz
Doullens Flesquières
XXXX Masnières

3 Bapaume
Byng Gouzeaucourt
R Ancre

Miraumont Ytres

St Q
Villers-
du N al
ord
Can
St-Pierre Guislain

uent
Vaast
Albert Wood

in Ca
2
XX
Bray XX 18 XXXX

nal
Sailly-
Laurette PÉRONNE
AMIENS 18
3 Bellenglise Hutier
XXXX Hamel
4 Villers-
XXXX Bretonneux Tertry
4 Beauvois
Rawlinson Chaulnes ST QUENTIN
RS

BEF
om

XXXXX
Rouy-
me

Moreuil
FR le-Grand
St Simon
Nesle Cr
Ham oz
at
Roye Ca
na
R Avre l
rre
R Se
ye
R No

Cantigny Montdidier
Chauny La Fère
XXXX

3 FR
Humbert Barisis XXXX
Noyon
7
Boehn
e

Key R
R Ois

Ai
let
Start line 21 March 1918 te
23 March 1918
27 March 1918
Final position 5 April 1918 25 miles
25 km
XXXX Army boundary © BAT 2016

12. MICHAEL 1918
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Key XXXX
Forest of Staden
German Lines: 4

Yse
Houthulst

r-Yp
9 April 1918 Sixt von Armin

res
10 April 1918 XXXX

Can
11 April 1918

al
12 April 1918 BE
25 April 1918 Albert Steenstraat
Poelcappelle
Formations shown at time of initial assault.
Langemarck
Passchendaele

3 miles N
3 kms GIA
BEL XXX
XX F
N BE

Vlamertinghe
YPRES
Poperinghe
Hooge Becelaere

Gheluvelt
Dickebusch

Steenvoorde
Scherpenberg
Godewaersvelde

Yp
res
La Clytte

–C
Mont des Mont
Wytschaete

om
Cats Rouge Kemmel

ine
XXXX Locre Mount COMINES

sC
Mont Kemmel

an
Noir Dranoutre
2 BR

al
Messines
Plumer Wulverghem
Caëstra

Neuve Eglise Ploegsteert Warneton

Meteren Wood
BAILLEUL

Strazeel

4
HAZEBROUCK XXXX
6
Steenwerck

ARMENTIERES
Forest of Vierhouck Bac-St-
Nieppe Maur
2
Neuf-Berquin
Lys

XXXX
er

1
Riv

Estaires
Fleurbaix Bois-
Merville Grenier

St Venant
Radinghem
Fromelles

Robecq Aubers
Vieille Chapelle
Neuve Chapelle XXXX
Fournes
Locon 6
XXXX Hinges
l
na

Quast
Ca

1 BR
le
Lil

Horne
Don
Festubert
aB
Aire-L assée Ca
BÉTHUNE nal
Givenchy La
Bassée
Cuinchy
© BAT 2016

13. GEORGETTE 1918
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

XXXX Courtrai N
Ypres Menin
2
XXXX
Hazebrouck BELGIUM
4
XXXX Roubaix

He sitio
8 Aug

Po
Armentieres

rm n
XXXX

an
9 Sept Lille Tournai

n
5
XXXX
Béthune
6 Mons
XXXX e
a rp re
Lens
XX
XX Sc mb
Sa
1 Valenciennes
30 Aug Maubeuge
Douai
XXXX Forest of
9 Sept Mormal
Arras
17
XXXX Le
Cambrai
25 Cateau
3 Aug Havrincourt
Bapaume Avesnes
XXXX
XX XXXX
XX
4 Épehy
Albert 2 Guise Oise
Péronne
Amiens
F R A N C E
Som

9 XX
18 XX St
me

9 Sept Quentin La Serre

8 Aug Aug Aug
XXXX
Hu
Roye nd
18 XXXX ing
XXXX 10 Aug 9 Sept La Fere
-Brunhild
Positio
Noyon XX 9 n
1 FR XX
Laon XXXX
XX

Montdidier 22 Sissonne
XX

21 Aug
8 Aug Aug 9 Sept 7
XXXX

e
XXXX Aisne Aisn 1
Compiègne
XXXX
3 FR se
Oi Soissons Vesle
10 FR
Reims
Key
Wotan Line
e Épernay
Hindenburg/Siegfried Line Château- Marn
Thierry
Armistice Line 11 November
30 miles
XXXX Army boundary 30 km © BAT 2016

14. The Hundred Days

OUP CORRECTED PROOF – FINAL, 19/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Introduction

C rown Prince Rupprecht of Bavaria first came alive for me a few years
ago. I was working on a book about the last few months of the war on
the Western Front and had gone to Munich to explore the archives of the
Bavarian army. These somehow escaped the Second World War firebombs
which destroyed many of the Prussian records and as a result constitute one
of the best sources for information on the German army during the First
World War. On the last day of my trip, as I was idly flicking through a folder
of apparently random documents, one sheet of paper caught my eye. It was
a passport, issued by the Dutch embassy in German-occupied Brussels,
allowing the Marquis de Villalobar and his party to enter the Netherlands
on 12 November 1918. What was it doing in the personal files of Crown
Prince Rupprecht of Bavaria?
I already knew a little about Rupprecht. I knew, for instance, that he
served as one of Germany’s most senior generals between 1914 and 1918
and  spent most of the war fighting against the ‘tommies’ of the British
Expeditionary Force (BEF), leading the German Sixth Army, and later an
army group of one and a half million men, in famous battles such as Ypres,
the Somme, and Passchendaele.1 It was Rupprecht’s men who launched
Germany’s last-gasp offensives in March 1918, and when they failed it was
his army group that felt much of the weight of Allied counter-attacks dur-
ing the so-called ‘Hundred Days’ campaign in August–November 1918.
Rupprecht’s closest British counterpart and most consistent enemy was
probably Sir Douglas Haig. Like Rupprecht, Haig rose to command an army
group: the BEF; and everywhere Haig and the BEF turned from October
1914 on, Rupprecht and his men were in the way. A good parallel from
the  Second World War might be the mythical rivalry between Bernard
Montgomery and Erwin Rommel, although Haig and Rupprecht, who
never met and were in no sense personal rivals, would have despised any
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

2 I n t roduct ion

propaganda of personality of that kind. What did the Marquis de Villalobar

have to do with Rupprecht, though, and why was he travelling to neutral
Holland the day after the Armistice had been signed?
In the early days of November 1918 Germany was in double crisis. At
home, Socialist agitators were whipping up revolution, pushing crowned
heads from their thrones and seizing power. Among those deposed were
Kaiser Wilhelm II and Rupprecht’s father, King Ludwig III of Bavaria. At
the front, the famed discipline of the German army was breaking down
after a series of heavy battlefield defeats. Some units were already setting up
Russian-style soldiers’ councils (‘soviets’). The people of Brussels, where
Rupprecht had his headquarters, cowered behind locked doors as gangs of
drunken German soldiers rampaged through the streets, looting and worse
wherever they could. Rupprecht’s life might be in danger. Less than four
months before, after all, Bolsheviks had pitilessly murdered the Russian Tsar
and his entire family. The Spanish ambassador to Belgium, Rodrigo Marqués
de Villalobar, came to Rupprecht’s rescue. He offered sanctuary in the
embassy and made arrangements to smuggle him, disguised as plain Mr
Landsberg, to safety in the Netherlands. That was what the passport was for
and that explained why it had found its way to the archive in Munich.2
Answering one question, however, raised others. How was the proud heir
to an ancient throne, a man of consequence in Germany and across Europe,
a field marshal who had led hundreds of thousands of men in battle, reduced
to skulking across borders under cover of darkness and a false name? This is
the first question this book seeks to answer. It offers the first biography of
Crown Prince Rupprecht in English, concentrating on his military career.
He has had several German biographers but they have tended not to be
specialists in the history of warfare. Only a couple of scholarly articles in
German have addressed this aspect of his career before now.3 As we will see,
the human story is a fascinating, moving, and possibly tragic one. It also,
however, raises important questions with broader ramifications than the
purely individual. Studying Rupprecht’s military career allows us to inter-
rogate and challenge several of the preconceptions we have about the First
World War and about modernity itself.4
We can, for instance, see the war from a new perspective if we look
through Rupprecht’s eyes. Too often we have been told the story of the
Western Front from the British point of view alone.5 The French, and even
the Germans, often fade into the background or even out of the picture
altogether. This book aims to help correct this distortion and to give the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

I n t roduct ion3

proper weight to all three armies, enabling us to understand the war in a

more balanced way.
We can also explore important questions about the war. One of the key
ones is: why did Germany lose the First World War? Inevitably, a range of
factors played a part. Germany did not lose the First World War only because
her army was defeated on the Western Front. Events on other fronts and at
home, some of which had little or nothing to do with Rupprecht, played a
part in determining the outcome. So there are limits to what studying him
can tell us. For instance, his contribution to grand strategy, the level of war
where politicians set overall aims and allocate resources, was, as we shall see,
limited. He did, however, know all the key decision-makers and was
extremely well informed. Consequently, he is a useful source of information
about the direction of the war. We will see that the German leadership
made strategic mistakes at every stage. Few historians would disagree with
that. The more interesting question is: why?
Rupprecht himself spent the war fighting on the Western Front, working
at the operational level of war. This is where generals construct and fight
campaigns, design and sequence battles, allocate resources and manage
reserves to achieve the mission they have been set by the strategists. Below
the level of operations lies that of tactics, where soldiers fight their battles.
Get the tactics right, and you win battles. Get the operational art right, and
you win campaigns. Get the strategy right, too, and you win wars. The
Western Front was one campaign of a larger war and for both sides winning
there was necessary, even if it was not sufficient, for victory. Control of
France and Belgium was precisely what many thought they were fighting
for in the first place and it was here that both sides deployed their most
powerful forces. The focus of this book, therefore, is on the operational level
of war on the Western Front, where seeing events through Rupprecht’s eyes
can be very helpful.
This is an aspect of Germany’s war which historians have tended to
neglect, concentrating instead either on the view of the supreme com-
mander or of the individual soldier in the trench.6 Germany has even less of
a tradition of operational military history than British academia.7 In neither
English nor German has any historian yet followed a single senior German
general, and the formations he commanded, through the whole war as this
book does. Doing so shows how the German military reacted to the stresses
of conflict and the need for rapid and sometimes radical change. Many mili-
tary historians tend to assume that the German army of the first half of the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

4 I n t roduct ion

twentieth century was highly skilled tactically, largely because it operated a

flexible system of command known as Auftragstaktik (‘mission command’)
which delegated authority and ensured that the man best placed to make
any decision was the one issuing the orders. Other strengths included high
levels of training and motivation. However misguided or downright evil the
purposes to which the army was put, they argue, it constituted a fine tech-
nical instrument. Here I challenge that interpretation and suggest that the
German army of 1914–18 had structural weaknesses and conceptual short-
comings which inhibited innovation, distorted decision-making, and
eventually helped destroy it. Some of those weaknesses were unique to the
military, but others carried over from broader German society.8 Studying
armies as institutions in their own right is interesting but when, as in the
two world wars, they are mass citizen forces, they open a window onto
society as a whole which allows analysis of the military to take on wider
significance.
If historians have sometimes overlooked the operational art of the German
army, the same cannot be said of the BEF, which has been studied under the
finest of microscopes. One of the most influential and persistent myths in
history, first sketched out by Winston Churchill and David Lloyd George and
enthusiastically developed and promulgated by Basil Liddell Hart and others,
is that which sees the Western Front primarily in terms of British failure.9
The traditional narrative assumes that what went wrong on the Western
Front did so because the generals were too stupid, or the army too hide-
bound, to adjust to the modern reality of barbed wire and the machine gun.
The commemorations of the war’s centenary have repeated the stereotypes
encapsulated in Oh! What a Lovely War and Blackadder more often than they
have challenged them, showing how deeply seared into the public con-
sciousness this myth is. The gulf between popular perceptions of the war and
recent academic research has never been wider, because since the 1980s a
generation of British and Commonwealth military historians have been
mining the archives to revise that traditional view.They have dissected almost
every aspect of the BEF in minute detail and argue that its leaders were not,
on the whole, fatheads and knaves who never learnt a thing. Instead, the
British army did remarkably well to overcome the immense challenges it
faced and adapt to radical change in the most trying conditions, by marrying
the best civilian expertise with military know-how to climb a metaphorical
‘learning curve’, until by 1918 it was capable of fighting the toughest army in
the world toe-to-toe—and beating it.10
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

I n t roduct ion5

The ‘learning curve’ idea captures the central truth that the British army
of 1918 was certainly more effective than it had been even a year or two
previously. It is not without problems, however, largely stemming from
exploring the BEF in isolation without sufficient comparative research into
the French and German armies. This is understandable given the mass of
detail involved in the study of just one army, much less three, even before
taking into account the language barrier. Nonetheless, this risks making
British learning look exceptional. Also, it can exaggerate the British achieve-
ment by ascribing too much of the BEF’s success to its learning curve and
taking insufficient account of any deterioration in the German army. One
of my core contentions here is that it is a mistake to see the war as a one-
sided (British) struggle to learn a new way of fighting and break the stalemate.
Although the front was deadlocked for much of the war, it was so only
because there was a continuous dynamic of push and pull, measure and
counter-measure, between the two sides. This produced a deceptively stable
equilibrium. This book argues that the battlefield defeat of the German
army in 1918 was at least as much the result of the Germans getting worse
as it was of the British getting better. It therefore challenges us to revise our
views of the BEF, which was neither a collection of ‘butchers and bunglers’
nor an institution made up exclusively of forward-thinking quick learners,
but somewhere between the two.
So, Rupprecht’s war enables us to see the war and the armies which fought
it in a fresh way. It also prompts us to think again about the relationship
between the First World War, modernity, and the agency of the individual.
One of the things that first attracted me to Rupprecht was that he seemed to
stand for a generation which was destroyed by the events of 1914–18. By the
time the war finished, his eldest son and many of his friends were dead. He
had lost his throne and, instead of leading a triumphant homecoming parade,
he sneaked back into Bavaria incognito. The threat of trial for war crimes
hung over his head for years. Never again would he enjoy true power or
influence. Throughout the 1920s and 1930s he watched as political violence
convulsed his fatherland. Under Hitler, he lost his home, his family, and
nearly his life. Like the cosmopolitan aristocrats of central Europe so vividly
evoked in Patrick Leigh Fermor’s travel books, Rupprecht can be seen as one
of the many relics of a bygone age swept away by the tsunami of modernity.11
How far does such a romantic picture correspond to reality, though?
Seeing Rupprecht as a victim fits neatly into popular perceptions of both
the First World War and modernity itself. Most of us first encounter the war
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

6 I n t roduct ion

at school, through poets such as Wilfred Owen and Siegfried Sassoon. The
worm’s-eye view they offer inevitably drains meaning from the existence of
the soldiers they depict. Unable to discern the larger patterns which make
sense of their lives—and deaths—of course the war seems bewildering and
futile. These poets were keen to reject what they saw as the twee conserva-
tive pastoralism of Edwardian Britain and so they eagerly highlighted the
impersonal, the mechanical, and the modern in ways which have further
influenced our view of the war. In Owen’s poem, the Western Front is
where, amid the monstrous clamour of modern industrial warfare, ‘doomed
youth’ are herded to their deaths with all the passivity of cattle. Similarly, it
is a mark of the insignificance of the individual in All Quiet on the Western
Front that the novel’s ironic title derives from the official bulletin issued on
the day its everyman hero dies. In other self-consciously modern depictions
of the war, such as C. R. W. Nevinson’s painting ‘La Mitrailleuse’, soldiers
have lost their humanity altogether. Flesh and steel fuse into war-fighting
machines. The fact that we concentrate our commemorations today on
memorials to those who were killed, not those who survived, only rein-
forces the victimhood motif. Also, to view the participants as victims rather
than agents has the advantage of being psychologically easier on us.12 The
victimhood we ascribe to the soldiers of the First World War is, therefore,
multi-faceted: it combines inevitable and pointless death with a sense of
alienation from a modern world increasingly dominated by the machine,
where humans have shrunk to insignificance and have no control over their
own destiny. For all the efforts of some historians to correct it, this percep-
tion of the ordinary soldier as victim remains firmly dug in.13
Even if it is correct that ‘the soldier in the trenches can be characterized
as the victim of modern warfare par excellence’, does that apply equally to all?
Were there active subjects, as well as passive objects, on the First World War
battlefield? To whom can we ascribe moral responsibility?14 Was Rupprecht
entirely caught up by forces he could neither understand nor control? Did
he conform to the romantic stereotype? Or was what happened to him the
consequence of his own decisions?15
The sources this book draws on are mostly new to an English-speaking
audience. I have worked through a range of archival collections, primarily
in Germany. First, Crown Prince Rupprecht kept a war diary, which is pre-
served in the Bavarian Household Privy Archive in Munich. It extends to
4,197 handwritten pages and was mainly put together as the war went on.
There are two exceptions: he reconstructed the first few months of 1914
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

I n t roduct ion7

from notes over the following winter, and he also had to re-write the last
week of the war, having burned the original pages before going on the run.
This manuscript diary is full of detail, not only about what was going on in
his sector but also about the wider war. The entry for a typical day, 1 October
1917, for example, covers: his tactical and logistical situation (one page); an
intelligence report concerning a recent top-level Anglo-French meeting to
discuss offensive plans (another page); diplomatic developments (one page);
and the domestic situations in Britain, France, Russia, and the United States
of America (nearly six pages). During the 1920s Rupprecht’s generalship in
August and September 1914 attracted criticism in the early volumes of the
German official history. The retired staff officers at the Reichsarchiv in
Potsdam who were compiling it blamed errors he made for the failure of
the Schlieffen Plan. To put his side in the long-running controversy which
ensued, Rupprecht decided to publish a version of his diaries. He asked
some of his former staff officers to read and check the manuscript, which
was then edited down to a quarter of its original length and published in
three volumes in 1929. Much of the material cut out was repetitive or, like
the sometimes lengthy domestic and foreign press summaries Rupprecht
used to write up, not relevant. On the whole, the published diary maintains
the sense of the original closely, although on occasion Rupprecht was ­unable
to resist tweaking it to make himself look better, as we shall see. He also
toned down, or left out completely, some of his criticisms of individuals.
Some historians have drawn on the published diary, but few have bothered
with the manuscript version, or with other unpublished papers including
Rupprecht’s correspondence with his family, friends, and politicians.16
Another major source was the official records of the formations Rupprecht
led: Sixth Army and Army Group Crown Prince Rupprecht. These are held
in the Bavarian State War Archive, also in Munich, and cover most aspects
of both day-to-day operations and longer-range policy in considerable
detail.17 They relate not only to Bavarian formations but also to all those
under his command, so they give us a broad view across the German army.
The after-action reports compiled by units at all levels after battle, in an
effort to learn lessons for next time, proved particularly valuable. These
papers were supplemented by other official documents from the federal
military archives in Freiburg-im-Breisgau, where any surviving records
from Potsdam are stored, along with other private papers belonging to
Rupprecht’s associates and subordinates. Published primary sources include
the fourteen-volume German official history, Der Weltkrieg, the Bavarian
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

8 I n t roduct ion

official history, Die Bayern im Großen Kriege 1914–1918, the series of battle
monographs compiled by the Reichsarchiv between the wars, German regi-
mental histories and officers’ memoirs, and, of course, the voluminous
French and British official histories.18
This book is divided into five Parts. The first four tell the story of
Rupprecht’s war from the beginning through to the end. Seeing the Western
Front through German eyes in this way invites us to review existing, largely
British, narratives of the battles fought there. We will discover, for instance,
that battles which loom large in British military history sometimes hardly
even feature as footnotes in the German equivalent. This also enables us to
explore some of the broader themes already mentioned. Prominent among
these is how the character of warfare evolved and the different armies adapted
to change. Many chapters conclude with a section on this strand. Another
recurring theme is the nature of command on the First World War battlefield.
Such themes help explain both the war’s outcome and the underlying
strengths and weaknesses of the forces fighting it. The fifth and final Part
assesses Rupprecht as military commander and political actor and offers
answers to some of the questions raised in this Introduction. An Appendix
on military terminology seeks to guide readers who want help telling the
difference between a corporal and a colonel or a company and a corps.
The war which emerges from this book is one of radical and dynamic
change where the ability to out-think your enemy was just as important as
being able to out-muscle him. It took both sides time and immense effort
to adapt to the new ways of war, not least because every time they came
close to an answer, the enemy changed the question. Eventually, however,
the German army was dragged down by its inherent weaknesses and fell
behind in the race to adapt.The results were disastrous. Sometimes Rupprecht
and his colleagues were constrained in the choices they could make, but the
decisions they took were nonetheless too often poor. The First World War
was not just a war of bodies and a war of machines: it was a war of brains,
too. The Germans lost all three.
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

PaRt I
To war 1914
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

1
Rupprecht’s Road to War

P rince Rupprecht Maria Luitpold Ferdinand was born in Munich shortly

before nine o’clock on the morning of 18 May 1869.1 He was the new-
est member of the Wittelsbachs, the dynasty which had ruled Bavaria for
nearly 700 years. Rupprecht was the first of thirteen children born to Prince
Ludwig and his wife, the Habsburg Princess Maria Therese. Rupprecht was
related to most of the crowned heads of Catholic Europe: not only the
Habsburgs but also the Braganzas, the house of Luxembourg, and even the
Stuarts. Indeed, some saw his mother as the rightful Stuart heir, tracing her
ancestry back to Charles I. When Rupprecht visited London for Queen
Victoria’s diamond jubilee in 1897, members of the Jacobite Order of the
White Rose tried to pay homage to him as the future King Robert I and IV.
Rupprecht never took this claim seriously, but his very name had a Stuart
link: King Charles I’s nephew and famous cavalry commander during the
English Civil War, Prince Rupert of the Rhine, was a Wittelsbach ancestor.
Rupprecht was not born to the throne. The King of Bavaria in 1869 was
his great-uncle, Ludwig II. Just twenty-three years old, he appeared per-
fectly vigorous and there seemed no reason why he should not live long and
build a family of his own. In fact, however, Ludwig’s enthusiasms lay else-
where. He was Richard Wagner’s patron and was more interested in building
fairy-tale castles than in continuing the royal line.The towers of his castle at
Neuschwanstein later inspired Walt Disney. Known as ‘Mad King Ludwig’,
he died in murky circumstances in 1886, childless and unmarried. His
brother Otto succeeded to the throne. Unfortunately, while there was room
for doubt about Ludwig’s mental health, there was none in Otto’s case. He
was indisputably insane. He had already been locked away for years and
remained so until he eventually died in 1916.The uncle of Otto and Ludwig,
and grandfather of Rupprecht, Prince Luitpold became the real ruler of
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

12 To Wa r 1914

Bavaria as regent. When he died, in 1912, Rupprecht’s father took over,

finally setting Otto aside and declaring himself King Ludwig III the next
year. Thus, at the age of forty-four, Rupprecht became heir to the throne
and acquired the title by which he was known for the rest of his life: Crown
Prince of Bavaria.
The throne which Rupprecht should one day inherit was a constitutional,
rather than absolute, monarchy. The King was supposed to stand above
­politics and rule by consensus with his government and the two houses of
the Bavarian parliament, the Landtag. In reality, he wielded considerable
power. Although from 1906 the Lower House was elected by universal male
suffrage, the King dominated the Upper House, appointed and dismissed
the government at will, and dominated policy.2 The biggest constraint on
Bavarian royal power was external, rather than internal. In 1871 Bavaria
became part of the new imperial Germany. She was now just the second
state among many, lying far behind Prussia in importance. Of the total
German population, over 60 per cent were Prussian, while only a little over
10 per cent came from Bavaria. Prussia had the power to veto constitutional
change on her own whereas Bavaria could only do the same by building a
broad coalition of the smaller states. That said, the government in Munich
retained considerable powers and rights, including control over direct
­taxation. In peacetime it remained responsible for raising and running its
own  army. Officers pledged allegiance to the Bavarian King, rather than
the German Kaiser. The army, however, was built on Prussian lines and in
the event of war the Kaiser would become Supreme War Lord of all the
German armies. Most crucially of all, the decision to declare war belonged
to the Kaiser alone.
The first half of Rupprecht’s life saw dramatic change for all Germany,
with rapid population growth, a rush to the cities, and huge industrial
expansion. Bavaria was affected too, but differences in the patterns of growth
and development across the new country exacerbated pre-existing regional
tensions. For example, although the population of Munich grew six-fold
between 1850 and 1910, towns such as Essen in the Ruhr expanded more
than thirty times over. Bavaria remained the least urban of the major
German states, and invested less in industry and scientific research. Overall,
Bavaria could not match the pace of growth in the north. In 1870, the
population of Bavaria had been 4.8 million. By 1910, it had increased by
40  per cent, to 6.9 million. This was huge growth in a short period, but
remained well below the German average of 60 per cent.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t’s Roa d to Wa r13

Bavaria retained deep cultural roots in its agrarian past. Seventy per cent
of the population was Catholic, with 28 per cent Protestant. In Prussia, the
religious proportions were almost exactly reversed. At the risk of generaliza-
tion, Bavaria remained relatively agricultural, Catholic, and conservative
even as Germany as a whole was becoming increasingly urban, Protestant,
and socially progressive. Even her conservatism, founded in the values of the
smallholder farmer, was very different from the Junker- and industrialist-
dominated conservatism of the imperial ruling class. All these differences,
new and old, created fault-lines which divided Bavaria and Prussia. Even in
peacetime, the relationship between the two was not always easy. Prussians
tended to treat Bavarians with condescension, while the latter were prickly
about anything which seemed to be eroding their rights and privileges. As
we shall see, wartime only made these tensions worse.3
Rupprecht later complained that his childhood was a strict one and to
modern eyes his upbringing seems indeed to have been harsh. It is not clear
that it was particularly so by the royal standards of the time, however. As a
young child he saw his parents only at mealtimes, had lessons daily from
eight a.m. until five p.m., and was frequently beaten, both by his short-
tempered father and by his tutor. In the finest tradition of German royalty,
his relationship with his father remained difficult throughout his life. It is
perhaps telling also that he reacted against the religious atmosphere of home
in later life: at least as a young man his Catholicism did not go much beyond
observing the proprieties. At thirteen he became the first member of the
Bavarian royal family to be educated at a public school. He graduated from
high school in 1886 and was commissioned into the Regiment of Foot
Guards. He later also spent time serving in artillery and cavalry regiments.
He seems to have enjoyed to the full the relative freedom of life in the sub-
alterns’ mess, as well as the perks of being a young royal prince. Alongside
his military duties he spent several terms studying at universities in Munich
and Berlin. He also began his public career. For instance, he attended Kaiser
Wilhelm I’s funeral in March 1888.
When fully grown, Rupprecht was rather taller than average, at five feet
ten inches. He took immaculate care over his appearance and carried him-
self with the upright bearing of the natural athlete. In his youth he had
enjoyed swimming and skating and throughout his life remained a devotee
of physical exercise and gymnastics. Well into middle age he could sit down
on the ground and get up again without using his hands or arms. He was a
keen sportsman, in the old, aristocratic sense: nothing so vulgar or English
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

14 To Wa r 1914

as team games for him. Instead he enjoyed game birds and hunting, both in
the Alps and abroad. In India he killed a tiger from the back of a moving
elephant with his second shot. Like most members of the upper classes of
the time, he rode often. Even during wartime, a daily ride remained part of
his exercise regime.4
In 1891, while serving in the 1st Heavy Cavalry Regiment, Rupprecht
began going to lectures at the Bavarian War College (Kriegsakademie),
where officers hoping to join the elite General Staff underwent three years
of intensive training. He attended some of the classes but was excused the
strenuous exams. The aim, after all, was not to produce a professional staff
officer, but to teach him command and leadership. In 1893 he was promoted
to command a squadron as Captain (Rittmeister). He rose rapidly through
the ranks until in 1900 he was promoted to Major-General (Generalmajor)
in charge of the 7th Infantry Brigade at Bamberg. In 1904, as a Lieutenant-
General (Generalleutnant) he became General Officer Commanding (GOC)
the 1st Division and soon thereafter took over the 1st Bavarian Corps. He
was ­promoted to full General (General der Infanterie) in 1906. He seems to
have taken his military duties seriously and to have been a respected and
able commander of men.
Away from his military career, Rupprecht was well travelled. Italy proved
a special and enduring love. He went there almost every year of his adult
life. He knew most of Europe well but had also journeyed extensively
around the Mediterranean and Middle East, Asia, and America. He even
wrote travel books. Some of his travel, of course, was for royal business.
For example, in 1911 he attended the coronation of King George V, writing
to an artist friend:
Time and again I admired the English race and its unique culture. The
­celebrations were great, but also absolutely exhausting. It is extraordinary how
England is on the one hand so modern yet on the other so exceedingly con-
servative. I wished you could see the coronation in Westminster Abbey with
me, the picture was so artistic, a real symphony of colours.5
He was a man of culture and taste. Although his reading was ‘intense and
critical’ and he enjoyed music (but not Wagner), his passion was for art. He
earned a reputation as a connoisseur and many of his friends were artists or
collectors. His first enthusiasm was for the Italian Renaissance, but he was a
keen patron also of modern artists.
His private life can only be described as tragic. After a playboy youth,
Rupprecht decided that, once he passed his thirtieth birthday, it was time to
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t’s Roa d to Wa r15

settle down. In the course of 1899 he began to court his twenty-year-old

cousin, Marie-Gabriele. She was receptive enough but her parents were less
convinced.When they sent her off on a trip to Verona and Florence with an
aunt, Rupprecht pursued her. Travelling under a pseudonym, he somehow
managed to get past the chaperone, proposed, and was accepted. After a furi-
ous family row, Rupprecht finally secured the approval of his father and
grandfather and the couple were married on 10 July 1900 in the court
chapel of the Residenz, the old royal palace in Munich.
As far as we can tell, the marriage seems to have been a love match. Of
five babies, however, one was still-born and another two died in infancy.
Marie-Gabriele died in 1912, aged just thirty-four. Rupprecht was left alone
with their two boys and his career as a soldier. He threw himself into
his work with redoubled energy. In 1913 he became a Colonel-General
(Generaloberst) and commander designate of the largely Bavarian-manned
Sixth Army in the event of war.
On 28 June 1914 Archduke Franz Ferdinand, heir to the thrones of
Austria-Hungary, was assassinated in Sarajevo. Since he had close family ties
with members of the Bavarian royal family, and had visited them only two
months previously, the news came as a great shock in Munich. Over the
weeks that followed, crisis mounted as the powers of Europe jostled for
power and prestige.The Bavarian royal family and government played a pas-
sive role as onlookers. Although they had good contacts in both Berlin and
Vienna, and remained well informed throughout, they made no attempt to
rein in Germany’s reckless brinkmanship during the July crisis. The sympa-
thies of King Ludwig III and his Austrian queen lay, unsurprisingly, with
Vienna and Berlin and in early July Bavaria intervened to block an attempt
by the state of Württemberg to question the handling of the crisis in the
federal Upper House foreign affairs committee.
By the last week of July passions were starting to run high across Europe.
Outside the royal palace, crowds gathered to sing patriotic hymns and shout
slogans against the Serbs and Russians. The royal family had to go out onto
their balcony again and again to wave to the crowd.The atmosphere on the
streets grew feverish. Enemy agents were rumoured to be poisoning the
reservoirs; air raids were expected any minute; every woman in the street
might be a cross-dressing spy.The mob harassed anyone who looked foreign
or in any other way suspicious and threw stones through the windows of
the Serbian consulate.6 When Austria-Hungary declared war on Serbia on
28 July, the Munich mob cheered. The next day all officers of the Bavarian
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

16 To Wa r 1914

army were ordered to return to their units. In this jittery environment,

Rupprecht later noted, ‘one was entirely aware of the seriousness of the
­situation but also fully confident and believed, that, given how things had
turned out, it was best to appear utterly determined’.7
At five o’clock in the afternoon of 1 August 1914, Kaiser Wilhelm II sat
down at his desk, carved, ironically, from the timbers of Nelson’s Victory.The
Chancellor, Theobald von Bethmann Hollweg, handed him the order that
would mobilize the German army and set the world at war. The Kaiser and
his advisers had discussed war time and again over the previous few days.
The time for thinking and talking was now past. He did not consult Bavaria
or other princely states. He did not need their agreement. The Kaiser was
Supreme War Lord of their armies, too. Wilhelm, with a flourish, signed the
mobilization order. As soon as the news reached Munich, Ludwig III issued
the proclamations which sent Bavarian soldiers scurrying to their depots to
report for duty. Reservists returned to the colours, drew their equipment,
and filled the ranks up to war establishment. Everything had been planned
for and prepared, and within hours the first trains began to roll, carrying the
army to the frontier.8
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

2
The Battle of the Frontiers

R upprecht set up Sixth Army headquarters in the luxurious surroundings

of the Hotel Bayerischer Hof in Munich. His staff began to organize
the myriad details involved in moving 220,000 men to war, working under
the chief of staff, Konrad Krafft von Dellmensingen.1 Ever since 1813, each
field commander had been assigned a chief of staff. In theory, the division
of labour between commander and chief of staff was clear. As one of the
leading historians of the German army, Robert M. Citino, has put it, ‘the
chief of staff was not a co-commander, a surrogate commander or even a
vice-commander. He was simply a highly trained officer . . . who could give
sound advice. . . . The army or corps commander still had a great deal of
leeway in deciding how to proceed. Although he was ultimately responsible
for making the decision, a wise commander also listened carefully to the
advice of his chief of staff.’ In practice, however, things were not always so
simple. Conflicting views existed of the responsibilities of chief of staff and
army commander. When the thirty-three-year-old Crown Prince Wilhelm
of Prussia was given command of Fifth Army, for instance, the Kaiser told
him to do whatever his chief of staff advised. The Chief of the General Staff,
Helmuth von Moltke the Younger, who was also present, hurried to correct
the Kaiser. He told the young prince:‘never forget that the Army Commander
is and remains the responsible head. The Chief of Staff has to advise, and
now—God be with you!’ The relationship between army commander and
chief of staff was ambiguous and fluid and much depended on the person-
alities involved.2
Major-General Konrad Krafft von Dellmensingen was fifty-two years old
and the Chief of the Bavarian General Staff when he was sent to work
alongside Rupprecht. This was a surprise appointment, the result of a mix-
ture of politics and luck. Originally, the Prussian Lieutenant-General
Konstantin Schmidt von Knobelsdorff was pencilled in as Sixth Army chief
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

18 To Wa r 1914

of staff, with Krafft down to work with the Prussian Colonel-General

Eichhorn at Fifth Army. Eichhorn, however, was not fit to take the field
after a riding accident. Crown Prince Wilhelm was promoted to take his
place, despite his youth and inexperience: he had never commanded any-
thing larger than a division. Clearly, he would need considerable support.
Schmidt von Knobelsdorf was the obvious man to step in. Not only was he
the most senior member of the General Staff after Moltke, but he had been
teaching Wilhelm tactics and operations at the Kriegsakademie and knew
him well. Teaming Schmidt with Wilhelm would also avoid the awkward-
ness of the heir to the empire taking instructions from a Bavarian. So Moltke
swapped Schmidt and Krafft over. Rupprecht was not only glad to have a
Bavarian colleague, but knew Krafft personally from their time together at
the Munich Kriegsakademie and respected him as a ‘smart and energetic
man’. The two forged and maintained a close bond from the start. The next
three most senior staff officers were also Bavarians Rupprecht knew, the rest
being a mix of Bavarians and Prussians.3
Indeed, four of the five army corps which comprised Sixth Army were
made up of Bavarian troops. The other was Prussian. Each corps fielded two
divisions of about 18,000 men each. In addition, there was a Bavarian cavalry
division, plus a brigade of older Landwehr troops to guard the lines of com-
munication. Rupprecht’s orders were to set up his headquarters in St Avold,
east of Metz in German Lorraine. He was to take overall command of the
whole German left wing, including Seventh Army to his south in Alsace.
Under the famous Schlieffen Plan, designed by Moltke’s predecessor, most of
the German army would swing through Belgium and across northern
France. Meanwhile, Rupprecht and the left flank were to tie down as many
French troops as possible until the German right had outflanked and defeated
them. In this way Schlieffen and Moltke planned to knock France out of the
war within weeks, freeing up troops to shift east against Russia. Specifically:
The mission of the overall commander [of the left wing] is to advance towards
the Moselle between Frouard and the Meurthe . . . to tie down the French
troops concentrated there and prevent them being transferred away to the
French left wing.
If the French for their part attack between Metz and the Vosges in superior
strength, this mission may become impossible. In the case that it becomes
necessary for the parts of our army in the Reichsland [Alsace-Lorraine] to
withdraw, they should so direct their movements as to prevent the left flank of
the main German forces being threatened. . . .
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le of t h e F ron t i e rs19

Should Sixth and Seventh Armies not encounter superior French forces, then
the possibility arises of operations by Sixth Army . . . past Metz or to the south
on the left bank of the Moselle.4

Rupprecht claimed in his diary that he knew neither his order of battle nor
his mission until the deployment orders arrived on 2 August. This may have
been true in the very narrow sense that he may not have seen his specific
orders in advance. It is less plausible that he knew nothing at all. The com-
position of Sixth Army, its area of operations, and its mission had all been
fixed for nearly ten years. Staff rides and wargames had examined many
aspects of the plan. Bavarian officers had been closely involved in exercises
practising operations in Lorraine. Even French intelligence had managed to
obtain a copy of Germany’s plans. It hardly seems likely that Rupprecht knew
nothing, therefore. From later comments in his diary, where Rupprecht dis-
cusses deviations from the Schlieffen Plan, he was clearly aware of its outline.
He had also taken part in a mobilization exercise in Berlin in January 1914,
where further intentions must have been discussed. In fact, during that exer-
cise Moltke specifically briefed Rupprecht on the most controversial aspect
of the whole Schlieffen Plan: the violation of Belgian neutrality. Rupprecht
may have been playing down his foreknowledge to understate his involve-
ment in an aggressive war plan which ultimately failed.5
The first few days of August saw a flurry of staff activity as the Bavarian
War Ministry and Sixth Army headquarters toiled to build up the army
and get it moving. The peacetime strength of the Bavarian army was 4,089
officers and 83,125 other ranks. It suddenly expanded nearly five times, to
a total of nearly 416,000 men. Two-thirds of these constituted the field
army, with the rest remaining in reserve at home. To deploy these men
where they were supposed to go, the authorities commandeered some
5,500 locomotives and moved 285,000 carriages and wagons in the first
two weeks of the war. We can begin to imagine the scale and complexity
of the task facing Rupprecht’s staff if we remember that a single army corps
would occupy 140 trains or fill nearly 65 kilometres of road. Each corps
required 130 tons of food and fodder per day, just to survive. Sixth Army
had five such corps, as well as other attached units, all of whom had to get
to the right place at the right time.6
While Krafft oversaw this administration, he also took the lead in drawing
up detailed plans for the first campaign. He suggested that the best way to tie
down the French would be to seize the initiative and go on the offensive,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

20 To Wa r 1914

rather than wait for the enemy to move first. Rupprecht wholeheartedly
agreed. The news that the British had joined the war worried Krafft little:
‘We’ll easily deal with them, too.’ Rupprecht was not so sure: ‘You don’t
know the English’, he replied.7 Rupprecht found this period of preparation
stressful and had trouble sleeping. The febrile atmosphere in Munich did not
help. The weather was oppressively hot and the night-time streets were noisy,
with crowds singing patriotic songs and hymns. He was relieved to be on the
move at last when, in the evening of 7 August, he tucked his two sons up in
bed, drove to the station, and boarded his train for the Western Front.8
Nearly thirty-six hours later, just before eight in the morning of Sunday
9 August, Rupprecht and his staff disembarked at St Avold. A grimy indus-
trial and garrison town some 50 kilometres east of Metz, it is today in
France. In 1914 it was German, in the part of Lorraine annexed in 1871 after
the Franco-Prussian War. As he got down from his train, Rupprecht received
news of the first casualties. Two German officers had cracked up and shot
themselves. The war had already begun to extract its blood price.9

Figure 2.1.  Rupprecht with his two sons before the war
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le of t h e F ron t i e rs21

Of greater military significance were reports of early combat. Cavalry skir-

mishes along the border on 5–6 August had been followed by a major French
move down in Alsace. General Auguste Dubail’s French First Army sent
General Louis Bonneau across the border to seize Mulhouse. This Bonneau
did on 8 August. French joy at this propaganda success was short-lived. The
German Seventh Army under Colonel-General Josias von Heeringen was
responsible for the area from Strasbourg to the Swiss frontier. It soon bundled
the French out again and within forty-eight hours the French were back
where they had begun. Bonneau became the first of many French generals in
1914 to be sacked by the commander-in-chief, General Joseph Joffre.
On 10 August, Heeringen formally came under Rupprecht’s command.
To delegate responsibility for the left wing to a local commander in this way
was a sensible plan. Communications were expected to be difficult, which
would make it hard for Moltke to control everything from Army Supreme
Command (Oberste Heeresleitung: OHL). Telephones and radios were
scarce and both were liable to interception; on the other hand, given the
distances involved, face-to-face liaison was time-consuming and tricky.10 No
formal command mechanism such as a separate Army Group headquarters
had been planned for, however, and the whole arrangement was jury-rigged.
Already stretched Sixth Army staff were simply expected to take on more
work. Nor had the two headquarters practised working together. Rupprecht
thought this a fatal mistake. Although the two armies exchanged liaison
officers, coordination remained poor. Strained relations between the com-
manders did not help. Rupprecht thought Heeringen ‘extremely nervous’.
Heeringen, on the other hand, a decorated veteran of 1870 who had served
as Prussian War Minister and was nearly twenty years older than Rupprecht,
felt that he was the senior of the two army commanders on the left wing and
resented being subordinated. The use of a similar arrangement on the
German right wing proved disastrous on the River Marne in September.11
Elements of Sixth Army began to probe forwards. Near the border, sharp,
small-scale engagements took place at Lagarde (11 August) and Badonviller
(12–13 August). The former was a neat combined arms fight, where infan-
try, artillery, and cavalry worked together well to take the village and nearly
1,500 prisoners. Losses, however, were heavy. At one point, about 1,000
Bavarian Uhlans charged, lances couched, downhill and into the village.
Cavalry on horseback immediately proved extremely vulnerable to modern
rifle fire: nearly a quarter of the men were lost. Badonviller was a more
confused affair. Here, combined arms cooperation broke down. The Bavarian
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

22 To Wa r 1914

Foot Guards, desperate to prove their courage, refused to wait for artillery
support. Charging across 400 metres of open ground, the regiment lost
twenty-one officers and 398 men to capture a town of little account.
Rupprecht was angry at this bravado: ‘How often during peacetime exer-
cises did I tell people not to charge until much closer to the enemy!’12
The main campaign opened on 14 August. In a thrust designed to syn-
chronize with the first Russian offensive into East Prussia, the French
Second and First Armies, under Generals Noël de Castelnau and Auguste
Dubail respectively, pushed across the frontier, heading north-east towards
Morhange and Sarrebourg. The French army is often criticized for tactical
naivety in the early battles of 1914. The impression we have is of white-
gloved officers, deeply indoctrinated in the cult of the offensive, repeatedly
leading red-trousered infantry forwards to ineluctable death, pitting fixed
bayonets, heartbreaking élan, and flying tricolours against machine-guns.
There certainly were events like that. Two French infantry regiments were
ripped to shreds by Bavarian firepower when they tried to capture the town
of Cirey with a bayonet charge across 2,000 metres of mainly open ground,
for example. More generally, however, it was not the recklessness, but the
caution, of the French attack which was remarkable. Despite generally only
light opposition, Castelnau insisted on ‘a methodical organization of the
advance’, with ‘painstaking artillery preparation’ before any infantry assault;
all ground taken was to be carefully consolidated in case of enemy counter-
attack. Meanwhile the Germans withdrew as planned, hoping to suck the
French forward into the trap.13
At first the Germans thought the Lorraine attack marked the main enemy
effort. Strategic intelligence suggested that as many as eighteen of the
twenty-two active corps in the French army were involved. Only after three
days of combat did Sixth Army correctly identify that the enemy was only
half that strong. Collating intelligence in mobile warfare was never easy:
terrain and enemy cavalry got in the way and it was often hard to tell the
difference between active and reserve formations. These factors made it dif-
ficult to estimate the enemy’s true strength. The Germans also had a prob-
lem of interpretation, however. They thought the Lorraine offensive was
Joffre’s main effort primarily because that was what they wanted to believe.
The idea of a major French drive to regain Alsace and Lorraine played too
neatly into what the Germans dreamt their enemy would do.14
The French were advancing ‘so extremely slowly and methodically’—
only 5–8 kilometres per day—that Sixth Army was beginning to think the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le of t h e F ron t i e rs23

chances of springing a trap were slight. The realization that they were facing
only a subsidiary attack reinforced Rupprecht and Krafft’s feeling that the
best way to tie hostile forces in Lorraine was to attack, not defend. They had
been pushing OHL to approve an offensive for days but now they stepped
up the pressure. In a series of telephone calls and face-to-face meetings
between 16 and 18 August, with a shifting cast of OHL characters including
the chief of operations, Lieutenant-Colonel Tappen, the head of the
Political Section, Lieutenant-Colonel von Dommes, and the Quartermaster
General, Lieutenant-General von Stein, Rupprecht pushed vehemently for
permission to attack. He was growing frustrated at OHL not giving him a
free hand to carry out his mission. One encounter, with Dommes on
17 August, grew so heated that Dommes forgot his sword and helmet in his
hurry to get away. Finally, during the afternoon of 18 August, Krafft spoke
again with Stein, who told him that OHL would not forbid an attack, that
Sixth Army must do what it considered right, and ‘you must bear the
responsibility’. They decided that Sixth and Seventh Armies would attack.
The date they chose was 20 August.15
By the early morning of that day, the French had pushed up to 30 kilo-
metres across the border. Dubail’s men were in Sarrebourg, preparing
to  drive on the heights north-east of the town. Castelnau, buoyed by
the apparent ease of his army’s advance, discarded his earlier caution on
18 August and called for ‘all possible vigour and speed’ and ‘the energy and
élan of all’. Despite this encouragement, only Ferdinand Foch’s XX Corps,
the famous ‘Iron Corps’, made much progress on 19 August. Crossing the
plain of the River Seille, they arrived in front of a steep ridge. On the
heights above perched the small town of Morhange. Foch, despite orders
from Castelnau to wait, decided to storm this ridge at six o’clock on the
morning of 20 August.16
At five o’clock, as ‘the sun rose blood red above the ground haze’, Foch’s
men began forming up. Suddenly, German shells from gun positions on the
high ground rained down among them. Bavarian infantry swept down the
slopes. ‘Our troops, in their enthusiasm, attacked with flags flying’, wrote
Rupprecht. Never again would they do so, but on 20 August, with surprise
on their side, the Germans swept all before them. XX Corps held up rela-
tively well but the rest of the French Second Army collapsed. Within just
ninety minutes the Bavarians swarmed across the valley and overran the
French artillery lines. Further east, on the French First Army front, the pat-
tern was slightly different. French morning attacks around Sarrebourg were
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

24 To Wa r 1914

shot to pieces by German defenders dug in on the hills, who then counter-
attacked in the afternoon. Here, too, the French fell back in disorder. Some
French units lost half their strength. German losses were also heavy, how-
ever: one Bavarian division lost nearly half (45 per cent) of its men.17
Within forty-eight hours the French armies lost everything they had
gained in nearly a week and were back where they had started on 14 August.
Joffre had to shelve plans to transfer a corps from Nancy up to the north.
The first major engagement of the war had resulted, as the French official
history admitted, in ‘a serious defeat’. Moltke wept at the news of Rupprecht’s
success. The Kaiser awarded Rupprecht the Iron Cross both Second and
First Class.18
On 22 August the Germans won another major victory, crushing French
forces in the Ardennes and adding to the general euphoria. Rupprecht and
Krafft thought that they had done enough damage to the French that their
mission in Alsace-Lorraine was complete. It was time now to stand pat there
and send men north to join the main attack on the right. When Krafft sug-
gested this to Tappen over the telephone in the afternoon of 22 August,
however,Tappen asked him to hold on the line while he went off to consult
the Chief of the General Staff. When he came back, Moltke’s answer was so
surprising that Rupprecht later asked for it to be repeated in writing. Rather
than redeploying northwards, Sixth Army was to exploit success in the south,
with a drive on Epinal. They were to try and cut off the 120,000–150,000
Frenchmen still thought to be in the Vosges Mountains. Rupprecht was
worried that such a move risked stretching Sixth Army very thin and leav-
ing its flank exposed to attack from Nancy.
OHL had misjudged events on the German left flank. Communications
had broken down in the hilly terrain and the French had retreated so fast
that the Germans had lost contact. The situation was not as healthy as it
seemed: the defence had buckled but not broken. Both French armies had
caught their breath and reorganized and they were now securely established
behind tough defences along the imposing heights of the Grand Couronné
de Nancy. From up there they could watch every move the Germans made
on the broad plain below. The French also still held Fort Manonviller,
behind the German lines. This blocked communications and only surren-
dered after several days of bombardment by monstrous Krupp 420-mm
howitzers. By that time, Sixth Army was already tiring and the swing south
had slowed. The French appeared to be slipping out of the trap in the Vosges.
OHL, as always, pushed for greater speed, but Moltke seemed increasingly
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le of t h e F ron t i e rs25

out of touch. Rupprecht complained in his diary that OHL ‘under-estimates

the difficulties of the pursuit.The Kaiser thinks it’s sufficient to lead the cav-
alry through!’ A request to send troops to help Crown Prince Wilhelm’s
Fifth Army west of Metz further irritated him. Not only was he already
juggling several missions, but he felt Wilhelm was making his own difficul-
ties by handling his forces poorly.19
It soon became impossible for Rupprecht to help Wilhelm, in any event.
In the early morning of 25 August reports arrived of a major French attack
out of Nancy, into Sixth Army’s right wing. Here the commander of
III  Bavarian Corps, Lieutenant-General Ludwig von Gebsattel, had been
ordered to sit on the defensive, well out of range of the French heavy guns
in Nancy, and cover Rupprecht’s flank. Gebsattel, however, decided that the
best way to execute his mission was to ignore his orders and attack. A series
of savage encounter battles ensued. Gebsattel’s men advanced ‘as if they
were on manoeuvres with live ammunition. The full seriousness of the war
had so far come home to few. By the evening of that day some had certainly
become acquainted with its full weight.’ French artillery fire overwhelmed
the German attackers, who fell back in disarray. As rumours of the setback
spread through the streets of Lunéville, panic broke out. Shots were fired
and nineteen civilians killed. In all, seventy houses were burnt down.
A member of Rupprecht’s staff, Major Rudolf von Xylander, wrote that ‘it
is incredible how a victorious army can be reduced in so short a time to
such a state by one single reverse. The nerves of our troops, who have been
fighting bravely for several days without a break, are simply overstretched.’
It was not just the Germans who made battlefield mistakes that day. Men of
General Émile Fayolle’s 70th Division attacked Hoéville at seven o’clock in
the morning. Fayolle described the result as a ‘catastrophe’: there had been
no reconnaissance, no artillery preparation, and too many men were sent
forwards in the battle line. Four thousand men were lost. ‘It’s crazy’, he con-
cluded. Nonetheless, as the day went on, the scale of the French threat grew.
By the evening of a ‘dramatic and critical day’, during which Rupprecht had
not been able to leave his operations room even for a moment, it seemed
that this was another of Joffre’s big pushes.20
At nine o’clock, a liaison officer arrived from OHL with new orders to
turn west and drive through the Charmes gap between Toul and Epinal.
Rupprecht was now to join hands with Fifth Army behind Toul. This
­unrealistic order was followed next morning by another: Sixth Army was to
attack the fortifications north and east of Nancy. This difficult task would
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

26 To Wa r 1914

require significant heavy artillery and thus take time to prepare. Rupprecht
was getting increasingly frustrated: ‘despite the detailed reports we send
them, OHL seems not to see the real picture and to be possessed by
unfounded optimism’.21 There was no way Rupprecht could execute these
grandiose and conflicting orders. Instead, Sixth and Seventh Armies both
went onto the defensive. The infantry of both sides had fought themselves
to a standstill.The guns, however, still killed. French artillery observers seemed
remarkably efficient. If even a single German was spotted in the open, shells
rained around him.The only response was to dig.Trenches, dugouts, barbed
wire: all the diabolical architecture that was to define the Western Front
made its first appearance here in Lorraine. As yet it remained improvised
and ad hoc but, already, trench warfare had begun.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

3
The End of the Campaign
in Lorraine

T he first sign that something was wrong was a telegram from home, late
in the evening of 26 August. It told Rupprecht that his elder son,
Luitpold, was seriously ill. The boy’s condition deteriorated rapidly. At one
o’clock the next afternoon, the doctors diagnosed the ‘Swedish disease’:
polio. Little more than an hour later came the shattering news that Luitpold
had died. He was just thirteen years old. Rupprecht had now lost his whole
family except for nine-year-old Albrecht. Even in an age more accustomed
to children dying young than our own, and for a man with the stiffest of
public upper lips, this was a cruel blow. Rupprecht tried to be stoic: ‘there is
no time to grieve, there’s too much to do. Work is my consolation.’ He told
his father that ‘duty requires action, not grief ’. When Major-General Karl
von Wenninger saw him, however, he thought ‘the poor man looked dread-
ful’, crushed by the dual burden of news from home and the stress of the
campaign. Luitpold was buried in Munich, with his father still far away.1
On the battlefield, both sides launched local attacks and the artillery
­continued to roar, but the front settled into an uneasy stalemate with little
change in the overall situation. On 1 September the French began with-
drawing into their fortifications around Nancy, trading space for time and
allowing Joffre to pull troops out and send them to his threatened northern
flank. Seven divisions had left Alsace-Lorraine by 2 September, some of them
in time to take part in the Battle of the Marne. This was precisely what the
Germans feared most. Although Oberste Heeresleitung (OHL) continued
to press Rupprecht to keep attacking to forestall such a transfer, disagree-
ment continued about how best to proceed. Poor intelligence and these
disputes hindered decision-making. OHL wanted Sixth Army to break
through the Charmes Gap and cross the River Moselle. Rupprecht, however,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

28 To Wa r 1914

thought this impossible until Nancy had been neutralized, which would
require siege guns and time to prepare. On 30 August the Bavarian military
representative at OHL, Wenninger, had just returned from a visit to
Rupprecht when Moltke ambushed him as he walked into headquarters.
‘You were with Sixth Army: when are they finally going to attack?’ Moltke
barked.Wenninger explained that enemy artillery fire was intense, losses had
been heavy, and the men were very tired: some units had been in action
constantly for nineteen days. Moltke had no sympathy: ‘That’s the same for
the other armies too, yet they attack!’ In any case, he said, ‘Sixth Army’s
casualties were in many cases unnecessary. The units run into enemy artil-
lery fire.That must cease!’ Moltke stalked back into his office, hardly waiting
to hear Wenninger’s reply.2
While Tappen and Moltke continued to push for the drive into the
Charmes Gap, others at OHL took a different line. Major Max Bauer, for
instance, the OHL heavy artillery expert, agreed that it was important to
take Nancy first and promised to find siege guns to do so. Sixth Army heard
what it wanted to hear from the babel of voices and began work on plans
for capturing Nancy. Krafft managed to overcome the nerves of his staff
officers and corps commanders, reeling from casualties as high as 50 per cent
in just two weeks, and organize an attack on the Grand Couronné for the
night of 4/5 September.3
Sixth Army hoped that a night attack would maximize surprise and
reduce the impact of enemy firepower. Unfortunately, the sky was clear and
the moon full, so the attackers were easy to spot.The experience of the 14th
Bavarian Infantry Regiment (5th Bavarian Infantry Division), attacking the
village of Réméréville, a dozen kilometres east of Nancy, was typical. Two
battalions advanced an hour before midnight, each deployed with four
companies abreast, covered by a few skirmishers. Meanwhile, artillery bom-
barded the village. After about an hour, with the attackers still 500 metres
short of the buildings, the French detected them. The Bavarians tried to
charge, but were forced to the ground by heavy fire from well-dug-in
defenders to the front and left flank and the assault soon foundered.Attempts
to clear the French from woods on the flank failed. Losses were mounting.
Nearly all the officers were killed or wounded and, around three a.m., order
broke down completely. Even when the Bavarians shelled the village with a
couple of field guns firing over open sights, the situation remained dangerous.
Further reinforcements in the shape of a battalion of light infantry (Jäger)
did not help much and the French finally evacuated the village only around
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e E n d of t h e Ca m pa ign i n Lor r a i n e29

noon. Casualties had been brutal: the Bavarians lost 750 men, or probably
about half their strength, including the two commanding officers.4
The German attack on Nancy made little progress elsewhere, either. OHL
now seemed to be losing interest in Lorraine. On 5 September Tappen told
Krafft to send two corps to the German right flank in the north. Sixth Army
had never thought its task easy but now, it seemed, OHL was going to make
it harder. Rupprecht and Krafft felt they were being hamstrung but their
protests were ignored. From OHL’s perspective, the quick results hoped for
had not materialized and Moltke now had much bigger problems than
Nancy or the amour-propre of the Bavarian crown prince. On 5 September
the Battle of the Marne began with a French drive into the extreme-right
flank of the German army, followed the next day by an offensive by over a
million Entente soldiers from Verdun all the way to Paris. On 9 September,
the Germans began to withdraw, only eventually stopping 60 kilometres
back along the River Aisne. Moltke’s attempt to knock France out of the
war with a single deadly blow had failed.5
The situation on the Marne, and the increasing difficulties of Crown
Prince Wilhelm around Verdun, threatened to draw further resources away
from Sixth Army.When Rupprecht was told to give up further ammunition
and divisions, he drove at once to OHL in Luxembourg to complain.There
he met Moltke who ‘impressed me as a sick, broken man. His tall body was
hunched and he seemed extremely haggard.’ The whole interview struck
Rupprecht as surreal: ‘Moltke went on to say things I didn’t properly under-
stand. He spoke of a new mission for me and stared at me wide-eyed.’
To Rupprecht’s surprise, Moltke agreed that the attack on Nancy could
­continue after all, on condition that the heavy guns be given up in a week at
most. Surprise soon turned to even greater confusion, however. Even while
Rupprecht had been motoring to Luxembourg that morning, an OHL staff
officer arrived at Sixth Army bearing new written instructions. All forces
in Alsace-Lorraine were to go on the defensive. Active formations were to
disengage, reorganize, and prepare to be moved by train to northern France,
leaving the front held by second-class units. This was the new mission
Moltke had spoken so cryptically about. He probably had not realized that
Rupprecht knew nothing of it. When Sixth Army asked whether it was to
follow the verbal instructions to attack, or the written ones to disengage, it
finally received an unambiguous and clear answer: the latter.6
On 10 September, Rupprecht’s men began pulling out of the front line.
Three days later the first men left for Belgium. On 17 September, Rupprecht
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

30 To Wa r 1914

handed over the Lorraine command to General von Falkenhausen, who

established a defence line more or less along the old frontier. And there, for
the next four years, the front lines sat.
Losses on both sides had been heavy. Of every eight German soldiers
who arrived in Lorraine at the beginning of the war, approximately three
became casualties and one died. In front-line combat infantry, who bore the
brunt of the fighting, the proportion of casualties was inevitably higher.
Some units lost as much as 60 per cent of their infantry strength. French
losses are harder to pin down but were probably similar.7
Neither side had much to show for these losses. The front lines settled
pretty much along the 1914 border. No crucial resources or strategic positions
changed hands. That was not the point, however. Neither army had fought
there for such reasons. In the minds of both Joffre and Moltke, this  had
always been a secondary theatre. The aim of both armies in Alsace-Lorraine
was primarily to tie down the enemy while they won the war elsewhere.
Neither side got what it wanted, though. Both German and French troops
were able to slip away and bolster the defence in the north at critical
moments that autumn.The only decisive battle fought in the west in 1914 was
the defeat of Germany’s attempt to overrun France.The Marne was not the
battle which won the war. But it did ensure that France would not lose it.
Before we leave Lorraine and explore the events of autumn of 1914, we
need to address the issue of war crimes. John Horne and Alan Kramer have
demonstrated that, within days of the war’s outbreak, civilians as well as
soldiers were suffering its violence. German soldiers massacred townsfolk at
Dinant and Tamines and destroyed the ancient library at Louvain. Men
under Rupprecht’s command in Alsace-Lorraine also committed atrocities.
For example, after the bloody capture of Badonviller on 12 August, they
accused the villagers of helping the French army. Bavarian troops burned
down houses, knocked down the church tower from which partisans,
known as franc-tireurs, had supposedly been sniping, and killed ten villagers,
including the wife of the mayor. On 20 August, men of Sixth Army attacked
and took Nomény. During confused street fighting in the town, they came
under fire. Blaming the civilian population, they torched the town and shot
forty-six inhabitants dead as they fled the flames. Two more were fatally
wounded, while another seven suffocated as they cowered in their cellars.
Four days later, other German soldiers came under fire in Gerbéviller. They
burned down the village and killed sixty inhabitants, some of them in two
mass executions. Another incident took place at Lunéville on 25 August.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e E n d of t h e Ca m pa ign i n Lor r a i n e31

As Bavarian troops retreated through the town in confusion, shooting broke

out. Major Berthold Schenk von Stauffenberg took sixty hostages and
nineteen civilians were killed. Some seventy houses were destroyed.8
These atrocities were not isolated incidents but part of a widespread pat-
tern. In August–September 1914 across Belgium and France there were 129
incidents involving the summary execution of ten or more civilians. About
half of all the regiments in the German army, from every corner of the
Reich, were implicated. Probably some 6,500 civilians were killed. Nearly a
thousand of them were French, the rest Belgian. Many thousands of homes
were destroyed. In some cases, senior officers took the lead in committing
war crimes. For instance, Major-General Stenger, a brigade commander at
the Battle of Sarrebourg on 20 August, the next morning ordered some of
his officers to kill all their wounded French prisoners. Five days later he
repeated a similar order, this time in writing. More notably, on 12 August
Moltke himself ordered that anyone not in uniform resisting the German
advance was to be shot at once. This level of direct involvement was rare,
however. More often, as Isabel Hull explains, ordinary soldiers or very jun-
ior officers carried out the atrocities. More senior officers got involved only
when they chose either to ignore or to approve such acts, or indeed to
systematize them, as Moltke did.9
Rupprecht’s involvement in atrocities was, it seems, limited, but he cannot
escape all blame. Although he was number 33 on the French post-war list
of  890 wanted war criminals, his case never came to trial. One of the
charges was that he had been responsible for the ‘methodically organized’
crimes committed by units under his command in occupied territory in
August–September 1914. Evidence that he was personally culpable is thin.
No order survives to link him to the killing of civilians or the burning of
houses. The most accurate accusation was that he had been present on
22 August when Dieuze was pillaged. Ten houses were set on fire and the
mayor and local curé were shot. Even if it cannot be proved that Rupprecht
had blood on his own hands, however, we have already seen that men
under his command did. Perpetrators were not disciplined and indeed
some, such as Stenger, were soon promoted.The failure of the Crown Prince
properly to investigate, much less take decisive action against, those implicated
must rank at least as a sin of omission. Whatever the stresses on him, and
even if it was not his finger on the metaphorical trigger, we cannot absolve
him of, at a minimum, moral responsibility for the crimes committed under
his command.10
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

32 To Wa r 1914

To judge those who have long been dead by modern ethical standards
achieves little beyond stoking the fires of our own self-righteousness. It is
more interesting to try and understand why they did what they did than to
condemn or praise them for it. Untangling Rupprecht’s attitude to what we
call atrocities is tricky. He became more sensitive to criticism on this score
after the failed Leipzig War Crimes Trials of the early 1920s. Consequently,
his published diary differs in intriguing ways from the wartime evidence he
left. For instance, Rupprecht mentioned the Dieuze episode in his published
diary, but not in the manuscript version. Presumably by 1929 he felt he
needed to counter French allegations. According to him, after shots were
fired during the night he permitted a house-to-house search. Fifteen French
soldiers were hauled out of hiding places around the village. The civilians
were, therefore, he implied, far from innocent of war crimes themselves and
reprisals were well deserved. He said nothing of executions, however.11
Rupprecht’s attitude to Dieuze specifically, and the use of violence against
civilians generally, was moulded by three factors. First, it reflected how German
officers conventionally interpreted the laws of war at this time. Soldiers, they
thought, possessed ample latitude to override even the established law of the
Hague Conventions (1899 and 1907) if military necessity demanded it. As
Isabel Hull has demonstrated, this view of military law had diverged radically
from that of the liberal European states for decades. Both British and French
officers were taught that suspected partisans must be tried and could not be
shot out of hand, whereas we have already seen that Moltke’s stated policy was
the opposite. When Rupprecht’s men shot French civilians, they could argue
that they were only following the orders of the Chief of the General Staff.
This attitude to law was a sub-set of the second factor: the broader ethos of
the German army. This idolized the objective of annihilating the enemy. All
other considerations were secondary. Civilians, therefore, were at best instru-
ments and more likely obstacles. The expectation that civilian franc-tireurs
would oppose the German advance, as they had during the Franco-Prussian
War of 1870–1, found in such a military culture rich soil in which to flourish.
German soldiers saw snipers behind every bush and shutter, largely because
they expected to. The third factor, the particular pressures of the 1914 cam-
paign, ratcheted up the level of violence further. The whole design of the
German campaign depended on rapid movement and the ruthless destruction
of resistance.Without speed, the Schlieffen Plan was nothing. If the demanding
timetable was to be met, opposition must be crushed at once. There was no
time for niceties such as trials of enemy civilians.12
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e E n d of t h e Ca m pa ign i n Lor r a i n e33

So, Rupprecht probably saw little wrong with most of the violence his
troops visited on the local population. He did retain some sense of proportion,
however. There were limits to what he thought acceptable. For instance, he
criticized some of the house-burning carried out by Bavarian troops on
20–1 August as unjustified. More generally, Rupprecht was publicly frustrated
by the continued burning of French villages. He tried to reduce or stop the
practice. On 24 August, for instance, he issued an Order of the Day:
Wanton destruction of enemy property of any kind, whether belonging to
military or civilian authorities or to private citizens, is unworthy of the
German soldier.
Necessary searches of offices of enemy military or civilian powers for secret
material or weapons must be carried out thoroughly and energetically but
should restrict themselves to the objects of the search.
It is the absolute duty of each commander to maintain, even using the harshest
measures, the renowned discipline and good name of the German soldier at all
times, even under the most difficult circumstances.
Even if the enemy is guilty of allowing outrages, this in no way justifies
German soldiers following suit.
Further, German soldiers should continue their victories and spread fear and
terror only through their determination for victory, not through other means.

Presumably from concern that they might not be obeyed, these orders were
repeated on 29 August and 9 September, with a similar Order of the Day
issued on 19 September, as Sixth Army redeployed to France, and again on
2 October. Arson became increasingly subject to formal process, with the
last of the series ordering that whole villages could only be burned down
by order of OHL or an Army commander, while the authorization of Corps
HQ must be sought for any partial destructions.13
Rupprecht issued these high-profile orders from four main concerns,
which were more pragmatic than ethical or legal. First, burning a village
took time and damaged food and shelter which might be needed one day.
Second, it was a punishment which hurt innocent and guilty alike and left
the population homeless and jobless ‘which is definitely not in our interest’.
Third, it risked undermining army discipline and might provoke the locals
to riot. Fourth, he had the breadth of vision to see that not everyone shared
the German military’s interpretation of law and that the enemy was likely
to pick up on anything that looked like an atrocity and use it in propaganda.
Many others lacked the imagination to see this. Even civilian statesmen in
the Foreign Office or Chancellery either followed the military line or knew
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

34 To Wa r 1914

that it was futile to attempt to intervene. After all, the Kaiser himself had
long used the most bloodthirsty of language against his enemies. Rupprecht
was proved right. The Entente shouted quick and loud about German
atrocities in Belgium and France. Perhaps the German authorities failed to
investigate war crimes because they wanted to avoid giving their enemies
ammunition for further propaganda. Equally, however, they may well have
expected that victory would simply solve the problem for them.14
Once Sixth Army moved to Picardy and the River Somme, it operated in
a less densely populated landscape from which much of the population had
already fled. As a result, there were fewer opportunities for friction with the
civilian population.The number and scale of atrocities against civilians soon
fell away. The fighting, however, only grew harder.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

4
The First Battle of the Somme

O n Friday 18 September 1914, Rupprecht and Krafft travelled up to

their new headquarters at Namur in Belgium. On the way, they
stopped at Oberste Heeresleitung (OHL) to discuss their mission with
Moltke’s replacement as Chief of the General Staff, the War Minister, Erich
von Falkenhayn. Moltke had been sacked after the Marne but for propa-
ganda purposes the news had not yet been made public. Awkwardly, Moltke
was present at the meeting. Rupprecht found him ‘considerably calmer’ than
the last time they had met. To Krafft, however, Moltke seemed ‘very shy, he
hardly said a word to us: not even hello or goodbye’. Falkenhayn explained
that a series of attacks were underway all along the line to pin the French
army in place, while Rupprecht and Sixth Army were to swing around the
open flank north of the River Oise and land the decisive blow. To begin
with, Rupprecht would have at his disposal XXI Corps together with I and
II Bavarian Corps. After 23 September, XIV Reserve Corps, largely recruited
from the industrial areas of Baden, would join them. Not all the troops were
yet in place, though. Rear area communications remained sketchy. The
bridges over the River Meuse at Namur had not yet been repaired so four
whole German armies would have to rely on a single railway line via
Brussels. Most troops would have to get off their trains at Namur and march
the 160 kilometres to the area of operations around St Quentin. The ques-
tion was whether Sixth Army had a few days to wait, concentrate, and strike
in strength; or must advance as soon as possible with whatever was to hand.
Rupprecht and Krafft preferred the first option. Falkenhayn insisted they
throw troops in as they arrived, for two reasons. First, there were already
signs of French pressure on General Alexander von Kluck’s First Army on
the right of the German line. The open flank there must be protected as
soon as possible. Second, he under-estimated enemy resilience and exagger-
ated German strength. One more push and the French, Falkenhayn thought,
would collapse.1
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

36 To Wa r 1914

Such intelligence estimates were based on little more than guesswork. In

mid-September, the Germans had little clue where enemy units were, never
mind how strong or ready to fight they were. Indeed, they were having
trouble keeping track of their own men. For instance, three of the cavalry
divisions supposed to be reconnoitring the open right flank had completely
disappeared. Communications had advanced little since the days of Frederick
the Great or Julius Caesar: wireless sets were too bulky and scarce to be used
below army level at best; it was impossible to build out telephone networks
quickly enough to keep up with the rapidly moving troops; pigeons were
no use for any but the shortest communications. The best way to send a
complex message remained to send a man with it, for all that he might get
captured, killed, or simply lost. In any case, the German cavalry had suffered
heavy losses and lost the initiative to the stronger and more cohesive French.
Poor weather prevented aerial reconnaissance from filling the gap.2
Joffre was playing the same game as Falkenhayn. He also was shifting
formations north to try to outflank his enemy. Even as Rupprecht headed
for that open flank, so did the Frenchmen of his old adversary Castelnau’s
Second Army. Joffre had the advantage of an intact rail system which enabled
units to be unloaded close to the action, while the Germans had to make
long approach marches from railheads in the rear. Thus, while French units
could transfer from Lorraine to the new battlefields in five or six days, it
took the Germans nine. This extra speed was sufficient to allow the French
to block German outflanking manoeuvres but not great enough to allow
their own moves to succeed. Instead, time and again the same pattern played
out over the next few weeks. One side or the other would advance, trying
to get around the enemy flank. At first, they would encounter only light
opposition, mainly from cavalry and artillery. Soon, however, enemy resist-
ance consolidated, the weight of fire increased, and it became ever harder to
push forward until eventually the attackers went to ground and began to dig
themselves in, while to their north, the process began again. The fighting
between Rupprecht and Castelnau in September and October 1914, in what
we might call the First Battle of the Somme, neatly shows how that process
unfolded. It also introduces place names which later played an ill-starred
role in British military history.3
The first elements of Sixth Army went into action on 23 September.
Advancing south-west from St Quentin, XXI Corps clashed with French
units coming up from Montdidier. Over the next few days they fought
their way forward north of Roye, suffering heavy casualties as they did so.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e F i rst Bat t le of t h e Som m e37

Rupprecht sent I Bavarian Corps to block any French attack from Amiens
by seizing Péronne and the crossings over the River Somme and occupying
the Flaucourt Plateau on the left bank of the Somme. In the face of fierce
French attacks these Bavarians became increasingly bogged down around
Lihons. On 25 September, therefore, Rupprecht ordered his II Bavarian
Army Corps to attack along the right bank of the Somme, working around
the enemy from the north. Two divisions of cavalry were to march on
Albert to protect the infantry’s flank. Meanwhile XIV Reserve Corps was
to march on Bapaume from Cambrai.
II Bavarian Corps attacked north of the Somme, fighting its way forward
in the face of toughening resistance from French Territorial troops, through
Morval to Maricourt and Ginchy. Rupprecht was keen to exploit this suc-
cess and push the enemy back beyond Amiens. While cavalry probed to the
south of Arras, he ordered XIV Reserve Corps to advance through Bapaume
on Albert while II Bavarian Corps, pivoting on its left, wheeled right
towards the line of the Somme. With the rest of Sixth Army apparently
stuck, if victory were to be won anywhere, it could only be north of the
Somme. Unfortunately, not everything went according to plan. Attempts to
work forward under cover of darkness to storm the defences early next
morning only left the attackers horribly exposed in the open when the sun
rose on 26 September. Tired troops advanced only slowly and, although the
Germans made some progress and captured Mametz Wood, for example, the
front was beginning to congeal once more. XIV Reserve Corps, com-
manded by Lieutenant-General von Stein, one of Schlieffen’s closest disciples
and until very recently Moltke’s deputy at OHL, came in for particular
criticism. According to Krafft’s diary, Stein was ‘very soft even compared
with his extremely soft troops—the Badeners have done poorly in the
whole campaign right from the start—one can detect the social-democrat
taint!’ Stein and Rupprecht had an awkward interview the next day.4
Rupprecht remained determined to get across the river, turn the French
flank, and deliver the longed-for decisive victory. On 28 September XIV
Reserve Corps closed up to the River Ancre, with German cavalry occupy-
ing Thiepval. From here they could watch French troops moving around
the houses of Beaumont, Hamel, and Serre on the other side of the valley.
Further enemy forces were reported in Arras and Douai. The next day II
Bavarian Corps occupied Fricourt after a night of fighting from house to
house, at one point using a searchlight to provide illumination. The French
defenders turned out to be from the élite XX Corps, the same ‘Iron Corps’
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

38 To Wa r 1914

which had covered the retreat from Morhange under General Foch five
weeks previously. Meanwhile, Rupprecht went to watch units of XIV
Reserve Corps fighting in the outskirts of Albert from the windmill at
Pozières. He could clearly see the golden Madonna atop the basilica in
Albert and watched an artillery duel between a German howitzer battery
and what they thought were British horse artillery. It felt, he later wrote, as
though the Germans were no longer the ones doing the outflanking. Others
felt an ominous shift in the psychological balance between the two sides.
One German colonel noted that
the enemy had changed too. The battles in Lorraine had given our troops a
feeling of superiority. They received an extremely rude shock when they had
to accept near Maricourt that the French, even though they were from the
same corps that we had brushed aside in Lorraine, had changed. They were
tough, daring and self-confident. The Miracle of the Marne had so raised the
morale of the French that the order to withdraw must be regarded as a crime
against all things German. Against such an opponent [our] ill-prepared, unco-
ordinated and over-hasty minor attacks were dashed to pieces.5

As it became ever clearer that the enemy would not be beaten on the
Somme, Rupprecht’s attention began to drift north. He handed command
of his units south of the river over to General Fritz von Below and began
work to capture Arras. The situation there remained obscure. He had little
idea of the likely strength of opposition. Consequently, he turned down a
proposal for a night-time coup de main to seize Arras. Instead, he ordered a
pincer attack. On 1 October IV Corps was to march up from Bapaume and
the south while I Bavarian Reserve Corps swung through Douai and
approached Arras from the east. The Prussians of IV Corps, however, found
their way blocked by both French territorials and élite mountain troops of
the Chasseurs Alpins. At Douai, the Bavarians faced only light opposition
and took the town easily, but they, too, were not moving quickly enough.
Resistance grew the next day and progress remained sticky. Rupprecht was
becoming impatient. He drafted a sharply worded order telling units to
report back at once if they failed to reach their objectives and reminding
cavalry that they must bypass enemy concentrations, rather than stop at the
first sight of them. On reflection, however, he decided not to send it, largely
from fear of upsetting his Prussian subordinates: ‘if I were a Prussian,
I wouldn’t have thought twice’.6
The real problem, however, was that the troops were getting increasingly
worn out. Krafft wrote in his diary:‘The men see “fortifications” everywhere.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e F i rst Bat t le of t h e Som m e39

One cannot hide the fact that the offensive spirit has declined since the sad
time in Lorraine. But it is the same amongst the troops of the other armies.
Everyone has his limit. The troops have had too much asked of them.’ Each
of the Bavarian corps had now lost between 7,000 and 9,000 men since the
start of the war. Some divisions were down to just 5,000 men. ‘The sad
truth’, Krafft noted, was that it was proving impossible to dislodge entrenched
defenders with field artillery and tired infantry alone. Only if heavy guns
were brought up and a set-piece attack developed was it possible to make
progress. The cavalry were a particular problem. Their staff officers blamed
slow progress on the difficult industrial terrain. Reports to Krafft, however,
suggested that the real problem was sybaritic officers who were unused to
the rigours of campaigning and would not leave their baggage behind.7
Rupprecht did not want to get sucked into large-scale street fighting. So
he continued to try to surround Arras before entering the town. To the
north, I Bavarian Reserve Corps drove back the French defenders and
occupied Vimy Ridge and Souchez late on 4 October. In the south, the
Guard Corps struggled forward. At one stage aerial reconnaissance reported
that the French were pulling out. Attempts by Rupprecht to take advantage
of this, however, collapsed when his troops met renewed strong opposition.
In fact, Castelnau had been on the brink of a large-scale retreat but he was
overruled by the new commander of all French forces north of the River
Oise, his former subordinate, Ferdinand Foch. General Foch forbade any
further trade-off of space for time. Instead his men were to hold firm
­everywhere and, if possible, counter-attack to regain the initiative. One such
counter-attack threatened the northern flank of the Bavarians scrapping
their way westward along the ridge of Notre Dame de Lorette on 5–6
October. Other French forces were reported to be detraining further north
again, at La Bassée. At the same time, strong enemy counter-attacks halted
German progress south of Arras. Rupprecht threw the newly arrived XIV
Corps into the gap between Lens and Lille on 8 October. They encoun-
tered heavy opposition around and north of Loos. Again, progress was less
than hoped. The same day, a major French assault south of the Somme
forced the extreme left wing of Sixth Army to postpone a large-scale break-
through attempt it had been preparing. That night the over-extended
I Bavarian Reserve Corps came under severe French pressure. Fighting in
Carency was heavy and forced the Germans onto the defensive. They had
to abandon thoughts of further advances between Arras and Notre Dame
de Lorette for now.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

40 To Wa r 1914

The difficulties facing the German army were multiplying. Ammunition

supply, for example, was becoming an ever greater challenge. By 9 October
Sixth Army had enough shells for just one more day of battle. Stalemate set
in around Arras. General von Fasbender, the commander of I Bavarian
Reserve Corps, summed up the difficulties he faced: he was advancing against
a strongly entrenched enemy. His troops were exhausted and he lacked heavy
artillery support. How he was to take the four villages next in his way, he
knew not.8 The capture of La Bassée on 11 October, and Lille the next day,
extended the line further northwards, but progress was becoming ever slower
and harder. Even the cavalry was encountering more solid enemy resistance.
Most significant of all in the long term, on 12 October they exchanged shots
with British troops around Estaires and Merville. Yet again, it was clear,
attempts to outflank the enemy had failed and Sixth Army was ordered onto
the defensive. The battle for Arras was over—for now.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

5
To Ypres

S till the search for an open flank went on.The eyes of both Falkenhayn and
Joffre now turned to Flanders, the last free space on the map. Once again,
they pulled troops out of the trenches in stalemated sectors, loaded them
onto trains, and sent them north.Antwerp finally surrendered on 12 October,
freeing up further German units to advance towards the French frontier.
On 14 October Oberste Heeresleitung (OHL) ordered Sixth Army to
stand on the defensive for the moment, holding a line from Menin through
Armentières and La Bassée and then on south. Falkenhayn hoped that this
would tempt the British and French to try to overlap him in Flanders.
He then planned to strike them in the flank using a new Fourth Army.
This ­comprised four newly raised reserve corps under Duke Albrecht of
Württemberg and was forming up near Brussels. Falkenhayn had read his
enemy perfectly: Field Marshal Sir John French was already directing the
British Expeditionary Force (BEF) eastwards through the gap between La
Bassée and the sea. The harder the British pushed Rupprecht’s right wing
back, the further they would be sticking their head into the noose. If, on
the  other hand, the BEF awoke to the threat from Albrecht and turned
north-east to face him, then Sixth Army would be able to break through
at  Arras, drive west and north, and split the enemy in two. Either way,
Albrecht and Rupprecht would surround and destroy the Entente forces
in northern France.
Rupprecht did not think much of this plan. ‘The mission we’ve been
given’, he grumbled, ‘is analogous to what we had at the beginning of
the campaign and it remains as questionable as before whether the enemy
will run into the trap. This return to Schlieffen’s ideas, under very different
­circumstances, seems to me very dubious. We’re ceding the initiative completely
to the enemy.’ It was all very well for Falkenhayn to tell Rupprecht to break
through at Arras, but he gave no clue as to how this might be achieved.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

42 To Wa r 1914

Rupprecht dismissed the orders as ‘a purely amateurish view of the situation!’

He would have preferred to attack with his right wing, cross the River Lys,
and occupy the high ground south and east of Ypres. This offered better
positions in which to await Fourth Army, which could not arrive before
21 October at the earliest. Krafft had similar concerns. Further, the defensive
line Falkenhayn had chosen possessed only poor fields of fire and, at 40 kilo-
metres, was very long for the manpower available. Even the strongest corps
was at less than half strength and some were much worse off.1
Rupprecht’s subordinate commanders, such as the GOC IV Corps,
General Friedrich Sixt von Armin, were very worried about their ability to
break through at Arras. Sixt suggested complaining to OHL, but Rupprecht
advised against: this was OHL’s idea and therefore its responsibility. In fact,
Falkenhayn told Rupprecht and Krafft not to be too restricted by these
orders. They were to take any chances that came along. While the prince
welcomed the apparent freedom of action this gave him, Krafft was more
cynical. He saw this as Falkenhayn’s way of shifting blame if anything went
wrong: ‘this fully lived up to the scheming character of this man!’ In the
event, Falkenhayn’s plan nearly worked. It also led to the First Battle of
Ypres, one of the most famous battles of British military history.2
Over the week beginning 13 October, the British clashed with Rupprecht’s
men for the first time. One British corps advanced only hesitantly and made lit-
tle progress towards La Bassée. Further north, the British had better success. By
17 October they had overcome light opposition to capture Armentières and
the next day they gained a foothold on the Pérenchies Ridge. From here they
could look down on Lille, less than 8 kilometres away. Meanwhile, British
and French troops moved through Ypres and took up position on the
­crescent of gentle ridges east of the town. Sir Douglas Haig’s I Corps was
concentrating to the west, ready to advance despite intelligence warning of
a growing German threat. The BEF seemed to be playing so neatly into
German hands that Rupprecht found its movements hard to comprehend.3
Rupprecht first encountered British troops for himself in the courtyard of
Douai town hall. He was struck by their soldierly bearing, especially relative
to the dirty and ill-disciplined French. He intervened to stop the guards
‘requisitioning’ the excellent ‘British warm’ overcoats.Tactically, he found them
less impressive: at La Bassée on 14 October ‘the English blundered forwards
in columns and suffered heavy losses to artillery’. Nonetheless, the British,
together with the French Chasseurs Alpins, were the toughest opponents his
men faced.4 On 19 October Rupprecht issued an order of the day to his men:
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s43

Soldiers of Sixth Army!

We now have the fortune to be facing British troops on our front, troops of
the nation which, consumed by age-old jealousy, has been working to encircle
us with a ring of foes and throttle us. So, when you go forwards against this
enemy, make sure you make him pay for his malevolent treachery and the
heavy sacrifices he has caused us. Show him that the Germans cannot be
wiped out of world history so easily! Show him that with a specially German
thrashing! Here is the enemy who, more than any other, stands in the way of
a return to peace!
Onwards!
Rupprecht, Crown Prince of Bavaria5

This order provoked outrage in the London press. The Daily Mail inter-
preted it as a call to ‘Annihilate the English’. A leader in The Times quoted
it  as another example of deep-seated German Anglophobia. This order,
together with another a week later which described the British as ‘our
most  hated foe’, ensured that the London newspapers gave Rupprecht a
reputation for intransigence and bloodthirstiness which proved hard to shift.
The Daily Mail took to calling him ‘Bloody Rupprecht’ or ‘Rupprecht the
Bloody’. In May 1915 British newspapers gleefully reported allegations that
Bavarian soldiers had murdered British prisoners on Rupprecht’s express
orders, depicting him as the Super-Hun.6
The new German Fourth Army came down through Roulers and began
to press the French and British east of Ypres. Albrecht’s men showed their
enthusiasm and inexperience by often advancing without waiting for proper
artillery support. Consequently, they suffered heavy losses and made slow
progress, even against only light opposition. On 20 October Rupprecht
ordered his men between Armentières and La Bassée to attack, trying to
split the BEF and drive some of the British into the pocket Fourth Army
was creating. Over the next twenty-four hours Sixth Army pushed the
British back off the Pérenchies and Aubers ridges, regaining most of the
ground recently lost. On 22 October, however, the German offensive stalled.
The two sides dug in south of Armentières, their trenches just 100 metres
apart. Rupprecht was annoyed by the performance of his men and worried
how well Fourth Army would get on. Albrecht’s troops, he feared, were
poorly trained and led by rather old, retired, or reserve officers. How
effective would they be in full-scale combat?7
He soon had an answer. On 21 October, east and north of Ypres,
Albrecht’s men ran straight into the advancing French cavalry and Haig’s
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

44 To Wa r 1914

British I Corps. The British quickly dug themselves in wherever they were.
Albrecht’s attacks on Ypres the next day were bravely carried out but
poorly coordinated. They gained little ground for heavy loss. Over the
next two days his inexperienced reserve troops struggled to make progress
in the teeth of hostile fire. These were the battles which gave rise to the
later myth of the Langemarck Kindermord, the ‘massacre of the innocents’,
which supposedly saw battalions of student volunteers sacrifice themselves
for Germany, marching to their deaths singing ‘Deutschland, Deutschland
über Alles’.8
Fourth Army’s slow progress was beginning to make OHL nervous.
Albrecht was apparently stuck east of Gheluvelt on the Menin Road: he
seemed unable to dislodge British cavalry from the Messines Ridge. To the
north, the Belgian army was holding the line of the River Yser all the way
to the sea. South of the River Lys, Sixth Army was doing rather better. It had
pushed the British back some 5 kilometres on a 15-kilometre front. Horace
Smith-Dorrien’s II Corps, in particular, was feeling the strain, although fresh
men of the Indian Corps were now starting to join the especially ferocious
fighting around Neuve Chapelle. Nonetheless, Falkenhayn’s first attempt to
encircle the BEF had yielded little and his frustration and friction mounted.
To maintain what momentum he had near Ypres, he first stripped divisions
from Second Army and set them marching north. Then he told Krafft that
he intended to form an independent new command of some six divisions
under General Max von Fabeck to attack into the gap from Ploegsteert to
Gheluvelt, between Sixth and Fourth Armies.
When Falkenhayn began requisitioning Rupprecht’s divisions for Fabeck’s
command, however, the Crown Prince objected violently. He argued that
this would disrupt his promising attack south of the Lys. In fact, however,
progress there had begun to slow and Rupprecht was simply expressing
general anger. First, OHL was humiliating him by countermanding the
order he had just issued, which called for a continued energetic offensive.
Second, the scale of his command was being eroded. Third, he had heard
that Falkenhayn was being disrespectful about him behind his back. Lastly,
he resented a clumsy attempt to ‘bribe’ him with an offer of the Grand
Cross of the Iron Cross. If OHL tried to take more of his men, he resolved
to drive to see the Kaiser:
Either I command the army, or I resign. This cannot go on. Falkenhayn lets
himself be influenced by every Chinese whisper and jumps to conclusions
which are in every way damaging, which weaken the offensive spirit of the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s45

men and undermine their trust in their superiors. If only Falkenhayn would be
replaced by Colonel-General von Bülow or one of the senior generals. On the one
hand army commands are kept too much in the dark about the general
­situation, on the other OHL interferes in their business, instead of contenting
itself, in the manner of the great Moltke, with issuing general directives and
leaving the armies to carry out the missions they’ve been assigned.9

A showdown was only averted when Falkenhayn put Fabeck’s task force
nominally under Rupprecht’s command.
In reality, however, Sixth Army’s control over Fabeck was limited. It was
Fabeck, his chief of staff, Lieutenant-Colonel Fritz von Loßberg, and
Falkenhayn together who planned the next major attack for 30 October.
‘Only later, when the attack got more and more bogged down, did General
Falkenhayn once again involve the [Army HQ] in command.’ Having beaten
off British and French counter-attacks from 25 to 28 October, Falkenhayn’s
scheme involved Fourth Army crossing the Yser and bypassing Ypres to the
north while Fabeck swung south of the town. The two pincers would then
link up, surrounding part of the BEF.This was a much more limited objective
than in Falkenhayn’s earlier attacks. His ambitions had previously encom-
passed destroying all the enemy in northern France and Flanders. Partly, this
was because many of Fabeck’s units were weak. For instance, the battalions
of II Bavarian Corps were only 500–700 men strong.10
The German attack in the north stalled almost at once when the Belgians
opened the sluices in the dykes, allowing the sea in and flooding the land
from Dixmuide to the coast.To the south, Fabeck’s men drove to within just
5 kilometres of Ypres itself, although he thought the attack could have
achieved more if units had cared less for their flanks and pushed on more
aggressively. Rupprecht noted that ‘we seemed to have forgotten how to
fight on the move, how to outflank the enemy and how to stop him out-
flanking us by deploying in echelon—everything we learnt on staff rides
and the manoeuvre grounds’.11
The next day, 31 October 1914, proved one of the most dramatic of the
BEF’s war. It is most famous for the actions around Gheluvelt on the Menin
Road, where German troops, including Adolf Hitler’s regiment, captured
the village and tore a hole in the British line.The situation was only restored
by a brave charge by the 2nd Battalion Worcestershire Regiment across
1,000 metres of open ground. This recaptured Gheluvelt Chateau, halted
the German advance, and bought time for the situation to be recovered.
Important fighting also went on further south. Fabeck’s main attack aimed
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

46 To Wa r 1914

to seize the Messines Ridge and drive westwards on the high ground at
Mount Kemmel which dominated the whole salient. Success was only par-
tial, however. Bavarian soldiers seized but lost Wytschaete, while troops from
Württemberg gained no more than a foothold in Messines in the face of
opposition from, among others, the ‘London Scottish’ (1st/14th Battalion,
London Regiment), the first British Territorial soldiers to see significant
action. Heavy fighting continued along the Menin Road and on the
Messines Ridge over the next few days. It ‘had almost the savagery of the
Middle Ages’. By 2 November the lines had stabilized once more. Fabeck’s
men held Wytschaete and Messines but were unable to push further ahead.
Rupprecht watched the action from his battle headquarters at Linselles,
accompanied by Kaiser Wilhelm II and the famous Swedish explorer, Sven
Hedin. Meanwhile, his men launched local attacks further south, designed
to pin French forces in place and prevent reinforcements being sent north.
Among the few gains made was the chapel of Notre Dame de Lorette, cap-
tured on 2 November.12
In Flanders, the fighting was intense everywhere. It often occurred at
close quarters. In Ploegsteert Wood, for instance, the two sides were only
thirty paces apart. Casualties were heavy.The Germans probably lost 80,000
men in Flanders between the middle of October and beginning of
November. On 3 November, their heavy artillery, now ranged on three
sides of Ypres, began to bombard the town. Among the buildings des-
troyed, controversially, was the beautiful medieval Cloth Hall. In his diary,
Rupprecht tried to argue that the guns were targeting the railway station
and claimed he gave orders to preserve the Cloth Hall. This seems disin-
genuous. The station and Cloth Hall were only 400 metres apart. The
destruction of the Cloth Hall may not have been intended, but the pos-
sibility must have been anticipated.13
After 3 November, the tempo of operations slowed somewhat, although
local actions remained frequent and sharp. For example, the British launched
eight counter-attacks on the factory in Ploegsteert Wood, five of them
during the night of 9/10 November alone. Generally, though, the weather
was wet and both sides were running short of artillery ammunition. Sir
Douglas Haig had so few shells that he sent a third of his guns to the rear.
Even so, on 5 November he had to ration the remaining field guns to
twenty rounds each daily. On the German side, too, all immediately available
stocks of field artillery ammunition had been expended. By 9 November,
half Fabeck’s heavy artillery was unable to fire.14
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s47

Between 3 and 10 November the Germans contented themselves with

pinning the Entente forces in place while they concentrated fresh forces
for a new attack. By now Falkenhayn had given up hope of a war-winning
victory on the Western Front, but he remained keen for some kind of
morale-boosting local success. He formed another ad hoc task force under
General von Linsingen, including a new composite corps, led by General
Plettenberg and incorporating units of the Prussian Guard. Their mission
was to drive up the Menin Road at Gheluvelt. Fourth Army would lend
support on the right; Fabeck and Sixth Army on the left. When front-line
commanders objected that ammunition was short and their men tired while
enemy defences were strong, OHL offered little sympathy: ‘Command
knows the problems very well: they must be overcome.’15
Again, coordination was poor. Fourth Army attacked on 10 November,
taking Dixmuide and gaining some ground near Bixschoote and Langemarck.
Plettenberg and Linsingen, however, postponed their assault by twenty-four
hours to allow extra time for reconnaissance.They advanced on 11 November.
The best known incident of the day was the attack of the Prussian Guard
north of the Menin Road. This briefly broke the British line near Polygon
Wood before stalling under heavy fire. Again a brave counter-attack restored
the situation,this time by the 2nd Battalion Oxfordshire and Buckinghamshire
Light Infantry. Fourth Army was apparently exhausted and unable to provide
much support. Units were down to between a third and a half of establishment
and Albrecht’s request for reinforcements elicited a curt ‘no’ from Falkenhayn.
Much of Sixth Army seems hardly to have attacked at all.16
The next day, Falkenhayn told a meeting of the chiefs of staff of each
army that, after a few more days’ work to capture Ypres, the army in the west
would dig in on the defensive for the winter. This would free up troops
to be sent east, where Austro-Hungarian demands for reinforcement were
becoming daily shriller. When Falkenhayn suggested another attack on 13
November, both Sixth and Fourth armies argued that their forces were too
weak to continue. Each corps of Sixth Army was down to about one-quarter
of its August strength. On 17 November Albrecht formally suspended offen-
sive operations. The following day, OHL ordered Sixth Army to take six
divisions out of the line to be transferred to the Eastern Front. As the
German official history explained, this ‘meant in reality, even if the words
did not really say so, the suspension of all offensive operations in the west,
forced partly by the exhaustion of the troops and lack of ammunition, partly
by the need to intervene in the east with strong forces’. Formal orders to go
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

48 To Wa r 1914

Figure 5.1.  Rupprecht setting out for a morning ride

over to the defensive followed on 25 November.Within days, eight divisions

had been sent to the Eastern Front, almost all of them from Flanders.17
Casualty figures are never easy to nail down, but in Flanders they were
certainly heavy. According to the British official history, the BEF lost 58,155
men.18 The French and Belgians had casualties of 50,000–85,000 and
18,000–19,000 respectively, while the Germans probably lost some 134,300
men, of whom 19,600 were killed. It seems conservative to suggest
Rupprecht’s army lost about 50,000 men overall between La Bassée and
Ypres. Around a quarter of a million men on both sides were probably
killed, wounded, or captured in this first clash in Flanders.19
The First Battle of Ypres was a tactical and strategic draw. A draw was
enough, however, to confirm the verdict of the Marne: Germany would not
conquer France in 1914. The grandiose schemes of envelopment which
both sides came up with had collapsed as Flanders filled with troops and
room for manoeuvre disappeared. In the encounter battle that ensued,
­neither army was initially able to batter a way through the rapidly petrifying
front. Increasingly, the fighting took on the character of desperate Entente
defence against German thrusts, until eventually the attackers were worn
out and forced to give up.The Entente defence held partly because they did
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s49

at least three things right. First, Anglo-French cooperation was better than
it had ever been before in 1914 or would be again until 1918. Liaison was
excellent and reserves were promptly sent where they were needed,
regardless of nationality. Second, the defence itself was boldly and skilfully
handled. With Ypres close behind and reserves scanty, the more the defend-
ers could hold in place, rather than give ground and allow the battle to
open up to manoeuvre, the better. When the Germans broke in, as a rule,
aggressive counter-attacks proved effective at re-establishing the front.Third,
when combat broke down into small-unit actions, as it often did, individual
junior officers and other ranks fought bravely.
The Germans also did several things wrong, however. At the tactical and
operational levels, as Krafft admitted in his diary, they had attacked on
too  narrow a front, in insufficient strength, and without enough artillery
support. The troops were over-tired. Lack of ammunition was critical:
no-one had anticipated how fast modern warfare would eat through supplies.
Falkenhayn erred in changing the place of main effort and axis of his attacks
three times. Heavy losses in junior officers hit the combat effectiveness of
the infantry, who became too reliant on the artillery clearing the way for
them. Exhausted staffs and units, often thrown together in ad hoc task forces,
struggled properly to coordinate their attacks.20
There was a more fundamental problem of intelligence and analysis at
work, too.The British official historian, James Edmonds, believed that German
intelligence overestimated the strength of Entente reserves and that this
contributed to excessive caution by the attackers. Douglas Haig went further:
he believed that the Germans’ nerve had failed just as they stood on the
threshold of victory on 31 October. If only they had pushed on,Ypres might
have been theirs.The real mistake which cost the Germans the Battle of Ypres
involved under-estimating Entente strength, however, not exaggerating it,
and it was made on 15 September. When Falkenhayn and Tappen decided,
even with their tired and weakened forces, to resume the offensive after the
retreat from the Marne, they set their army on the road to Flanders, to frus-
tration and failure, and to the hospitals and graveyards. By the end of the
First Battle of Ypres, Bavaria alone had suffered casualties of about 100,000
men, nearly half of those who had marched away in August.21
Before we leave 1914, let us consider in more detail how the combatant
armies responded to the shock of the new warfare.The first issue they faced
was how to manoeuvre tactically in the face of twentieth-century firepower.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

50 To Wa r 1914

The idea that the impact of modern weaponry came as a complete shock
has long been discredited. Militaries had obsessively studied the wars of the
previous two generations and knew full well that quick-firing artillery,
machine-guns, and rapid smokeless rifle fire might make it harder for infantry
to attack and seize ground. Nonetheless, the evidence of the Russo-Japanese
and other wars seemed to demonstrate that attackers could still overcome
the defence. All they needed to do was to adopt more open formations,
combine their own firepower intelligently with movement, and, most
importantly, win the battle of wills. With enough courage and skill, any
attacker could triumph.22
So, when commanders on both sides began to find their infantry having
trouble manoeuvring on the battlefield and getting pinned down by enemy
fire, their first response was a form of denial. After all, they thought, this was
a problem they had fully anticipated and trained their troops to overcome.
If attacks were nevertheless breaking down, then one of three things must
be true. Perhaps the defensive had indeed become invincible. Maybe they
had trained their troops badly. Or, the troops were failing to follow their
training. The generals could hardly accept that the first two possibilities
could be true.Therefore, the attackers’ tactics must be faulty.Thus, as we saw
in Chapter 2, the heavy casualties suffered by the Bavarian Foot Guards at
Badonviller (12–13 August) were seen as the result of impetuous officers
ordering their men to charge prematurely and contrary to their training.
Likewise, the failed French attack on Cirey (15 August) was blamed on the
poor tactics of a single divisional commander. It was easier to blame individuals
than accept that problems might be systemic and hard to resolve. No-one
ascribed failure to commanders’ shortcomings more enthusiastically than
Joffre, who sacked generals liberally. The idea that individuals had fallen
short of training standards remained a recurring theme in both French and
German explanations of defeat throughout the rest of 1914. For example, a
Sixth Army lessons-learnt report of 1 October listed seventeen detailed
areas for improvement but at the top of the list still featured that staple of
peacetime doctrine and exercises: artillery–infantry coordination.23
Discerning observers, however, were beginning to realize that they
had  been under-estimating the strength of the defensive. They saw that
they were going to have to change their approach and pay firepower more
respect. The French General Dubail suggested overturning some of the
principles of peacetime training. He argued for more night manoeuvres to
offset enemy firepower. Greater use should be made of counter-battery fire,
even if that meant less direct support for the infantry. On the German side,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s51

the failure to capture Maricourt on 27 September showed Krafft that field

artillery was useless against a dug-in enemy and it was heavy artillery which
was required in much greater quantities. Falkenhayn soon revised pre-war
artillery principles in the light of the new circumstances. He suggested new
approaches including more use of indirect fire directed by forward observers
and guns engaging less by line-of-sight.24
Both sides soon realized that, in addition to the tactical difficulties, there
were operational problems at work, too. Joffre thought the French army was
trying to do too much at once. It was dispersing effort when it should be
concentrating it. Rather than launching a series of feeble attacks up and
down the line, it would be better to proceed more methodically, focus effort,
and capture one objective after another. This would gradually wear down
the enemy until he broke. On the German side, Rupprecht still saw oper-
ations in more traditional terms. He had not yet realized that million-man
armies could not manoeuvre in the old way anymore. The space available
was too narrow; even the most improvised trench defences were awkward
obstacles; and the defender could always use the railways to plug gaps in his
line faster than the attacker could open them. Krafft’s view was closer to
Joffre’s, having already identified that the central problem of the Western
Front was the operational one of how to create and maintain momentum.
On 22 October he wrote that:
the operations are unfortunately beginning to stagnate once more. . . . It is quite
clear that what is not accomplished in the first rush can never be achieved on
the second, third or subsequent days. The enemy has learnt too well to dig
himself in quickly so that the struggle becomes an exhausting local battle from
village to village and hamlet to hamlet which after a few days drains away into
complete stasis.
This showed a good grasp of the character of the new warfare and the prob-
lems it posed, even if he had no solution to propose as yet.25
In the early days of the war the way both sides learned had been fairly
informal. Sixth Army’s initial deployment orders had included instructions
to pass on any lessons learnt as quickly as possible. Early in the war, if this
happened at all it often did so either verbally or in hastily scribbled notes
which have not survived. Thus, for example, we know from Rupprecht’s
diary that on 29 August he verbally ordered all corps to imitate the relatively
successful methods of III Bavarian Corps, waiting for heavy artillery sup-
port before attacking fortifications and deploying in depth when on the
defensive. His men were to garrison the front line only thinly, to provide a
security screen, while most of the strength was held in reserve ready to
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

52 To Wa r 1914

counter-attack enemy advances. Some of these ideas were reflected in the

first formal lessons-learnt report which survives in the Munich archives,
submitted on 31 August.26
Over the course of autumn 1914, learning became increasingly formal
and developed into a process. Once the front solidified, the German army
began systematically collecting lessons-learnt reports. On 11 December
Sixth Army told all of its corps to submit theirs by January. These were not
just about what had gone wrong. They were also to include examples of
good practice to be emulated.Thus, on 26 December Rupprecht sent all his
corps commanders a report on a successful local action to seize British posi-
tions at Festubert, stressing the importance of careful preparation and ‘exem-
plary tactical execution’. Captured enemy documents containing tactical
hints were translated and distributed. On 27 November, for example, Sixth
Army intelligence circulated the British 7th Division’s after-action report.
Enemy actions were also analysed closely. A report written by an observer
with the German 89th Grenadier Regiment on French attack tactics
between 21 and 25 December 1914 was passed up to First Army and thence
to OHL, who distributed it back down to all army headquarters in the west
on 4 January 1915. Sixth Army sent it on down to its subordinate commands
three days later. In other words, within two weeks the whole German army
had been informed of the latest French tactics. One of the most important
functions of the General Staff was to collate such reports and sift them for
valuable lessons to disseminate and incorporate into doctrine for both train-
ing and fighting.27
French army learning displayed the same mix of formal and informal
methods. After-action reports became increasingly important.Within forty-
eight hours of a local attack by French VIII Corps on 8 November, for
instance, the army commander had read and was commenting on their
report. This approach was supplemented by liaison officers attached to each
army who reported back to French supreme headquarters (Grand Quartier
Général: GQG), as well as all manner of unofficial friend-to-friend or
­colleague-to-colleague communication.Thus, on 9 November GQG issued
two doctrine notes on how best to use artillery. These were explicitly
designed to spread experience from other sectors of the front. In the much
smaller BEF, informal transmission of lessons learnt was easier, but, as we
have already seen in the case of the 7th Division's after-action report, formal
approaches were also gaining ground.28
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

To Y pr e s53

All three armies, therefore, after an uncertain start, had quickly appreci-
ated the need for radical change if they were to survive on the Western
Front battlefield. They had begun to develop a three-dimensional approach
to managing change. First, they used pre-existing informal networks to
­disseminate best practice, as they always had. In addition, however, they had
begun to develop coherent processes to capture and distribute lessons from
their operations. This second dimension worked vertically. Typically, infor-
mation was passed up from the bottom of the chain of command to the top,
where new ideas and doctrine were formulated and passed back down to
the men at the front. There was also, however, a third dimension, working
horizontally, where experience from one unit or sector was distributed
around the army, normally mediated by OHL or GQG. The scale and
­complexity of these learning processes, still in their infancy in the winter of
1914/15, would grow as the war went on.29
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

PART II
The Anvil 1915–16
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

6
A Difficult Winter

L ooking back now, the Western Front campaign of 1914 was clearly
decisive. But it had not been conclusive. Germany’s failure to break
Entente resistance, first on the Marne and then in Flanders, doomed her to
the long war she knew she could not afford. At the time, however, remark-
ably few members of the German high command were prepared to accept
this. Falkenhayn was one of those who could. He was unable now to imagine
how to gain a decisive victory either in the congested and fortified west or
among the huge expanses of the east. Therefore, he could see no military
solution to Germany’s problem. On 18 November he told Chancellor
Bethmann Hollweg that they must seek a diplomatic way out of their pre-
dicament. Negotiations should begin with Russia. Remove her from the
war, he suggested, and France might sue for peace. Even if the French fought
on, the resources freed up from the east would enable Germany to win in
the west.
This approach seemed unduly pessimistic at the headquarters for the
Eastern Front, known as OberOst. Here, the two ambitious and capable
officers who had already masterminded a string of victories, Field Marshal
Paul von Hindenburg and Lieutenant-General Erich Ludendorff, felt that
the German army could still win the war. A series of successful manoeuvre
battles could yet crush Russia and enable Germany to turn west.To win the
necessary victories, however, fresh troops and ammunition would be crucial.
Any new formations raised, they argued, should be sent to them. This fed
into the existing strategic debate within the highest reaches of the German
command. It also, however, formed part of a power struggle. Ludendorff and
Hindenburg were trying to bring down Falkenhayn, a man they detested,
and seize control of the army. Falkenhayn was relatively junior and not
much liked in the army anyway; his handling of the Battle of Ypres had
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

58 T h e A n v i l 1915–16

generated concern; and the idea of a compromise peace was anathema to

most of the German army.
Consequently, the OberOst fronde recruited an impressive list of sup-
porters. Generals commanding armies, such as Karl von Einem, Duke
Albrecht of Württemberg, Bülow, Gallwitz, and Crown Prince Rupprecht,
were quick to sign up. Rupprecht in particular had never been confident
that Falkenhayn was the right man for the job. His doubts had grown over
the autumn as victory kept slipping out of reach. Falkenhayn’s manner was
sarcastic and boorish and Rupprecht resented what he saw as unwarranted
interference in his own command. In fact, the three most senior heirs
to  German thrones—Rupprecht, Albrecht, and Crown Prince Wilhelm
­himself—all came out against Falkenhayn. Moltke and Bethmann Hollweg
both suggested sacking him. Even the Kaiserin (Empress) Augusta Victoria
wrote to her husband urging change. When Wilhelm II refused to move,
Hindenburg threatened to resign if Falkenhayn did not go. The Kaiser was
infuriated by this insubordinate ultimatum. Hindenburg, however, was too
popular to sack, so Wilhelm reached a compromise. Falkenhayn would keep
his position as chief of staff but lose responsibility for the War Ministry.
Wilhelm II compensated him with a promotion but four new recently raised
army corps were to be sent east.
This leadership crisis made it clear that Falkenhayn had lost the confi-
dence of the army. From now on, he would be entirely dependent on the
Kaiser’s support, with important consequences for the exercise of command
in the German army to which we shall return. It also showed Hindenburg
that he could challenge the Kaiser’s authority and survive. The reinforce-
ments took part in the Second Battle of the Masurian Lakes, fought between
7 and 22 February. This proved, as Falkenhayn had worried, indecisive.1
As 1914 faded away on the Western Front, the tempo of fighting
dropped. Both sides struggled to reorganize themselves. The Entente
powers tidied up their jumbled units in Flanders, with the French taking
responsibility for much of the Ypres Salient. The British Expeditionary
Force (BEF), meanwhile, side-stepped south and took over 35 kilometres
of line from Wytschaete down to Givenchy, near the La Bassée Canal. In
the German army, Rupprecht was given command of an ad hoc Army
Group covering the sector from the sea to the River Oise. He had Fourth,
Sixth, and Second Armies under his control. Heeringen took over First,
Seventh, and Third Armies in the centre and Crown Prince Wilhelm had
charge of Fifth Army around Verdun and the left wing all the way to
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

A Di f f icu lt W i n t e r59

Switzerland. With action over the winter likely to be fairly local in scale,
these Army Groups were designed to allow any reinforcements needed to be
shifted from one army to another without requiring Oberste Heeresleitung
(OHL) to get involved. It is not clear how thoroughly this idea was fol-
lowed through, however. As in August 1914, no provision was made for any
dedicated Army Group headquarters and this ad hoc arrangement was
unwound in early March, at least partly because better weather increased
the chances of larger operations again.2
Across Belgium and France, men dug. They deepened the scrapes hur-
riedly scrabbled away under enemy fire, linking them up into rudimentary
trench systems protected in front with barbed wire. The war was rapidly
taking on the character of a siege, Rupprecht noted with resignation. His
army was stripped of men. He had to send five divisions to the Eastern
Front and was left covering a 120-kilometre sector with fewer than one
infantryman per metre of front line, much thinner than pre-war rules of
thumb allotted. Reserves and ammunition were scarce. The weather was
rainy and cold and conditions for the soldiers in the line quickly became
abysmal. By 11 December, for instance, the trenches at St Eloi on the south
side of the Ypres Salient were up to a metre deep in water. Even on ‘dry’
land, men sank up to their ankles in mud with every step they took.3
For all the rain and mud, the end of the battles for Ypres did not mean
the fighting was finished. Far from it. Despite the general exhaustion, a
series of sharp and sometimes bloody actions followed as both sides strug-
gled to gain some kind of edge over their enemy.They patrolled, raided, and
launched local attacks to straighten the lines or gain points of tactical value.
They struggled for physical features, such as higher ground, but also for
psychological dominance. It was important to show the enemy who was
boss. Men fought also to try out new tactics or weapons, to keep up offen-
sive spirit and avoid lethargy, as well as out of ambition or the spirit of
competition. On some occasions, no doubt, they fought just for something
to do. The result was a low rumble of near-continual small-scale but often
deadly operations.4
In the middle of December, the level of violence stepped up. Between
14 and 16 December, French and British troops began probing Rupprecht’s
positions on the south side of the Ypres Salient. They made little progress.
The mud was thick and the German defences remained intact. At first these
attacks gave ‘the impression they were undertaken reluctantly and only for
the sake of doing something’, and indeed the British effort was half-hearted
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

60 T h e A n v i l 1915–16

by design. Nonetheless, losses were sometimes heavy. In front of their posi-

tions the 3rd Bavarian Division counted 350 French bodies, having lost
between eighty and ninety men themselves. By 16 December, however,
something more serious seemed underway. French probes were growing in
aggression and purpose. Rupprecht reckoned that in fact the enemy was
trying, but failing, to carry out a single coordinated attack. When prisoners
of war were brought in, they confirmed that a full-scale attack had indeed
been ordered. This was merely one of a number of attacks taking place all
along the front, part of a much bigger plan for major offensives in both
Artois and Champagne. Joffre’s intention was to take advantage of German
troop withdrawals for the Eastern Front to drive in the shoulders of the
German salient that bulged towards Noyon.5
On 17 December, French Tenth Army in Artois launched a major operation
to clear the Notre Dame de Lorette Spur and capture Vimy Ridge. Resources
were scarce. One French division possessed just fifteen pairs of wire-cutters
when they needed 100. Problems in coordination and lack of artillery
resulted in attacks being made piecemeal and on narrow fronts. The German
defenders of XIV Corps quickly regained any trenches lost on Notre Dame
de Lorette and the French made little progress. Within two days the attack
had been suspended. After a patch of poor weather, the French tried again
on 27 December, with no more success. Again the offensive was suspended.
Even now, the intention was to try again. Only a further fortnight of bad
weather finally killed off the idea. For paltry gains, the French had suffered
heavy losses. At no point were the German defenders unduly stretched.The
units on the spot were quite capable of containing the French assaults on
their own, without calling for help from Sixth Army.6
Rather than Lorette and Vimy, it was the heavy fighting between
Givenchy-lès-La-Bassée and Festubert which appears to have captured
Rupprecht’s attention. Here, on 20 December, men of the German 14th
Infantry Division attacked and defeated units of the Indian Corps, captur-
ing Givenchy and nearly 800 prisoners. Rupprecht heard that Indian troops
at Festubert had run away when the attack began, throwing away their
weapons. Soon afterwards, General Haig’s I Corps was brought in to replace
the Indians, who were thoroughly worn out after two long and bloody
months in the trenches.7
Also on 20 December, the French General Fernand de Langle de Cary
launched the operation which became known as the First Battle of
Champagne. The fighting here was long and brutal. It dragged on until
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

A Di f f icu lt W i n t e r61

20 March. The French lost nearly 100,000 casualties and, again, had little or
nothing to show for the sacrifice. Meanwhile, smaller-scale fighting con-
tinued on the Sixth Army front. It was a rare day which saw no action at all.
Some local truces did take place on Christmas Day, of course. In popular
myth these truces were all about swapping cigarettes and playing football.
The reality was normally grimmer and more pressing. Much of Christmas
Day was spent clearing corpses from the battlefield. Inevitably there was
some fraternization and there were occasional recurrences over the next
few days.There is even some evidence of French and German soldiers shak-
ing hands in No Man’s Land up on top of the Lorette Spur. This ridge, 150
metres high, which dominates Vimy Ridge, Lens, and the Douai Plain, was
one of the most hotly contested places on the entire Western Front. It was
the scene of near constant and intense close-quarters fighting throughout
1914 and 1915. The trenches were always within shouting distance of
each other and often nearer. Any truce up here could not last long. Artillery
fire soon broke it up. Especially heavy fighting took place at Lorette on
27 December, 15–24 January, 3–8 March, and again on 15–21 March.8
Other hotspots throughout the late winter of 1914/15 were at Écurie,
north of Arras, and at two places on the BEF front: the Cuinchy-Givenchy
area either side of the La Bassée Canal and near St Eloi, south of Ypres.
Trench fighting, however harsh, was inevitably not as bloody as mobile
warfare had been. Sixth Army casualties in December were less than half the
monthly average in August to November. In 1914, Sixth Army lost, on aver-
age, 37,630 men per month in combat. In December 1914 the number of
casualties fell by more than half to 15,799. Sickness rates, however, remained
high. No fewer than one in ten of Sixth Army’s total strength reported sick
in December 1914. Illness was a consequence of the armies still learning
how to live in the trenches. The primitive trenches offered only poor pro-
tection from bad weather and sanitation arrangements were as yet only
rudimentary. Learning how to live in such conditions was an immense chal-
lenge for all the combatants. For many, this first winter in the trenches was
the most miserable of them all.9
With the approach of spring the ground began to dry out and planning
for larger-scale operations could begin. With the Germans busy in Poland, it
was clear to all that the initiative on the Western Front lay with Marshal Joffre
and Sir John French. Rumours of British amphibious landings in Flanders
had come to nothing, but German intelligence knew that Canadian troops
were already arriving in France and thought fresh divisions of Kitchener’s
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

62 T h e A n v i l 1915–16

New Armies were, too.What they did not know was that British formations
which had been promised to the BEF in France were now being diverted
to the Mediterranean for the Dardanelles expedition. This prevented Sir
John French from taking over more of the line as Joffre requested. Out of
frustration, and to show that the British were doing something to help, he
ordered an attack at Neuve Chapelle for 10 March 1915. An advance there
would have the ­benefit of enabling his men to move out of the waterlogged
valley of the River Lys and up onto the higher and drier La Bassée-Aubers
Ridge. If they could do that, not only would they no longer be overlooked
by the enemy, but in turn British observers would be able to see all the way
to Lille.10
The Battle of Neuve Chapelle has always enjoyed a prominent place in
the history of the BEF. The British official history, for instance, dedicates
eighty pages to it: a lot for a fight which lasted three days and involved just
four British divisions. The German official history, in contrast, devotes only
two. Neuve Chapelle attracts attention from Commonwealth historians for
four reasons. It marked the first time during the war that the British army
launched a major set-piece offensive against a dug-in enemy. It was, there-
fore, a landmark in trench warfare, for the British at least. Second, it dem-
onstrated that it was possible to break in to the enemy’s fortifications. In
retrospect lessons can be discerned from Neuve Chapelle which would
prove important in the long run, even if they were not universally accepted
in the near term. Third, it greatly influenced British planning over the rest
of the war. Lastly, it was relatively successful, certainly compared with most
of the other British battles of 1915.11
The British attack at Neuve Chapelle surprised Rupprecht. His attention
had been further north, overseeing plans for an attack at St Eloi. Rain had
prevented German aerial reconnaissance, so he received no prior warning.
The first reports to reach Sixth Army headquarters in the morning of 10
March spoke of two attacks: one at Givenchy, another at Neuve Chapelle.
The former, carried out by the British 2nd Division, failed badly. British
artillery failed to cut the wire and the defences remained largely intact.
When the assaulting battalions went over the top, German machine-guns
cut them down in No Man’s Land. The Givenchy attack, though, was only
one of several attempted diversions designed to conceal the main effort.
This was made by the IV and Indian Corps on the village of Neuve Chapelle
itself. General Sir Douglas Haig was in overall command.
The German positions here were weak. First, they lay in a salient.The British
could fire in on them from almost three sides. Second, since the water table was
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

A Di f f icu lt W i n t e r63

less than a metre below the surface, the front-line trench was largely built
up with a breastwork of sandbags, a metre and a half high and nearly 2
metres thick. A double row of barbed wire between 2 and 5 metres deep ran
in front. The village itself was in ruins but not especially fortified. Further
back from the front ran the old British ‘Smith-Dorrien Trench’ of October
1914.This had not been maintained and was largely waterlogged. A thousand
metres behind, the Germans had built a line of concrete machine-gun nests.
These were spaced some 800 metres apart. They were to serve as support
points in case the enemy broke through the front-line defences. The com-
manders on the ground had asked for permission to link up these strong-
points with a continuous trench. Their requests had been refused, however,
on the grounds that the front-line defenders would not fight as hard if they
knew a second line lay behind them.12
The British launched the first assault with three brigades on a 2,000-
metre front. They only had enough shells for a short and sharp bombard-
ment. So, after 276 field, and sixty-four heavy, guns had dropped 3,000 shells
on the German defences in just thirty-five minutes, the infantry would
attack. Zero Hour was set at five past eight in the morning. Everything
depended on the guns. Where they had sufficient time to register their fire
properly, they cleared much of the barbed wire, knocked down large
stretches of the breastwork and also silenced the German artillery. In these
sectors the British infantry advanced swiftly. They overran five of the eight
German companies in the front line and swept through the village. A pro-
tective barrage to the east of Neuve Chapelle prevented the intervention of
German reserves. After a little over three hours the British had captured the
German front line and broken into their defences. Rawlinson wanted the
men of his IV Corps to advance on towards Aubers Ridge. Three major
problems arose, though. First, the assault units were predictably disordered.
They needed to reorganize before they could push on. Second, they had
outrun their telephone wires and could only be contacted by runner.
Inexperienced staffs under-estimated how difficult this could be on the
lethal and disorientating First World War battlefield. In fact, Rawlinson’s
orders never got through to the front brigades. Nonetheless, some of these
moved forward again on their own initiative, only to be hit by shells from
their own artillery, which was aiming at the German strongpoints. Third,
German reserves had managed to throw together a defensive position
around the machine-gun casemates. Thinly manned and sketchy as it was, it
more than sufficed to contain the uncoordinated attacks that were all the
British could manage before darkness fell.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

64 T h e A n v i l 1915–16

The British had mixed views on what they had achieved on 10 March.
Some, like Rawlinson, were delighted to have captured 748 prisoners and
penetrated 1,200 metres on a front about 4 kilometres wide. As he wrote in
his diary: ‘We have had a grand day—our plans succeeded admirably . . .
Douglas Haig is certainly pleased and we have received the congratulations
of Sir John French.’ Others, including Haig himself, were more downbeat,
convinced that Rawlinson had missed the chance to do more.13
From the German reaction, it is clear that they took this attack seriously.
As soon as General von Claer, commanding the defenders from VII Corps,
heard that Neuve Chapelle was lost, he asked Sixth Army headquarters for
reinforcements. Rupprecht sent more than a division of infantry and four
heavy artillery batteries to his aid on 10 March. The next day, when a cap-
tured copy of Haig’s orders showed the full scale of the British attack,
Rupprecht despatched another seven battalions. On 10 March the British
had outnumbered the Germans by forty-eight battalions to twelve. Within
twenty-four hours they faced thirty-two. The Germans had also strength-
ened their defences overnight, piling up breastworks and wire along much
of their new line.
Although Rawlinson and Haig were keen to see a further advance on
11 March, communication difficulties continued to make coordinated attacks
tricky. Moreover, visibility was poor and the artillery had been unable prop-
erly to register their fall of shot.The consequence was weak or non-existent
artillery preparation, piecemeal British attacks, and ‘a fiasco’. No progress
was made. Attacking units were swept away by machine-gun fire as soon as
they went over the top. The commanding officer of at least one battalion
refused to order his men to waste their lives. By mid-afternoon the artillery
fire was dying away. In the morning, Rupprecht had been able to hear only
a continuous roar from the guns. Now he could pick out individual shots.14
Rupprecht was close enough to hear well because he was visiting Claer’s
headquarters at Marquillies, 9 kilometres from the fighting. Here he could
avoid repeated jittery telephone calls from Falkenhayn and OHL.15 More
importantly, he could review VII Corps’ plans for an attack to win back
Neuve Chapelle on 12 March. Rupprecht suggested to Claer that there was
too little time for proper artillery preparation. Perhaps a further twenty-
four hours’ delay was called for? Claer disagreed. He argued that another
day would give the British time to dig in more firmly. In any case, to pre-
pare a full set-piece attack properly would take weeks. There was no point
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

A Di f f icu lt W i n t e r65

in waiting for all the heavy artillery reserves in transit: there were not
enough suitable observation posts for him to use them all, anyway. Rupprecht,
somewhat reluctantly, approved Claer’s plan. He clearly did not want to
interfere with decisions made by the man on the spot. In Krafft’s opinion,
this was a mistake and Claer should have been ordered to wait.16
Rupprecht and Krafft were right to be worried. After a thirty-minute
artillery bombardment, Claer’s attack went in at six a.m. on 12 March. His
infantry found that the guns had largely missed the defences. Vickers
machine-guns scythed swathes in the attacking units, which hesitated and
then fell back. Within two hours, the assault was abandoned. On 13 March
Rupprecht noted the British were bringing up wire and seemed to be con-
solidating the positions they already held, rather than gearing up to attack
again. Indeed, late on 12 March, Haig had ordered Rawlinson to suspend the
attack. The battle for Neuve Chapelle was over.17
The British had eliminated an enemy bulge into their lines and demon-
strated that, with careful preparation and sufficiently accurate and heavy
artillery preparation, they could break into the enemy’s fortifications. The
battle proved that the scale of any assault must be proportional to, and deter-
mined by, the amount of heavy artillery on hand and the availability of
ammunition. In three days Haig had used up seventeen days’ worth of total
UK shell production. The battle had also shown, however, that the difficult
bit was how to follow up a break-in. Rawlinson seems to have understood
that he had been wrong to carry on hammering away with increasingly
disorganized forces against ever stronger defences on 11 and 12 March.
Instead, from now on he argued for a policy of ‘bite and hold’, seizing a
small piece of the enemy’s position and defeating the inevitable counter-
attack. Some historians have seen in this limited-objective ‘bite-and-hold’
approach a genuine operational alternative to the search for that elusive
breakthrough. Consequently, Neuve Chapelle appears to mark an important
stage on the British army’s ‘learning curve’. As we shall see, however, if any
conceptual advance had indeed been made, it was far from fully accepted
and internalized, even by Rawlinson, much less by Haig or the rest of the
BEF. Practice lagged even further behind.18
The Germans were learning lessons, too. Neuve Chapelle got right to the
heart of the two core doctrinal debates about how to mount the defensive.
First, should the defence be rigid and forward, or elastic and in depth? The
former aimed to stop the attacker dead, ideally literally so, in No Man’s
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

66 T h e A n v i l 1915–16

Land, while the latter absorbed pressure, falling back until, exhausted and
extended, he could be counter-attacked and defeated. Second, should the
defenders counter-attack to win back lost ground and, if so, when? To begin
with, on 11 November 1914, Falkenhayn had laid down as a basic principle
that ‘to guarantee holding our position, the front line must be held with all
resources’.19
This principle, embodied in the catchphrase ‘Was gewonnen ist, muß
unbedingt gehalten werden’ (‘What’s been won, must be held at all costs’),
was the foundation upon which the whole German approach to the defen-
sive was built in late 1914 and the early months of 1915. This came through
in a series of orders and memoranda issued by both Sixth Army and OHL.
Nonetheless, where there was a danger the enemy might break through, a
second line should also be prepared. Quite how this second line might fit in
to any scheme of defence provoked debate. When Rupprecht and others
expressed concern that the front-line garrison might be less steady if they
knew there was a second line behind them, Falkenhayn reiterated that any
fall-back position was no more than a contingency measure. He had no
intention of mounting a more flexible defence unless he absolutely had to.
He repeated that it was crucial to hold the front line as firmly as possible.
The failure of the French winter offensives in Artois and Champagne
seemed to vindicate this principle.20
Rupprecht thought that at Neuve Chapelle VII Corps had left too many
men in the front line where they were vulnerable to enemy artillery and
could be overrun. Once the front line was lost, it had been a mistake to defend
as far forward as Claer’s men did. Rather than digging in along the line of the
blockhouses, it would have been better to conduct the main resistance further
back, up on Aubers Ridge itself. Here reserves could more easily assemble for
a decisive counter-attack. In fact, the more Rupprecht thought about it, the
more he felt Neuve Chapelle was an unimportant village not worth retaking.
Not only had Claer counter-attacked too soon, but he had probably erred in
counter-attacking at all. It would have been better to preserve strength and
launch a set-piece attack in his own time. It was obvious that the 12 March
counter-attack had failed for want of preparation.21
OHL’s comments on Sixth Army’s battle report found fault with VII
Corps’ execution of its counter-attack and blamed Rupprecht for not
having forbidden it, to his predictable annoyance. Otherwise, however,
Falkenhayn seemed fairly happy with the outcome. It was and is hard to
avoid the conclusion that the British had little to show for their efforts. BEF
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

A Di f f icu lt W i n t e r67

casualties were 583 officers and 12,309 other ranks. The British official his-
tory estimated similar losses for the Germans, but they were probably lower.
The German official history estimated casualties of ‘almost 10,000 men’ and
Rupprecht’s diary and Sixth Army records suggest a final total around 8,500.
Whatever the true figure, Aubers Ridge remained out of reach and u­ nfinished
business for the British.22
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

7
A Successful Spring

T he war in the East devoured men remorselessly through the early

months of 1915. The reinforcements combed out from the Western
Front after Ypres had proved unable to shatter the Russians at the Second
Battle of the Masurian Lakes (7–22 February). Any German progress was
undercut by repeated Austro-Hungarian defeats. On 23 March the great
fortress at Przemyśl surrendered to the Russians, its garrison of over 120,000
men heading into captivity. The Habsburg regime seemed on the brink of
military collapse. More German troops were desperately needed. Where,
however, could they be found? The field army was already consuming
180,000 replacements per month. The answer was to reduce existing divi-
sions from four regiments to three and top up the remaining regiments with
fresh replacements so that overall rifle strength remained the same.
Simultaneously, field artillery batteries were cut from six guns to four. The
infantry and artillery freed up in this way were then formed into fourteen
new, but experienced, divisions.1
There still remained nowhere in the west to employ these forces to any
purpose. Between February and April Falkenhayn solicited proposals for
offensives from some of the staff. The responses he received all required
unrealistic amounts of manpower and heavy artillery. Krafft, for instance,
suggested attacking with fifteen new army corps on a 25-kilometre front
either side of Arras, supported by nearly 200 heavy batteries. Such an oper-
ation would have dwarfed even the Battle of Verdun a year later. Indeed, not
until March 1918 could the Germans concentrate so much artillery for a
major assault. It is unclear whether Krafft was setting the bar so high on
purpose to imply that an attack was out of the question.2
Falkenhayn was already convinced that ‘we won’t see a war of movement
now again. It’s just trench warfare. We cannot make a breakthrough, because
then the enemy can breakthrough somewhere else.’ Rupprecht’s response
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng69

when Falkenhayn said this was brutally direct: ‘if that’s your view then we
must get a peace as soon as possible and at any price’. Rupprecht had already
begun to show signs of pessimism at the end of November. He had warned
Krafft then that the weight of enemy manpower, especially from the British
Empire, now meant they were in for a long war. Unless they could restore
mobility, the Entente would slowly squeeze the life from Germany. He also
complained that stress was turning his hair prematurely grey. As we shall see,
however, Rupprecht’s view of the chances of victory changed rapidly and
frequently during the war and otherwise he remained upbeat. Indeed, when
Rupprecht heard rumours that his father was considering asking the Kaiser
to seek an early peace, he rejected this defeatism. His reply to Falkenhayn
seems intended more to discomfit a superior he disliked and distrusted than
to constitute a serious proposal for peace.3
The fall of Przemyśl soon rendered further discussion moot. The newly
raised divisions had to go east to prop up Austria-Hungary with a fresh
offensive at Gorlice-Tarnow. This was planned for 2 May. A major attack
might be beyond reach in the west, but that did not preclude operations
with more limited objectives, especially if they might divert attention from
the main German effort against Russia. The most famous of these was the
attack on French and British troops north-east of Ypres on 22 April 1915.
This marked the first use of lethal gas on the Western Front. Weeks of
violent fighting ran on almost to the end of May, in what became known as
the Second Battle of Ypres. About another 100,000 men were killed,
wounded, or captured.4
Rupprecht had no direct role in the planning or execution of the Second
Battle of Ypres. He knew all about it, though. On 1 March Dr Fritz Haber,
mastermind of the German chemical warfare programme, had visited
Rupprecht’s headquarters. According to Rupprecht’s published diary, he
warned Haber and Falkenhayn that they were gambling against the prevail-
ing wind and, more importantly, with Germany’s reputation. How could
they be sure that the British or French chemical industries might not catch
up with the Germans? Tellingly, this passage does not exist in Rupprecht’s
manuscript diary. It is a later insertion, designed to distance Rupprecht from
the adoption of a controversial and unsuccessful weapon. The manuscript
diary shows that he fully supported the use of gas, seeing it as a legitimate
reprisal for British atrocities and expecting the German chemical industry
to be able to outperform that of the Entente countries. He thought the
attack itself was poorly executed: it took place on too narrow a front, too
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

70 T h e A n v i l 1915–16

late in the day for proper exploitation, and there were too few reserves to
hand. Still, German pressure forced the Entente back some 5 kilometres, to
within 3,000 metres of the walls of Ypres.5
Rupprecht toured Flanders and the captured trenches in early May. He
was unimpressed by the enemy’s standards of engineering and hygiene. The
trenches were too wide and looked unfinished. They were too shallow and
there were not enough dugouts. Communications trenches were in a poor
state and there was not enough wire. British trenches were a little better
than French ones but still very dirty. Food leftovers, tins, and jars littered the
dugouts. There were no latrines nearby. Corpses lay buried in the trenches
and breastworks with their arms and legs sticking out. Dead bodies from the
November battles still lay about in the open all over the place. The contrast
with the German trenches Rupprecht toured the next day was dramatic:
here ‘everything was excellently maintained and lovingly tidied’. Telephone
wires ran in neat channels in the trench walls; signposting was comprehen-
sive and artfully painted; here and there little gardens were planted with
snowdrops and other spring flowers. ‘Everything was as homely as possible,
and the men were in the best possible spirits.’ The Germans had buried
British dead from mid-April in a small cemetery on the side of the hill
under simple wooden crosses. The village of Zonnebeke, by contrast, was ‘a
picture of devastation’. In places the foundations were all that was left. The
bell tower of the church was split from top to bottom. The roofs of Ypres
could not be seen through the mist, but Rupprecht was able to make out the
succession of low ridges which made up the terrain of the salient. The
subtle swells of the ground hardly showed on the map but offered good
fields of fire to the defender, even if in 1915 enough trees still survived to
restrict visibility in places.6
From the last days of April onward, German intelligence began picking
up ominous signs of increased enemy activity opposite Sixth Army. By 2 May
aerial reconnaissance had shown that the British were digging new assem-
bly trenches opposite Aubers Ridge. Further south, French artillery fire was
increasing almost day by day. The sector around Vimy Ridge held by
I  Bavarian Reserve Corps was especially heavily shelled. Sixth Army felt
stretched thin. It had 90 kilometres of front to hold with thirteen divisions
in the front line and just two in reserve. Artillery support was provided by
858 guns, of which 174 were heavy. When, therefore, on 21 April, Rupprecht
was told to give up a regiment from the division responsible for defending
Vimy Ridge, he protested. ‘I told the Chief of the General Staff that I was
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng71

not in a position to accept responsibility for the weakness of the front and
that I took no responsibility for the consequences.’ Falkenhayn replied: ‘If
your Royal Highness cannot bear the responsibility, I will just have to add it
to my already heavy load. It cannot be changed.’ On 8 May French engin-
eers were spotted digging saps forward into No Man’s Land and clearing
lanes in their own wire. Clearly, an attack was now imminent.7
At quarter to five in the morning of Sunday 9 May, Rupprecht was woken
by a bomb blast near his headquarters in Lille. A bomb had dropped near
the garden wall of his old billet. Rupprecht had recently moved out after a
newspaper published its location, but he did not think this was an assassin-
ation attempt. More probably the enemy was trying to destroy the nearby
building which housed his telephone exchange to cut his communications.
Other bombs later fell on nearby railway stations. At headquarters, the only
real damage was to the composure of the cook. The blast had thrown him
out of bed and he was so badly shaken that he made ‘dreadful coffee’ that
morning. Rupprecht retained his poise better than his chef, rising and dress-
ing as usual at seven o’clock. By then his army was already under attack.8
First into action were the British of Sir Douglas Haig’s First Army. They
launched two attacks some 5,500 metres apart, designed to converge on and
capture Aubers Ridge. In the north, Rawlinson’s IV Corps aimed to break
through the 6th Bavarian Reserve Division near Fromelles, while further south
the 1st and Meerut Divisions attacked the Westphalians of the 13th Division
on a 2,200-metre front south of Neuve Chapelle. A short but sharp bom-
bardment by 637 guns began at six o’clock. The artillery hoped to cut the
wire and destroy both the front line and rear strongpoints. The forty minutes
allotted was nowhere near enough. None of these objectives was achieved
before the artillery fire, as planned, began to move off hundreds of metres
into the German rear. The British were still forming up for their assault.
When they did finally advance across No Man’s Land, they found uncut
wire and largely intact defences. The Germans were fully alert and poured
rifle and machine-gun fire into the British ranks. Any small groups of
attackers which managed to get into the German trenches were unable to
hold out long before being overwhelmed. Most survivors found themselves
pinned down in No Man’s Land, unable to move either forward or back.
Within twenty minutes the southern attack lost over 3,000 troops and had
entirely broken down. The northern pincer had done little better. Again,
some isolated companies had seized short stretches of the front line, but they
were fighting for their lives. The Germans laid down murderous fire on No
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

72 T h e A n v i l 1915–16

Man’s Land and the British trenches to prevent reinforcements coming up.
The 2nd Battalion Devonshire Regiment, for instance, suffered 30 per cent
casualties without ever going over the top.9
Repeated British attempts to get the attack moving again later that day
were either stillborn or equally abortive. Such chaos reigned in the commu-
nications trenches that it was impossible to relieve the shattered units in the
front line, even under cover of darkness. The few pockets hard won in the
German lines were pinched out during the night. When the casualty returns
came in, Haig had to accept defeat. Further attacks were suspended for now.10
The British were shocked by their defeat at Aubers Ridge. Confidence
had been high. They thought they had learnt the lessons of Neuve Chapelle
and had worked hard to improve planning and communications. Reserves
were closer to hand, ready to deploy when needed. The reasons for failure,
though, lay elsewhere. The British official history suggests that the initial
forty-minute bombardment proved too short. Planners had failed to appre-
ciate how much the Germans had built up their defences after Neuve
Chapelle: they had doubled or tripled the thickness of breastworks and
increased the depth of barbed wire, for instance. This explanation captures
some, but not all, of the truth. A longer bombardment would probably not
have solved the problem without measures to improve the accuracy and
effectiveness of the British artillery. There was no allowance made for the
reduced accuracy of worn gun barrels and insufficient time was allotted for
target registration. More importantly, as Robin Prior and Trevor Wilson
point out, the British failed to realize that, with fewer guns per metre of
front and effort split between more than one German defence line, the
intensity of their shelling was just 20 per cent that of 10 March. The tactical
consequence was intact defences. Expecting the infantry to march forwards
almost without having to fight in these circumstances was unrealistic. Aubers
Ridge also displayed a weakness in British operational art, which had not yet
internalized the need to match the scale of ends to the availability of means.
This applied in particular to heavy artillery but had nothing to do with the
shortage of shells Sir John French alleged in the political crisis over muni-
tions he provoked back home, however. Plenty of ammunition remained
available the next day. Although the ‘shell crisis’ hit on the deeper truth that
war production and procurement needed reform, its origin was a sordid
attempt by Sir John French to obscure his own command mistakes and
counter likely criticism from Lord Kitchener, the Secretary of State for War.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng73

Overall, Aubers Ridge, as Professor Gary Sheffield has written, ‘most closely
matches the stereotype of the First World War battle: soldiers advancing
across No Man’s Land only to be cut down by machine-gun fire; generals
frustrated by the lack of progress ordering further fruitless attacks; minimal
gains for huge losses’.11
The contrast between British failure and French success on 9 May was
stark. Aubers Ridge was the British Expeditionary Force’s (BEF’s) contribu-
tion to a much larger French offensive, the so-called Second Battle of Artois.
It is clear from Rupprecht’s diary that the best efforts of the BEF caused him
little or no concern on 9 May. Those of Victor d’Urbal’s French Tenth
Army, on the other hand, on the 20-kilometre front from the Lorette Spur
down to east of Arras, worried him deeply. Word of trouble began to come
in during the morning, with reports that Lorette was under such heavy
bombardment that the entire hill was shrouded in clouds of dust. An enemy
attack also seemed to be underway against I Bavarian Reserve Corps, whose
5th Bavarian Reserve Division held the front from Carency down to
Neuville St-Vaast, about 4,000 metres west of the crest of Vimy Ridge. The
1st Bavarian Reserve Division extended the line south around Arras. The
German front line constituted a network of trenches up to 150 metres deep.
A second line, 2,000 metres or so back, had been begun but was not yet
continuous. Many of the German defences had been badly knocked about
by a methodical French artillery preparation from 3 May onwards. At seven
a.m. on 9 May the French guns began firing once more with growing
intensity. Four hours later, the infantry assault began. As the French infantry
moved forwards, the guns either fired a rolling barrage in front of them or
shifted their attention to the second line. Five French corps took part in the
attack, supported by 782 field and 293 heavy guns. In all, the artillery
dropped 265,430 shells on Sixth Army’s positions between 3 and 9 May.12
By 12.30 on 9 May it was clear that the French had broken through the
Bavarian line. Souchez and Vimy Ridge were under threat. If they lost
Souchez, the Germans would have to evacuate the Lorette Spur. The loss of
Vimy Ridge, the last high ground before the Douai Plain, would be even
more worrying. Rupprecht asked Oberste Heeresleitung (OHL) for two
divisions to prop up his reeling Bavarians. OHL agreed at once. Reserves
would not be able to help before nightfall, however. In the meantime,
Rupprecht would have to scrape together what units he could to contain
what was increasingly looking like a major French breakthrough. No-one
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

74 T h e A n v i l 1915–16

thought to send reinforcements to Aubers Ridge, which further confirms

the insignificance of the BEF attack.
In fact, at around noon men of the French Moroccan Division were
digging in on the crest of Vimy Ridge.13 The Moroccan Division had
advanced 5 kilometres in ninety minutes. This was so remarkable an achieve-
ment by 1915 standards that d’Urbal’s staff officers could not believe it at
first. On the Moroccans’ left the 77th Division of General Ernest Barbot
made similarly spectacular progress, occupying Souchez cemetery.14
Their success was based on three foundations. First, the attackers did
several things right. Planning was painstaking and detailed. The artillery
preparation had cut the wire so well and battered the trenches so thor-
oughly that in places the only way to move in daylight was on all fours. The
French also used mines to blow large holes in the defences. The assaulting
troops moved fast, leaving strongpoints to be mopped up by subsequent
waves. Second, the men were well trained, experienced, and had high mor-
ale, partly due to good leadership. The commander of the Moroccan
Division, General Ernest Blondlat, for instance, issued an order of the day
which seems to have inspired his men: ‘I am your father, you know that
I  love you like my own children, if you want to show me that you love me
too, fight well and kill as many Boches as you can.’ General Barbot seems
equally to have personified French élan. When he was mortally wounded
on 10 May, his dying words are said to have been ‘Beautiful . . . Glorious
day . . . Victory.’ Third, the Germans they defeated had some very specific
problems. While some defending units were on the alert, others chose to
interpret French attack preparations as a diversion and were caught by
surprise. In the crucial Vimy sector, the 5th Bavarian Reserve Division had
been reorganized, with one of its original regiments replaced by a Landwehr
regiment full of older men transferred from occupation duties in Belgium
who resented their transfer. When the French attacked, this Landwehr
regiment broke and ran.15
Things were going badly for Rupprecht in the centre but his men were
doing better on the flanks, holding the French on the Lorette Spur and
maintaining a firm grip on key villages such as Ablain-Saint-Nazaire,
Carency, Souchez, and Neuville-Saint-Vaast. The Labyrinth redoubt also
remained in German hands. Rupprecht’s men were thus able to fire from
both flanks and make it difficult for the French to reinforce success on Vimy
Ridge. The defenders gathered together an ad hoc force to seal off the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng75

French penetration by early afternoon. Under pressure from three sides the
men of the Moroccan Division had been pushed back down the slope of
Vimy Ridge by six o’clock that evening.16
The next day, 10 May, combat raged from house to house, sometimes
from room to room, through the ruins of Neuville-Saint-Vaast, Ablain-
Saint-Nazaire, and Carency. Equally vicious hand-to-hand struggles were
fought out in the trenches of the Labyrinth, along the Souchez-Arras road
and up at Notre Dame de Lorette. Both sides were too disorganized, how-
ever, to put together any coordinated attack. It was all many units could do
to hold in place. Reliefs and resupply proved extremely difficult. In forty-
eight hours of action, the Moroccan and 77th Divisions received almost no
food, water, or ammunition. The French tried to improvise an attack in the
afternoon of 11 May but it was unrealistic to expect their exhausted men to
be able to restore momentum to an attack which had stalled, especially at
short notice, with limited intelligence and poor communications. Most
operations between 10 and 15 May remained small-scale and local.17
That did not mean they were not bloody. The Moroccan Division lost
over 5,000 men in three days. One German regiment lost all three battalion
commanders and nine of its twelve company commanders within thirty-six
hours of coming into battle. Some of its companies were left just thirty men
strong. Sixth Army counted 20,000 casualties up to 13 May.18
Partly to fill the gaps left by these casualties, but partly also with an eye to
retaking the lost ground, between 9 and 14 May Sixth Army received
reinforcements of four and a half infantry divisions, plus artillery. This was
more than half the reserves OHL held on the Western Front. By 18 May
another three and a half divisions had been sucked in, and OHL had just
one division left in reserve for all of France and Flanders. This demonstrates
how seriously the Germans took the threat in Artois. Tensions within the
German high command further underscore the point. At noon on 11 May,
Rupprecht received this telegram from OHL:
At the very least, his Majesty expects the army to hold its present line under
any circumstances. Whether this objective is accomplished defensively or with
secondary attacks must be judged locally. In any case, with reasoned deploy-
ments and firmly coordinated leadership, the available infantry and heavy
artillery ought to be strong enough to prevent a continued enemy advance.
His Majesty acknowledges the outstanding bravery displayed in the recent
battles with appreciation.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

76 T h e A n v i l 1915–16

This was Falkenhayn writing, not the Kaiser. Rupprecht’s reply was firm. He
was keen to reject any implication that the effort so far had been less than
wholehearted:

His Majesty the Kaiser may rest assured that Sixth Army will do everything in
its power to stop the enemy offensive. This commitment has been maintained
from the beginning. We will not voluntarily relinquish any ground to the
enemy. At this time a major counterattack does not hold any promise, but we
continue to reserve this option.19

Falkenhayn, however, remained worried that Sixth Army’s grip on the

situation was not all it should be. In the evening of 11 May two OHL
interventions in Sixth Army affairs underlined that fact, to Rupprecht’s
annoyance. First, a liaison officer arrived to provide a direct link between
OHL and Rupprecht’s two most menaced corps, bypassing Sixth Army
headquarters, and to advise on the feasibility of a counter-attack. Rupprecht’s
complaints that this threatened the integrity of the chain of command were
ignored. Second, a telegram arrived from Falkenhayn announcing that a
fresh corps was coming to reinforce Rupprecht, which should be used
‘under an energetic commander’ to regain the lost ground. Rupprecht
correctly read this as implied criticism of the general officer commanding
I Bavarian Reserve Corps, Karl von Fasbender. To Rupprecht’s mind, it was
not Fasbender, but Falkenhayn, who was most responsible for the setback of
9 May. The latter had ignored warnings that the units in this sector were
being stretched too thin. On 12 May, Falkenhayn was more explicit, wiring
that any counter-attack needed a leader who ‘is highly committed to the
thing and is confident it will succeed’. He named a general (Claer) whom
he knew full well Rupprecht thought unsuitable for special missions.
This was the last straw for the Crown Prince. He was no longer prepared
to tolerate his authority being undermined. He would complain direct to
the Kaiser. Before he could do so, however, he received further fuel for
his resentment, both on his own account and on behalf of Bavaria. First,
Falkenhayn appointed General Ewald von Lochow to command a semi­-
independent task force comprising three of Rupprecht’s corps. Lochow
was being squeezed into the chain of command rather as Fabeck had been
at Ypres, nominally under Rupprecht but in reality reporting direct to
OHL. Thus Rupprecht suddenly lost command of half his army. All his
Bavarian staff thought he should resign in protest, seeing this as a ‘war’
with OHL. The next day, Falkenhayn twisted the knife further. When
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng77

Rupprecht requested reinforcements, OHL told him to consult Lochow—

his subordinate—before deciding whether he really needed more men.
Second, Rupprecht heard talk of a new military convention which
would give the Kaiser power to appoint and promote Bavarian officers: an
infringement of Bavarian royal privileges which Rupprecht neither saw the
need for nor welcomed.20
Rupprecht wrote his letter to the Kaiser and sent it off on 17 May. He
received two very different responses. The formal answer arrived on 23 May.
It comprised two letters. One was from the Kaiser, agreeing with Rupprecht
on all points. The other, from Falkenhayn, apologized for any misunder-
standings. Rupprecht replied that he considered the matter now closed.
There had already been a less formal, but more important, response, how-
ever. At one o’clock in the morning of 19 May, Rupprecht’s chief of staff,
Krafft, was shaken awake and handed an urgent message. He was ordered to
depart the very same day for the Tirol, there to assume command of the
Alpenkorps, a formation of mountain troops collected to fight Italy. ‘That’,
thought Krafft, ‘is Falkenhayn’s answer to the Crown Prince!’ Rupprecht felt
that ‘Krafft’s transfer was a heavy blow for me: I’d known him well since the
War Academy and I lost in him not only a trusted adviser but a loyal friend.’
He also worried that it would create the impression that the command team
of Sixth Army had failed and that Krafft’s transfer marked another step
in  Prussian attempts to dominate the Bavarians. Krafft was replaced by a
Prussian, Colonel Gustav von Lambsdorff. No effort was made to consult
King Ludwig III about Krafft’s transfer.21
While Rupprecht and Falkenhayn played out their political battle, a more
deadly one continued to unfold in Artois. The British, under pressure from
Joffre to add their weight to the offensive, began once more to batter away
near Neuve Chapelle. This time, having seen a short bombardment fail, they
experimented with a prolonged one. During three days of deliberate,
observed fire, British guns fired over 100,000 shells, obliterating much of the
German front-line breastwork on a 5,000-metre front north of Festubert.
On the night of 15/16 May the BEF began its first night assault of the war.
Two of the attacking brigades were detected by alert defenders; they were
shot up and failed to get near the German lines, leading to the cancellation
of planned follow-up attacks in this area. The third brigade was more
successful, silently crossing No Man’s Land and seizing part of the German
front trenches. At daybreak, Major-General Hubert Gough’s 7th Division
joined the assault. It also took the German front and support lines, but by
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

78 T h e A n v i l 1915–16

ten a.m. the attack had stalled. Resistance and rain prevented further ­progress
that day but during the evening the Germans withdrew to a new line hastily
built to the rear. Although fighting continued until 25 May, the British were
unable seriously to threaten this defensive position. The BEF had advanced
up to 1,500 metres on a front 5 kilometres wide but lost some 16,000
­soldiers, while the Germans probably lost fewer than the 5,000 suggested by
the British official history. German VII Corps drew down one brigade of
reinforcements but was otherwise able to handle the situation with its own
resources. Although Rupprecht did mention these operations in his diary,
they clearly worried him little. He did not even bother to refer to another
British attack on 16 June at Givenchy-lès-la-Bassée. This proved a ‘rerun of
Aubers Ridge’, as Prior and Wilson put it: uncut wire and intact defences
cost the BEF another 3,500 casualties for no gain.22
The situation further south, around Notre Dame de Lorette and Vimy
Ridge remained Rupprecht’s major concern. Neither army was seeing much
return on the blood spilled in poorly coordinated attacks, hastily cobbled
together by tired troops and staffs, so both sides rethought their approach.
The German response, as we have seen, was to bring in Lochow to try to
coordinate efforts in the central sector. The French took a different path.
Foch told d’Urbal that improvised attacks must cease at once. Instead, a slow
and methodical approach was called for, with assaults launched only after
the ‘fullest and most detailed preparation, along the lines of the 9 May attack.
This preparation requires time.’ In the short term, Tenth Army was to focus
on local attacks to improve its positions and secure jumping-off positions
for a full-scale attack to take place later. French pressure was sufficient to
keep the initiative and prevent a coordinated counter-attack. Any fresh
German troops found themselves being fed in to hold the line almost at
once. By the end of May, despite heavy rain and much mud, the French had
cleared the Lorette Spur and captured Ablain-Saint-Nazaire. German troops
still held out, however, in Souchez, Neuville-Saint-Vaast, and the Labyrinth.23
By early June, OHL seemed to feel the pressure was easing somewhat.
Falkenhayn sent two and a half divisions to the Eastern Front to reinforce
the remarkable successes achieved in the Battle of Gorlice-Tarnow. He even
travelled there himself for a few days to plan further exploitation. Rupprecht
was not quite so sanguine. Between 9 May and 5 June, he noted, his troops
manning the Souchez-Neuville sector had been subjected to fifty enemy
attacks. On 7 June alone, no fewer than six French attacks were broken up
by German artillery fire before they could gain any ground. Nonetheless,
the strain on German units remained intense, and worn-out units needed to
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng79

be rotated out of the line for rest. Rupprecht’s dissatisfaction with Lochow
grew: he found him lacking in dynamism and over-cautious in his prepar-
ations. On 15 June, in view of the apparent improvement in the situation,
he suggested that Lochow’s task force should be dissolved. Lochow, how-
ever, successfully argued for a fortnight’s extension.24
In fact, at exactly this time, after five weeks of fighting and preparation,
French Tenth Army was finally ready for a new attempt to capture Vimy
Ridge and break through on to the Douai Plain. D’Urbal launched his
assault on Wednesday 16 June. He tried to rectify what he saw as the three
greatest shortcomings on 9 May. First, the Entente scaled up operations
generally in an effort to tie down German reserves all along the line. The
BEF launched operations at Givenchy on the La Bassée Canal and at
Bellewaarde near Ypres. The former was a bloody failure, as discussed above.
The latter seized a few hundred metres of front-line trench at a cost of
nearly 4,000 men. More significantly, three French armies launched their
own attacks to support the Tenth Army effort. Second, although Joffre
could do little to reinforce d’Urbal’s artillery, he could ensure better supplies
of munitions. So, although the number of guns grew little, nearly twice as
many shells fell on the German trenches in the week before 16 June than in
the equivalent period in May. Third, this time the reserves would not be left
too far back behind the line. Indeed, the attacking units were to commit
every man they had to the first assault and keep no local reserve. Army
reserves would follow up the advance and mop up as required.25
Significant obstacles to French success remained, though. The German
defenders were fully alert. French troops were tired and intelligence was
poor. At least one French division had been in the front line every single
day since 9 May. Commanders did not always know the precise positions of
their own men, much less where their opponents were. The flanks were still
not properly secure. So any French advance would be exposed to German
enfilade fire from Souchez and Neuville-Saint-Vaast, unless these villages
could be rapidly neutralized when the assault went in. This proved impos-
sible. As a result, although the Moroccan Division again did well, driving
into the second line on Hill 119,26 and a few other footholds were gained in
the defences, yet again the attackers were exposed to cross-fire and the
attack stalled. In many places, indeed, intact wire and heavy losses had
stopped it almost before it had started.
The Germans were once more worried as ‘the situation was becoming
extremely precarious’. Rupprecht was gripped by insomnia. Lochow argued
that Sixth Army should pull back behind Vimy Ridge, lure the French
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

80 T h e A n v i l 1915–16

forwards, and then hit them from three sides as he had successfully at
Soissons in January. Rupprecht, on the other hand, felt that his men were
quite capable of holding in their current positions. In any case, he was not
sure he had enough strength to execute a plan on Lochow’s scale. Rupprecht’s
view prevailed. OHL did, however, think it worth sending significant
reserves, totalling some five divisions and thirteen heavy batteries, over the
next few days. In general, by the end of 18 June it was becoming clear to the
Germans that the French drive was losing punch. Indeed, French losses
between 16 and 18 June were 29,000, only slightly fewer than in the whole
month of May. Foch and d’Urbal suspended major operations on 18 June,
although heavy fighting went on locally until almost the end of the month.
By 22 June German intelligence had convinced the high command that the
Second Battle of Artois was over.27
Over 200,000 men became casualties during the Second Battle of Artois,
half of them French. The Germans had reason to be pleased with their
efforts. Although some ground had been lost, they had stymied Entente
breakthrough attempts. Any lessons the British and French had learnt had
been at a very high price, and both had shown themselves capable not only
of inventing new mistakes but also of repeating old ones. They would need
months to recover and get ready to try again.28
By the end of June the French had begun to move beyond the denial of
early 1915 and grasp the scale of the challenge they faced. In February, Foch
had still been maintaining that, if the defensive seemed superior to the
offensive during the winter fighting in Artois and Champagne, it must be
because the attackers had been doing something wrong. Similarly, Grand
Quartier Général (GQG) still felt nothing had fundamentally changed:
The present war has in no way undermined the principles which form the
foundations of our offensive doctrine, but, because of the novel forms it has
taken, operations are characterized by unfolding more slowly and methodic-
ally in both time and space. That is the result of firepower and the strength of
defensive organization. The principles remain intact. Methods of attack must
be adapted to this new situation.

Specifically, preparation should be thorough and methodical. With suffi-

cient artillery support, getting infantry into the enemy defences was pos-
sible, so long as assaults were not made piecemeal but properly coordinated
across a wide front. Success was to be followed up fast.29
Joffre developed the theme of speed further in his Note 5779 of 16 April,
written to set the tone for the Second Battle of Artois: ‘the aim of an
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

A Succe ssf u l Spr i ng81

­ ffensive action is not merely to capture a line of enemy trenches but indeed
o
to chase the enemy from the whole of his position and to defeat him without
giving him the time to re-establish himself ’. The initial assault must be
­‘sudden, violent and pursued ceaselessly and continually by the commit-
ment of fresh units to the fighting front until the final result has been
achieved’. In other words, Joffre was hoping to break through the German
lines with one sharp blow.30
On 9 May, breaking in to the German positions indeed proved relatively
straightforward but the results proved disappointing, as we saw above.
The first lessons-learnt reports, produced within a couple of weeks, sug-
gested moving reserves up nearer the front and some alterations to artil-
lery-targeting priorities, but broadly upheld Joffre’s operational approach.
Even after another major failure on 16 June, GQG reckoned that
the lessons of the most recent fighting prove that the first infantry attack,
minutely prepared and vigorously executed, always succeeds. The enemy front
is broken and his artillery unable to react. If our infantry stop, it is less because
its offensive force is spent than the exploitation of initial success cannot be
pushed far and fast enough.

As Joffre told Foch, more artillery was needed and any attack must take
place on a broader front. More effort must be made, especially by the British,
to tie down enemy reserves. Foch’s reply was guarded about the possibility
of a breakthrough. The German superiority in heavy artillery worried him
and to rely on the British would be an error. Foch hinted that the poor
results of 16 June, despite reinforced artillery and the improvements made
after 9 May, in fact suggested that the problem might go deeper than
poor execution or the application of insufficient resources. It was possible
instead that the whole approach might be mistaken. The French might
need to recalibrate their entire thinking. Not everyone yet agreed with
Foch about the need for wholesale change. Nor had he worked out what
reforms were necessary. But the fact that some, at least, within the French
army were beginning to think in ‘root and branch’ terms represented
­significant progress.31
The British remained a long way behind. With a smaller army in 1914,
they had been able to call on a narrower base of military expertise in the
first place. The impact of the heavy casualties of the war’s first months was
therefore especially intense and it proved difficult to bring the rapidly
expanding army up to par. Limited resources in manpower and artillery
left BEF attacks necessarily only small-scale in 1915. While this reduced
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

82 T h e A n v i l 1915–16

the absolute level of loss, if not the proportional rate, when they had the
opportunity to experience large-scale operations, learning was sometimes
erratic. The BEF misread the lessons of Neuve Chapelle at the tactical level,
for example, and persisted in using short bombardments long after the
French had given them up in favour of thorough preparation. When the
guns failed either to break down the German defences or to achieve surprise,
the consequence was the bloody failures of spring 1915. Operationally,
British staffs and commanders had little chance to confront the issues of
momentum and tempo in major battles which Foch was beginning to
ponder. The political significance of the British army’s effort in France as
yet outweighed its military impact.32
To the German defenders, wholesale reform did not seem necessary. The
heavy artillery had worked well. So had the newly rebuilt railways, allowing
quick and efficient reinforcement of the defenders. Most of the lessons
Sixth Army learnt were low-level tactical tweaks. For instance, units were
reminded to counter-attack from the flank where possible, rather than
frontally. What the Second Battle of Artois did close down, however, was
any chance of continued debate about the utility of a second line of defence.
A full second line was now clearly necessary. Falkenhayn ordered one to be
built all along the front, about 1,500 metres back, to prevent the enemy
overrunning both in a single rush. Both ‘lines’ in fact comprised linked
systems of several trenches, a hundred metres or so apart, with sufficient
dugouts to shelter the garrison. Specialist engineers and artillerymen were
made responsible for inspecting the front and disseminating best practice in
their spheres. The next time the British and French attacked, they would
find themselves facing even stronger defences.33
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

8
Further Victories

T he summer of 1915 passed relatively quietly for Rupprecht and his

Sixth Army, although the steady drip of casualties continued relent-
lessly. Sharp fighting broke out from time to time, especially in dangerous
flashpoints such as Souchez. Overall, however, the first anniversary of the
outbreak of war allowed time for reflection. For Rupprecht, despite the
defensive successes achieved, it had been a year of disappointment: ‘How
brilliantly the war began! What hopes we possessed a year ago! How dif-
ferently everything has turned out.’ To mark the anniversary of the Battle of
Lorraine, the Kaiser honoured Rupprecht with the Pour le Mérite, the order
founded by Frederick the Great informally known as the Blue Max.
On 29 June 1915 Rupprecht gave an interview to the New York Times,
presumably to repair any damage done to his reputation by allegations
publicized in May that his men had murdered prisoners of war.1 If so, he did
a good job. He described himself as ‘the anvil in the west’, whose job it was
to hold the enemy back while the hammer struck in the east. He appeared
‘simple, unpretentious and democratic’. He was ‘not a voluble talker, par-
ticularly about himself, and [was] modest to a fault’ but ‘one quickly discovers
that he has personality with a punch and an iron will. As a man and a fighter
with something of the shrewd canniness of the Scotchman and all of the
bulldog tenacity which is traditional of the British, relieved by native
Bavarian good humour’, he was ‘soft of voice and slow of speech’. He took
the opportunity to deny ordering the execution of captured British troops
and to complain that American artillery shells, sold to the French army,
were prolonging the war. Over a ‘very small glass’ of beer he showed his
human side, chatting about hunting and expressing a wish to return to
America after the war.2
The state of reserves would not allow the Germans to attack in force
before November at the earliest. Oberste Heeresleitung (OHL) had just six
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

84 T h e A n v i l 1915–16

Figure 8.1.  Kaiser Wilhelm II visiting Rupprecht’s headquarters

divisions available. Intelligence suggested that the French, on the other

hand, were actively preparing an offensive, perhaps in Alsace, with a possible
diversion around Arras.3 From late August into September, enemy activity
certainly seemed to be on the rise opposite Sixth Army. French artillery fire
was growing heavier and changing from routine harassment to something
more deliberate. Enemy infantry were digging new assault trenches and
pushing saps out into No Man’s Land towards the German lines, in some
places closing to within hand-grenade range. Prisoners of war began to
report the arrival of extra heavy artillery. The British seemed active around
Hulluch and the threat of a French offensive down in Champagne was also
growing.4
What the Germans were observing were the preliminaries of a large-
scale Anglo-French offensive which Joffre had been planning since July.
The precise shape and timing changed over the summer as the French and
British commands debated among themselves. The final idea, though, was
for Castelnau’s Groupe d’armées du Centre (GAC) to deliver the main
blow in Champagne on Saturday 25 September with some thirty-five
divisions, while French Tenth Army attacked on a 20-kilometre front in
Artois with seventeen. On d’Urbal’s left, Sir Douglas Haig’s First Army
would extend the offensive a further 10 kilometres north, driving on Loos
and Hulluch with six divisions in the first line and a further three in reserve.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

F u rt h e r V ictor i e s85

Haig possessed an average of eleven guns per kilometre, down from about
thirty at Neuve Chapelle and compared with twenty-two for the French in
Artois and thirty in Champagne. Haig hoped to offset his relative weakness
in artillery with chlorine gas, which would panic the defenders and allow the
British to carry both the first and second German positions in one bound.5
By 21 September, Sixth Army was braced for an imminent assault.
Oberste Heeresleitung (OHL) sent Rupprecht additional artillery ammuni-
tion and guns. Enemy preparations were most obvious from Souchez down
to the River Scarpe. Further north, in the sector of General Friedrich Sixt
von Armin’s IV Corps around Hulluch and Loos, the opposing lines
remained far apart, there was little British sapping going on, and an attack
seemed less likely. On Friday 24 September, however, the bombardment
reached a new pitch of intensity along the whole front from the La Bassée
canal south to Arras.6
The next day it took about three hours for the first news of major British
attacks between Hulluch and Loos to reach Sixth Army headquarters.
Rupprecht had not expected an assault in this sector, where mining villages
provided good defensive positions and No Man’s Land was open, overlooked
by German-held slag heaps from the mining industry, and up to 800 metres
wide. Thick clouds of gas covered the attack, however, cutting visibility to
3 metres in places. The British quickly overran IV Corps’ front lines, des-
troying whole companies of the defenders and capturing their guns. No-one
knew what forces, if any, were left to man the second line and fill the gap.
The situation seemed critical. Rupprecht put all his reserves at Sixt’s disposal
and appealed for more from OHL. Falkenhayn agreed to help at once. By
lunchtime it was clear that the British had captured Loos and were threat-
ening the German second line along the Lens-La Bassée road, with fighting
underway on Hill 70.7
With the situation around Loos obscure but troubling, further worrying
reports soon began to reach Rupprecht’s headquarters. This time, it was the
French on the move. After a morning of carefully observed artillery fire
their Tenth Army attacked at half past one in the afternoon. Seventeen
divisions went over the top from Angres to Ficheux. South and east of
Arras, Fasbender’s Bavarians cut them down in No Man’s Land and stopped
the attack dead. From Roclincourt up to Souchez, however, the situation
seemed more threatening. Again, as in May, groups of French troops broke
through and even reached the crest of Vimy Ridge in places. They were
isolated and wiped out in the course of the afternoon and evening, however.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

86 T h e A n v i l 1915–16

To the north of Vimy Ridge, the French overran the German trenches,
swept through Souchez, and drove up on to the lower slopes of the Gießler
Heights. Here the advance was brought to a halt by the second-line defences
and a round of local counter-attacks.
The situation around Loos was still Rupprecht’s biggest worry. Falkenhayn,
however, had larger concerns yet. Because of the pressure in Champagne, he
sent reserves, originally intended for Rupprecht, down there. Sixth Army
would have to hold out until the Guard Corps could arrive in forty-eight
hours’ time.
The question for all three armies was where to commit their reserves the
next day. In Artois, Foch ordered his right wing to hold fast and reinforce
the relatively successful left. Rupprecht overruled a complicated idea of
Lambsdorff ’s and ordered the Guards to push straight into the most threat-
ened sector, around Souchez. Meanwhile, the British thoroughly misjudged
the situation. Sir Douglas Haig, convinced that he could have broken
through the German second line around Hill 70 on 25 September if only
reserves had been to hand, sent forward two inexperienced New Army
divisions to attack that second line between Hulluch and Hill 70 in the late
morning of 26 September.
The German position the Kitchener volunteers faced consisted of no
more than a single trench, but it was intact and lay on a reverse slope, making
it hard to spot or shell. They would have to cross more than 1,500 metres of
open ground to reach it. Worse, both Hulluch and Hill 70 remained in
German hands, exposing the British divisions to crossfire. Haig gambled on
clearing both flanks as the main attack went in. After a very thin prelimin-
ary bombardment lasting just an hour, the British 21st and 24th divisions
began their attack. Both had been in France less than a month. This was
their first action. With staffs green and improvised, the troops underwent
some chaotic night marches and were sent into battle worn out and hungry.
The effort of the 21st Division in the south was also disrupted by a pre-
emptive German attack. When the two British divisions advanced at noon,
they were met by fierce and persistent machine-gun and artillery fire. They
suffered heavy casualties and soon lost both direction and cohesion. A few
groups of brave men reached the German wire but found it was uncut and
could not get through. Within minutes the entire attack collapsed and the
British broke. One German witness described the British advancing naively,
as if on exercise, in dense columns twenty to thirty deep with officers
leading the way. Field guns and cavalry trotted behind each column. Close
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

F u rt h e r V ictor i e s87

formations made easy targets. Soon some 10,000 British lay dead and
wounded in front of the German 117th Infantry Division. The rest ran away.
Although an English prisoner of war (POW) said ‘it was a very nice sport’,
Rupprecht noted that ‘we Germans have a somewhat more austere impres-
sion of war’. That evening both British divisions were taken out of the line,
each having lost some 4,000 men. The original assault waves of 25 September
were also suffering badly by now. One Scots brigade, for instance, had
suffered 65 per cent casualties. Any hopes Haig had cherished of breaking
through now evaporated.8 Further south, the French consolidated their
hold on Souchez but were unable to make much progress against Vimy
Ridge itself. Rupprecht was by now convinced that Sixth Army could
contain the enemy. A day of small-scale attacks by all three armies on
27 September seemed to bear this out, with the Germans getting the better
of the fighting, especially against the British.9
Events on 28 September shocked the German command in Artois out of
its complacency. Joffre had ordered Foch that Tenth Army should, without
letting the British Expeditionary Force (BEF) know what it was up to, scale
down its offensive. This was designed to free up troops to transfer to
Champagne, where the offensive was going relatively well. Nevertheless,
local operations had been continuing around Angres, Souchez, and on the
Gießler Heights. Slightly further south French troops continued probing up
the western slopes of Vimy Ridge, around where the Canadian Visitor
Centre now stands. Suddenly, during the early morning of 28 September,
German resistance opposite them crumbled. Two French divisions made
good progress towards the highest point of the ridge, the summit of Hill
140. General d’Urbal sensed an opportunity and ordered fresh attacks by
three whole corps from Angres down along Vimy Ridge. Results, however,
turned out to be limited. The French captured a few trenches on the Gießler
Heights. Again, a few groups reached the crest of Vimy Ridge around Hill
140 and La Folie, but were cut off and crushed. Reinforcements could not
have reached them even had any been available. Foch was normally an
incorrigible optimist but even he did not think exploitation was a realistic
prospect. The troops were too exhausted, the ground too chewed up, and
German reinforcements would soon seal off the front once more.10
This attack caused deep and genuine concern in the German lines,
however. When General von Pritzelwitz, in command of the Vimy Ridge
sector, lost his nerve and demanded to be allowed to retreat, he was dismissed
and replaced. Rupprecht’s chief of staff, Lambsdorff, also dithered. His advice
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

88 T h e A n v i l 1915–16

was, first, to abandon the ridge and pull back into a third line down on the
plain; second, to attack; and third, to let the Guard Corps commander do
whatever he thought best. ‘The chief of staff is completely incompetent’,
noted one of his staff officers. Even Rupprecht, receiving reports of French
troops in Givenchy, La Folie, and on Hill 140 and fearing a breakthrough,
thought that ‘this evening was the most frightening of the campaign so far’.
The only reserves he had immediately to hand that day were two com-
panies in training. He later considered that the French had indeed broken
through, if only they had realized it.11
Rupprecht’s response was, first, to send the Guard Corps in around
Givenchy and Hill 140. Second, he inserted I Bavarian Corps into the line
on the southern half of Vimy Ridge. Another corps was held in reserve in
Douai. So, while French Tenth Army had initially been less successful than
the British in pulling in German reserves, now they were beginning to have
a significant impact. This was especially important since it was not only in
Artois that the French were able to penetrate the German second position
on 28 September: their offensive in Champagne similarly made good
progress on that day.
By now, however, the French units around Arras were exhausted. They
could make no headway in heavy fighting all along Vimy Ridge over the
next week or so. Ammunition was scarce. Muddy ground, churned up by
constant artillery bombardment, made movement difficult. There were no
fresh French reinforcements to match those Rupprecht was able to throw in.
Although the French occasionally made small-scale break-ins over the days
that followed, by 29 September it was clear to Rupprecht that he had
contained their primary drive. The next day he noted that ‘the main crisis
seemed over’.12
Plans for a renewed Anglo-French offensive on 2 October had to be
postponed several times, partly due to weather, partly to allow time to secure
vital tactical features before the main assault. A German counter-attack with
five regiments at Loos on 8 October suffered heavy casualties and gained
little ground. It did, however, disrupt British preparations and force a further
postponement. Eventually, the French assault began in the afternoon of 11
October but progress was limited and by 14 October the Germans had
regained any ground lost. Even to Rupprecht, it was clear that Anglo-French
coordination had broken down: ‘it was striking that while the French were
making such efforts, the British were so passive’.13 Delays and disappoint-
ments on both sides had eroded mutual trust between French and British
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

F u rt h e r V ictor i e s89

commanders. Also, however, the inexperienced men and staffs of the BEF
were having trouble keeping up with the intense tempo of local fighting.14
There was a third factor at work, too. British senior commanders were
distracted by a row which flared up between Haig and French over the
opening of the Battle of Loos. The Prime Minister, Herbert Asquith, wanted
a scapegoat for the military failures of 1915 and Haig was determined it
should not be him. He wrote to Kitchener claiming that his men had
broken through on the first day, but that he had been unable to exploit their
success because the commander-in-chief was holding reserves too far
back and released them too late. Intrigues about the top job occupied the
higher reaches of the BEF until they were finally resolved in Haig’s favour
in December.
When at last three British divisions attacked on 13 October, the result was
another bloody disaster. In the unusually blunt words of the British official
historian, it ‘brought nothing but useless slaughter of infantry’. Although
attempts had been made to learn from recent mistakes at the tactical level,
operational planning remained muddled and over-ambitious, intelligence
poor, and execution slipshod. Nick Lloyd suggests that ‘in many ways
the attack . . . typified much that was wrong with the BEF in this period’. The
failure of this attack effectively closed down large-scale operations for the
British at Loos, with Joffre officially suspending the French offensive on
15 October.15
The French deployed half as many divisions during the third battle of
Artois as they did in Champagne. Artois was not solely a diversion from the
main effort, but it was firmly a sideshow in comparison. The BEF commit-
ted twelve divisions to a sideshow of the sideshow. Casualties were, as always,
heavy: about 150,000 men between the three armies. The Entente lost two
men for every German but again had little to show for the price paid. The
French and British lines had pushed a couple of kilometres east but had not
won any truly significant ground. The German lines had buckled, but not
broken. Nor did operations in the north significantly aid success in
Champagne. Not even the Entente could afford a strategy of attrition with
this loss ratio. Both British and French high commands were duly disap-
pointed and the Battle of Loos offered the final pretext required to remove
Sir John French. The German official history described the third battle of
Artois as a ‘clear—albeit costly—victory’.16
After the failures of autumn 1915, the French became more open to radical
changes in approach. Foch was more convinced than ever that German
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

90 T h e A n v i l 1915–16

defences were now too strong to break through with one ‘coup de main’.
Instead, a methodical approach, characterized by careful preparation, plan-
ning for multiple operational phases and close support from heavy artillery,
was required. His opinion now won converts up to and including Grand
Quartier Général (GQG). It was enshrined in a new series of doctrinal
instructions issued in January 1916: ‘In the current situation, specifically that
of attacks on fortified fronts which present successive lines of defence, a
breakthrough of the enemy front will not in general be possible in one
operation alone.’ These laid down the new offensive doctrine. New defensive
instructions, released in December 1915, paid the Germans the compliment
of imitation, by ordering units to establish a defence in depth.17
The Germans apparently had less to change. Autumn 1915 demonstrated
how well they had learned the lessons of trench warfare over the last year.
Their defensive tactics, it seemed, had been triumphantly vindicated.
Nonetheless, Sixth Army carried out a thorough analysis of the fighting. It
called in after-action reports and distributed a follow-up questionnaire. The
final product was a printed twenty-four-page booklet, including maps. This
distinguished between the enemy’s initial tactical success and his eventual
operational failure. The weakness of the defensive positions around Souchez
was blamed for what progress the French made there, while the initial
British break-in was put down to weight of numbers and surprise. It con-
firmed, however, that the basic principles of defensive tactics, such as the use
of reverse slopes, had worked well. At the operational level, although in
places enemy artillery had inflicted heavy losses on over-strong front-line
garrisons, overall the battle seemed to validate defence in depth. Strongpoints
scattered between the first and second positions had helped to break up
enemy attacks, although poor enemy communications and control had also
helped. Immediate counter-attacks on disordered and disorientated attack-
ers by German reserve units had proved highly effective. Germany, it seemed,
with a little help from its enemies, had learnt how to win defensive battles
in trench warfare. The closest the report came to recommending a major
change was to suggest that trenches should be dug more than 2 metres wide,
to prevent attackers from jumping over them.18
At Rupprecht’s headquarters, however, there was not much back-slapping
going on. It was clear that a way must be found to turn tactical victory of
this kind, a negative success which prevented the Entente from achieving its
aspirations, into a more positive step which would bring German victory
closer. Wars are not won by sitting forever on the defensive. To launch a
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

F u rt h e r V ictor i e s91

decisive offensive in the west, Rupprecht thought, required reinforcement

by twelve to fifteen army corps which simply were not available. To avoid
a very long war, he told Falkenhayn on 13 October, the aim should become
the best peace possible, rather than outright victory.19
The other factor contributing to an uncomfortable atmosphere in Sixth
Army headquarters was growing discord within the staff. The focus of irri-
tation was the new chief of staff, Lambsdorff. Anyone trying to fill Krafft’s
place was always going to find his relationship with Rupprecht difficult to
manage, especially given the circumstances of Krafft’s removal. Over the
summer, however, Lambsdorff had become increasingly alienated not only
from his army commander but also from the rest of the staff. When he
dithered under strain in September, then failed to brief Rupprecht that the
crucial Gießler Heights had been lost, and finally tried to cover his mistake
up with a lie, it was time for him to go. His replacement was the highly
experienced Prussian General, Hermann von Kuhl. Kuhl had served as
chief of staff to Kluck’s First Army in the early months of the war, and later
with Fabeck’s Twelfth Army in the east. Major Xylander of Rupprecht’s
staff was, as usual, dismissive: ‘that’s great! Things go from bad to worse!
“Papa Kuhl” has become Chief of Staff, that common, lying, little tactical
dead loss of a man. Nothing ever gets better.’ In fact, however, when
Xylander saw Kuhl at work, even he was impressed. Rupprecht and Kuhl
would go on to form a highly effective partnership which lasted for the rest
of the war.20
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

9
Verdun and the Road
to the Somme

T hroughout 1915, Falkenhayn’s main focus had been on the east, where
the fighting had carried the Central Powers deep into the Tsarist
Empire. Russia had lost two and a half million men but still her armies had
not collapsed. Displaying extraordinary resilience, they were still in the field
and fighting on. The German and Austro-Hungarian troops, on the other
hand, were exhausted and had done all they could for now. There was little
hope of them achieving in 1916 what they had not managed in 1915. If the
war were to be won, it would have to be won in the west. And wars are not
won by defensives. Somewhere, Germany would have to attack. The place
she chose was Verdun.1
The Battle of Verdun is often held up as one of the archetypal attritional
battles of the First World War. The briefest of glances into the ossuary at
Douaumont tells you why. Although Falkenhayn pretended that his inten-
tion had always been to ‘bleed the French army white’, that was not his
original plan, as his biographer Holger Afflerbach has shown. Kuhl and others
criticized Falkenhayn for not knowing what he wanted and, Micawber-like,
hoping something would turn up. In fact, though, Falkenhayn’s plans were
if anything too complex, rejecting the neat but artificial dichotomy between
attrition and manoeuvre warfare and instead trying to synthesize the two.
In its final form, his plan consisted of two, or possibly three, phases. First,
he would threaten or seize something the French could not afford to lose.
The target he chose was Verdun, partly because it had a reputation as a
strong fortress and partly because it sat in a salient. His artillery could fire
into it from two sides. This would help during the second phase, when he
expected the French army to counter-attack to try to regain ground they
had lost. The German heavy guns would chew up these French attacks until
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

V e r du n a n d t h e Roa d to t h e Som m e 93

morale collapsed. Falkenhayn hoped a third phase would ensue with a

British attack around Arras to relieve pressure on their ally. He expected to
have little trouble defeating this, but if necessary he would then launch a
final attack near Arras to complete the overthrow of British and French
armies. Within a few months, therefore, if all went according to plan, at a
minimum Germany could hope to have broken the back of the Entente in
the west, enabling Falkenhayn to send troops to finish off Russia. The key
to the plan lay in the estimate of enemy morale. If Falkenhayn was right that
the French army was close to collapse, Verdun might yield rapid strategic
success. If he was wrong, however, Germany would be locked in to a slow
grind with little hope of winning. Unfortunately for Falkenhayn and many
of his men, his assumptions proved flawed.2
Rupprecht’s part in the planning of Verdun was limited. By now, he and
Falkenhayn had been sniping at each other for a year. Their relationship had
deteriorated further in the wake of the third battle of Artois, when
Rupprecht accused Oberste Heeresleitung (OHL) of losing the Lorette
Spur, by stripping his army of reserves against his advice, in front of the
Kaiser. It is hardly surprising, therefore, that Rupprecht only heard of
Falkenhayn’s plans for an attack on the French late, at Christmas 1915, or
that he disagreed with them. As he had since the spring, Rupprecht argued
against dissipating strength in local offensives against the enemy’s strongest
points. Instead, he advocated concentrating all possible resources for a
knock-out blow, perhaps between Arras and the Somme, in the summer. It
was equally predictable that Falkenhayn would ignore Rupprecht. Within
days, Sixth Army was ordered to transfer all its supply lorries to Crown
Prince Wilhelm and to prepare diversionary attacks for the end of January.3
Further meetings and memoranda in early 1916 served only to emphasize
the gap in thinking between Rupprecht and Falkenhayn. The latter argued
not only that large-scale reserves did not exist, but also that the previous
year’s fighting had demonstrated that it was impossible to maintain control
of large forces on the modern battlefield. On 12 February, Kuhl came away
from a meeting with Falkenhayn with the impression
that Falkenhayn himself was not clear what he actually wanted and was waiting
for something fortunate to turn up to produce a good outcome. He wanted a
decision in the spring, but considered a breakthrough impossible and hoped
that with the operation at Verdun he could shift over from position to mobile
warfare. How else, I wonder, can that take place but with a breakthrough? Just
because the French and English leadership of mass armies failed, must ours, too?
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

94 T h e A n v i l 1915–16

Falkenhayn was talking of fighting a decisive battle at Verdun, but for fear
of a British counter-attack, which neither Rupprecht nor Kuhl thought
likely, was not prepared to concentrate his forces to fight it. ‘He who bets
nothing, wins nothing’, sniffed Rupprecht.4
On 26 February, Kuhl submitted a memorandum outlining his ambitious
scheme for a decisive breakthrough. He and his staff had been working on
this for weeks. It reiterated that ambitious objectives required sizeable
commitment and underlined their opposition to smaller-scale operations.
The memo remains interesting both as a summary of German thinking
about the attack in early 1916 and because it displays considerable foresight
about events later that year. Kuhl argued that the British at Loos and the
French in Champagne had done their best. They had prepared carefully and
brought maximum strength to bear. Both had come close to a breakthrough
but ultimately failed. The German second-line defences had held them up.
Entente attempts to break through that second line, whether improvised or
set-piece with full artillery support, had failed because the defenders had
been reinforced and surprise had been lost. Kuhl disagreed with the view
that breakthroughs were therefore impossible and attackers should adopt a
more methodical approach. He pointed out that proceeding step by step
would only give the defender time to strengthen his defences and build
whole new lines further back if necessary. Consequently, the Entente was
right to seek a breakthrough. This meant punching through the enemy first
and second lines and overrunning his gun positions, all in one bound, which
would require several things. First, surprise was essential. Second, the attacker
would need to work on a wide frontage: at least 20 kilometres. Third, he
would require between twenty and twenty-four divisions and 320 heavy
batteries plus field artillery. This represented roughly three times as many
heavy guns as the Entente had just used in Artois and nearly a quarter of all
the heavy artillery the German army possessed in 1916. Indeed, even the
densest concentration achieved on 21 March 1918 was less intense. Kuhl was
setting the bar too high and neither this memo nor its reasoning had much
impact at OHL.5
On 23 January, Rupprecht’s Bavarian troops along Vimy Ridge launched
a series of local attacks, setting off mines, seizing the craters, and driving the
French well back from the crest. Successful as they were, these actions indi-
cate how dysfunctional German command had become. No-one had
informed Rupprecht that the Verdun operation had been postponed so
these diversions were wasted. Rupprecht was ‘shocked at this irresponsible
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

V e r du n a n d t h e Roa d to t h e Som m e 95

carelessness!’ No less sloppy was the fact that, far from being a closely
guarded secret, an attack on Verdun was the talk of Berlin. Right away,
British intelligence had spotted that the German attacks north of Arras
were ‘diversions’. By the time the German attack at Verdun finally went in
on 21 February, the French had done their best to reinforce the salient.6
Within a week at Verdun, it was clear to Rupprecht that forward progress
was slowing. Although Falkenhayn remained convinced that a British attack
was imminent, analysis of the British Expeditionary Force’s (BEF’s) reserve
situation convinced Rupprecht otherwise. Intelligence suggested that Haig
was not employing the divisions now arriving from Britain to build up
reserves for an attack, but to take over more of the front line from the
French. No British attack was likely before April or May, at the earliest.7 By
the end of March, with Verdun deadlocked and still no Entente relief
offensive in prospect, Kuhl was ready to write off the possibility of a British
attack: ‘in that sense the plan certainly didn’t work. Now the whole thing
has become a test of strength at Verdun, a question of prestige.’ Falkenhayn
did not finally give up on the idea of launching his own attack at Arras until
the end of April, when it became clear at last that the so-called ‘mill on the
Meuse’ was grinding through reserves so fast that it would be impossible to
spare any for Sixth Army. Losses were so heavy that some divisions were
being kept in the fighting line even after they had lost half their men. Hope
began to drain away. ‘What distresses me most’, noted Rupprecht, ‘is that
General von Falkenhayn apparently no longer thinks a decisive victory can
possibly be won. In that event, how must the war end for us?’8
Prospects had not improved by early summer. ‘Verdun is a real witches’
cauldron which swallows our best troops one after the other’, Rupprecht
complained. Attrition at Verdun began to bite directly into Sixth Army,
with Rupprecht forced to transfer out seven of his fresh divisions and receiv-
ing only five worn-out ones in return.9 Even during quiet periods, the
steady drip of casualties continued for Sixth Army. Artillery fire, sniping,
patrolling, raiding, and accidents drained strength all the time. For instance,
Sixth Army, now rather fewer than 200,000 men strong, lost 1,370 killed,
wounded, and missing in the ten days between 11 and 21 January 1916.
When combat grew more intense, inevitably wastage increased. The next
ten-day period, which included the ‘diversionary’ attacks north of Arras,
saw 4,700 lost. Small-scale operations to secure local tactical advantages
were frequent. Thus, in the evening of 20 February Sixth Army recaptured
the Gießler Heights. Accidents could be costly, however. On 29 April, for
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

96 T h e A n v i l 1915–16

example, German gas drifted back into the assembly trenches of a Bavarian
unit waiting to go over the top. The assault troops were wearing gas masks,
but many were overcome nonetheless. The final death toll reached 307.10
Accidents or worse killed civilians, too. In the early morning of 11 January
a dump of engineers’ stores exploded in Lille, where Rupprecht’s headquar-
ters were based. Thirty German soldiers and seventy French civilians were
killed. That was an accident, but the local population also fell victim to the
stupidity or negligence of the occupiers. Also in Lille, an outbreak of
typhoid was eventually traced to the latrines of a German artillery battery
which had been built over the city’s main water reservoir. Other diseases
were also on the rise: no fewer than 400 Frenchwomen were detained in
Douai on 8 March with venereal disease. Finally, in June 1916 a French
teacher in Douai became the victim of a sex murder committed by a
German soldier. This was the first case of its kind that Rupprecht had heard
of, but no doubt not the last.11
The war remained the priority for both sides and the imperative was to
absorb the lessons of Verdun. To the French, the fighting there seemed to
reinforce the lessons of 1915 and validate the principles enshrined in their
January 1916 doctrine, rather than throw up new ones. This applied equally
to the attack and the defence. In particular, it showed that thorough prepar-
ation was more important than surprise. Otherwise, the only changes
suggested were minor tactical ones. For Joffre, indeed, the main lesson was
in fact an ancient one:
The moral factor is more preponderant than ever. As the mechanical means of
destruction grow in power, so the capacity for resistance grows in the soul of
the warrior. . . . Troops who are well led, supported, informed, looked after,
fed  are capable of unlimited resistance. . . . The battle of Verdun has, in the
view of the high command, taught us so far nothing we did not already know
and teach; she must fix more deeply in the heart of everyone this truth: not a
single inch of ground must be given up voluntarily under any circumstances.
Troops, even when surrounded, must hold out to the last man, without retreating.
The sacrifice of each is the very condition of victory.

Even on the industrial battlefield, Joffre was arguing, it was still human will
which counted most.12
Joffre may have felt this a point worth making because throughout 1915
and 1916 the ‘mechanical means of destruction’ were growing in power.
The war was becoming noticeably more scientific. For example, it was pos-
sible to locate enemy guns by taking bearings on the reports made as they
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

V e r du n a n d t h e Roa d to t h e Som m e 97

fired and triangulating back. The Germans had experimented with this
sound-ranging technique at La Bassée in late 1914. French scientists alerted
the British to the method’s potential the following March. The British were
slower to pursue it than the French, but by the end of 1915 all three armies
had established sections of both sound-rangers and flash-spotters. It would
take time to prove their effectiveness, but the British alone had some 200
top-flight scientists working on the project. Likewise, now that artillery fire
was taking place at longer range and without line of sight from gun to tar-
get, accuracy became both more important and harder to achieve. One of
the improvements artillerymen tried was to take account of atmospheric
conditions when calculating the trajectories of their shells. A more obvious
example was the move over from cloth or leather caps to steel helmets. In
autumn 1915, the iconic Adrian helmet replaced the képi in the French
front line and by the middle of 1916 steel helmets were widespread on both
sides of the wire. Most famously of all, the tank was coming. The British
had developed a working prototype by Christmas 1915 and Joffre signed off
on the first French designs around the same time. More broadly, generals on
both sides were applying science to the problem of breaking in to the enemy
front. Only after a careful calculation of what the available artillery could
achieve would infantry objectives be set. Thus, for instance, Foch’s instruc-
tions for the Battle of the Somme included the calculation that it would
take 600 shells to clear 500 square metres of barbed wire. Similarly, Rupprecht
noted that to prepare an attack on a front of 100 metres required 800 rounds
from a field, or 400 from a heavy, artillery battery. A creeping barrage to
cover an assault needed one field battery every 100 metres, or one heavy bat-
tery every 150. Once you knew how many guns and shells you had, you could
work out how ambitious an attack you could plan. Wishful thinking could be
challenged and ends could now be calibrated to fit the available means.13
One of the lessons Falkenhayn drew from the early fighting at Verdun
and circulated to the army was that the front line must be thickly garrisoned
and firmly held. Rupprecht disagreed. He thought May 1915 had shown that
this approach left too many men exposed both to enemy artillery fire and
to being overrun by the first assault. It was cheaper to hold the front line
sparsely and counter-attack to win it back if necessary. Rupprecht decided
to remind his corps commanders of this verbally, deliberately undercutting
Falkenhayn’s authority. It is rare to find a record of informal communication
trumping formal doctrine in this way. This reminds us that, just because the
written record survives when quick chats, raised eyebrows, and meaningful
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

98 T h e A n v i l 1915–16

glances have been forgotten, not everyone always followed official doctrine.
One wonders, in fact, how carefully the doctrinal literature was sometimes
even read. Officers had many other calls on their time, after all, and often
more pressing uses for paper.14
By the summer of 1916, all three armies were beginning to master at least
some of the problems thrown up by the new warfare. They had surmounted
the health and sanitary challenges of the first winter. Production was ramp-
ing up to offset shortages of equipment and weapons and they had learnt
the logistics of sustaining millions of men in the field. They had dusted off
old technologies, such as the mortar, and refined them for the new century,
and were rapidly integrating brand new technologies, such as the aeroplane.
On the battlefield itself, the tactical and operational approach of the
German army to defence had been thoroughly tested. Soldiers and com-
manders were confident in their methods and leaders.They had constructed
defences in depth in case they needed them. The basic principle, however,
at least officially, remained that every foot of ground would be fiercely con-
tested. If the original garrison was forced to give way and the enemy broke
in, reserves were to counter-attack and recapture the lost positions. French
attack tactics had discarded surprise in favour of an artillery-heavy approach
involving careful and lengthy preparation, with limited objectives set within
range of the guns. They had yet to unlock the operational secret of main-
taining offensive momentum, however. Debate continued over the best
use of reserves and, although advocates of attrition such as Foch were in
the ascendant for the moment, plenty of senior officers were unconvinced
and still favoured going for a breakthrough. The British, as we saw in
Chapter 6, remained behind the French and Germans both tactically and
operationally.
The way of war of all three armies was on the brink of the severest test
any of them had yet faced. Over the late spring and early summer of 1916,
the first indications began to appear that the Entente was gearing up for an
offensive in Picardy. Spies in London and The Hague reported that an
attack was being prepared. Tactical intelligence was also generating worry-
ing signs. General Fritz von Below’s Second Army on the Somme first
expressed concerns in mid-May. By early June Kuhl was persuaded. As
tented camps sprang up around Albert and rail traffic increased, it became
clearer that the British and French were making preparations. In mid-June
the French 39th Division, part of the crack XX ‘Iron’ Corps, was identified
in the line just north of the River Somme. When Joffre himself visited XX
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

V e r du n a n d t h e Roa d to t h e Som m e 99

Corps a few days later, he unwittingly but effectively confirmed that an

offensive was imminent.15
Falkenhayn took time to convince. To begin with, he was distracted by
the early success of the Brusilov offensive in Russia, which began on 4 June.
He had to transfer troops from the west to counter it. When the situation
in the east began to stabilize, for a while Falkenhayn remained concerned
that Alsace-Lorraine was the likely target of any French offensive. On
20 June, however, he sent three divisions of reserves to reinforce General
von Below if needed. The beginning of the Entente artillery bombardment
on 24 June made it immediately clear that this was the real thing and that
Second Army was the primary target. Falkenhayn sent further heavy guns
and Kuhl concentrated Sixth Army’s artillery reserves so that they were
ready to move at short notice. He offered them to Second Army, but
Tappen, still running the operations department at OHL, said no. Tappen
and Falkenhayn remained concerned that the Somme operations were only
a diversion, intended to distract from a full-scale attack at Arras. Not until 5
July did they acknowledge the severity of the threat facing Below and
accord Second Army top priority. By then the British army had suffered
‘the catastrophe of 1 July’.16
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

10
Early Days on the Somme

T he first of July 1916 is, in the words of Gary Sheffield, ‘the most notori-
ous day in British military history’. Of the fourteen British divisions
which attacked that morning, nine either never made any progress or
lost  any ground they had won within hours. A diversionary attack at
Gommecourt achieved nothing. Along much of the front when night fell,
the only British troops forward of the start line were the wounded or the
dead. Villages such as Thiepval, Beaumont Hamel, and Serre wrote their
fatal names into the history of the British army as the calvaries of the British
Expeditionary Force (BEF). By the end of that day at least 19,240 officers
and men lay dead. In all, 57,740 soldiers of Sir Henry Rawlinson’s Fourth
Army had become casualties. Roughly every other man who went over the
top on 1 July was hit. Only on the British right was lasting progress made.
Mametz fell on 1 July, Fricourt the next day. Lieutenant-General Walter
Congreve’s XIII Corps, next to the French, was the only British formation
to take all its objectives on the first day.1
It would be a mistake, though, to see the first day of the Somme through
British eyes alone. Three French corps (six divisions) extended the offensive
south of the river. They performed uniformly well and took many of their
objectives. They cleared the German first position and in places broke into
the second line. Fayolle, now promoted to command the French Sixth
Army, had the problem his British counterpart would have loved to face:
how to exploit success. Immediately south of the river, tough troops from
French North and West Africa, whom we last met rushing Vimy Ridge,
hesitated just long enough for the Germans to pull together local reserves
and stop up the gaps in their line. Further north, General Maurice Balfourier
was keen to push his XX Corps on but could not get the support he wanted
from the British. Congreve himself was keen, but Rawlinson had never
shared Haig’s optimism and ambition and he was surprised by success where
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

E a r ly Days on t h e Som m e101

none had been planned for. He would not permit the release of XIII Corps’
reserves, which in any case were not ready to be committed so soon. This
was an early but ominous portent that maintaining the coordination of the
British and French armies on the Somme would not always run smoothly.2
One man’s disaster is not necessarily his enemy’s triumph. If Haig and
Rawlinson did not realize quite how much they had lost, neither did
German commanders yet feel that they had gained a lot. While they gath-
ered that they had beaten off nine British divisions, the fact remained that
on much of the Anglo-French front the defence had been overwhelmed.3
In the popular imagination, the first day of the Somme has become, in
the words of the best recent historian of the battle, ‘a metaphor for futility
and slaughter; a national trope and tragedy that defies understanding’. In
fact, however, there is nothing incomprehensible about the British tragedy
that day. Multiple convincing explanations have been put forward. Almost
at once, wounded British prisoners told their German interrogators that the
defeat was the result of: ‘the inexperience and uselessness of the young offi-
cers’; a failure to reinforce success; and inadequate artillery preparation
which left German wire intact and machine-guns ready to cut down the
attackers. To what extent the interrogators heard or reported what they
wanted to hear, of course, is impossible now to reconstruct. The British
official historian Sir James Edmonds offered a deeper analysis in the 1930s.
For him, the problems began at the top, with the choice to attack where
German defences were strong, rather than weak. This decision was forced
on the British by the demands of coalition politics, as was the selection of
day and even time for the assault. Disagreements within the chain of com-
mand over the objectives of the offensive, and so the correct operational
method to pursue, introduced confusion. The French wanted to wear down
the German army and relieve Verdun, proposing a series of limited-objective
attacks to achieve this. Rawlinson also favoured a step-by-step approach, but
Haig pushed for a deeper drive to try for a breakthrough. At the tactical
level, the success of XIII Corps, where surprise was greatest and the artillery
preparation most effective, partly thanks to French help, showed that pro-
gress was indeed possible. Problems with manufacturing quality, however,
continued to reduce the effectiveness of British guns and shells. Lastly,
Edmonds regarded British tactics as clumsy, inadequate, and an important
cause of the high casualty rate. He saw this as the inevitable result of using
ill-trained and inexperienced troops prematurely. Modern historians have
tended roughly to work within the analytic template established by Edmonds.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

102 T h e A n v i l 1915–16

Although they do not always agree on the exact weight to be attached to

each factor, they tend to believe that fundamental conceptual differences,
both between Haig and Rawlinson and between the British and French,
underlay much of what went wrong. A litany of tactical and operational
problems ensued, most of them at some point to do with the dominant
weapon of the First World War: artillery.4
The German view at the time was much more straightforward: ground
had been lost where Anglo-French artillery, guided by superior air power,
had shattered the defenders’ trenches and suppressed the German guns.
Counter-battery fire became impossible and both protective barrages and
harassing fire were weakened. Although Rupprecht criticized Falkenhayn
for not sending enough reinforcements, the official historians, who were
not normally sympathetic to Falkenhayn, judged that not even fresh troops
would have saved the ground lost on 1 July.5
There was no time yet for proper post-mortems since Second Army
remained under intense pressure. On 2 July German troops abandoned the
second line and the Flaucourt Plateau, falling back in some confusion to a
third position along the line of the river. The French had advanced nearly
8 kilometres and could see the rooftops of Péronne. By 5 July, however, the
French advance on the left bank of the Somme had stalled. Further progress
in the south would clearly require another major effort. Meanwhile the
sector between the Albert–Bapaume road and the Somme saw a series of
local British and French attacks to clean up the line and gain jumping-off
positions for the next assault. It took the British six days and seven attacks
to clear Trȏnes Wood. Contalmaison held out for four days, Mametz Wood
for six. The three British divisions involved here lost over 12,000 men.
British intelligence had under-estimated the speed of the German reaction
and hence exaggerated the chances that ‘just one more push’ would bring
success. These fights were poorly executed, small-scale, and uncoordinated,
but eventually cleared the way for a British assault on the German second
position along the Bazentin Ridge on 14 July. After an artillery b­ ombardment
of unprecedented concentration and a night approach march, Rawlinson
achieved surprise and captured the German second position. As  ever,
however, exploitation remained problematic.6 Between 2 and 13 July the
British launched forty-six poorly planned and supported attacks. These
cost 25,000 British casualties. Christopher Duffy has described this period
as a ‘prolonged chapter of wasted opportunities’. It did have the effect,
however, together with the Bastille Day attack on Bazentin Ridge, of
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

E a r ly Days on t h e Som m e103

hammering home to the Germans that they had lost both their balance
and the initiative.7
German unease manifested itself in four main ways. First, Falkenhayn and
Below forbade any further voluntary withdrawals. Their men must hold
their ground and fight to win back what they had lost. Second, reserves
were pulled in with increasing urgency from other sectors of the Western
Front. For instance, Oberste Heeresleitung (OHL) overruled Rupprecht’s
complaints that he could spare no troops and ordered him to transfer two
divisions to the Somme. Third, reinforcements were thrown in as they
arrived to prop up Second Army’s line. This piecemeal approach created
difficulties of command and control and inevitably undermined cohesion.
As Rupprecht noted: ‘When reinforcements eventually came up, it was too
late. They arrived in dribs and drabs and had to be committed to action at
once to block the gaps. Consequently, units are now so mixed together that
no-one knows what’s going on.’ The headquarters of XVII Corps at one
stage had units from no fewer than eleven different divisions under com-
mand. In all, slightly over forty-three divisions were sent to the Somme
battlefront, between the end of June and the end of August 1916, with
twenty-two being rotated out: a net reinforcement of twenty-one. Of those
forty-three fresh divisions, twenty-one came from Rupprecht’s Sixth Army,
which received just fourteen worn-out formations in return. In just three
months, therefore, the Somme consumed three times as many of Sixth
Army’s reserves as Verdun had in half a year: a useful index of stress.8
The fourth sign of nerves was turbulence in the command chain. Corps
commanders and staff officers up to and including the Second Army chief
of staff were purged. Second Army was split in two, with Below effectively
demoted and left in charge only of the portion north of the Somme, now
renamed First Army. The defensive expert Fritz von Loßberg was sent to
keep an eye on him as chief of staff. South of the river, General Max von
Gallwitz took over a new Second Army and was also given day-to-day
command of the whole battle with Below a subordinate. Again the German
army was improvising command arrangements, even after this had worked
so poorly in 1914 and 1915.9
Rupprecht felt Falkenhayn was scapegoating Below to cover up his own
failure to heed warnings of an attack. If anyone should be relieved or
demoted, he thought, it was Falkenhayn himself. The cabal which had
intrigued against Falkenhayn in January 1915 now saw another opportunity.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

104 T h e A n v i l 1915–16

The Chief of the General Staff was vulnerable on many levels. He had failed
to predict the attack on Second Army. More seriously, 1 July destroyed the
last shreds of credibility of his entire strategy. The unexpected power of the
Brusilov and Somme offensives had shown that he had under-estimated
both the Russian and French threats. It also pointed up the bankruptcy of
Verdun. Now that the British had failed to fall in with Falkenhayn’s plans
and rush an offensive at Arras, Verdun made sense only if it tied down and
ate up the French army. Even this it was evidently unable to do. For the first
time in the war, Germany felt menaced by enemies who seemed finally able
to string together a series of coordinated offensives from the Somme to the
steppes. The intrigue was not all about policy, of course. Power and personal
concerns also played a part in uniting the cabal: Hindenburg and Ludendorff
were agitating for greater independence of command in the east; Bethmann
Hollweg was worried by repeated rumours that Falkenhayn had ambitions
to take his job as Chancellor; and Rupprecht, quite simply, as Kuhl noted,
‘hates Falkenhayn, anyway’.10
During the first few days of July Rupprecht heard from Bavarian dip-
lomats in Berlin and at Imperial Headquarters that dissatisfaction with
Falkenhayn was growing within the army, at court, and across the govern-
ment more generally. On 5 July he wrote to the Bavarian representative in
Berlin, Hugo von Lerchenfeld, arguing that Germany would lose the war
unless Falkenhayn went. Lerchenfeld, as he was meant to, showed the letter
to the Chancellor. Bethmann Hollweg’s coded telegram in reply was sup-
portive but cautious. Bethmann Hollweg said that, as a civilian, he could not
be seen to interfere in military matters and that the message would be better
coming from a soldier. Rupprecht was not discouraged. Instead he busied
himself recruiting further members to the anti-Falkenhayn cabal. The
Governor-General of occupied Belgium, General Moritz von Bissing, was
a willing co-conspirator who also knew the Kaiser well. He proved a useful
source of advice, as well as opening another channel of communication to
the Chancellor. On 22 July, for instance, Bissing showed Rupprecht a letter
from Bethmann Hollweg which made clear that the Chancellor was scared
of moving directly against Falkenhayn himself. Bissing suggested Rupprecht
approach the Kaiser directly. ‘But how?’ asked Rupprecht. Other help
was offered from further afield. A Dr Gildemeister wrote from Bremen,
offering to pay off Falkenhayn’s allegedly huge gambling debts and suggest-
ing Ludendorff as a replacement.11
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

E a r ly Days on t h e Som m e105

Meanwhile, on 7 July Hindenburg demanded overall command of the

whole Eastern Front. He and Ludendorff repeatedly pressed for more
resources as a way of keeping up the pressure on Falkenhayn. Bethmann
Hollweg overcame his former scruples about intervening in military affairs
and sent a memorandum listing Falkenhayn’s shortcomings to the head of
the Kaiser’s Civil Cabinet, Rudolf von Valentini. Gradually Falkenhayn’s
support began to ebb away. Influential courtiers such as Generals Hans von
Plessen and Moriz von Lyncker, commandant of the Kaiser’s headquarters
and head of his Military Cabinet respectively, began to distance themselves.
Even the Kaiser’s backing was becoming more conditional and in late July,
he reorganized command in the east to give Hindenburg and Ludendorff
greater independence.
Falkenhayn’s response was to work yet harder. He intervened in ever
smaller details and tried to micro-manage the disposition of every battalion
on both fronts. The result, predictably, was over-work, insomnia, and illness.
By August, he was physically worn out and looked like an old man. The
intrigue against him did not let up. On 9 August the Kaiserin wrote to
Plessen asking him to set up a meeting for the Kaiser with Rupprecht and
Duke Albrecht behind Falkenhayn’s back. Four days later Falkenhayn
told the War Minster, Wild von Hohenborn, that he would not be able to
continue as Chief of the General Staff unless he possessed the full confi-
dence of the Kaiser. The further Hindenburg’s stock with Wilhelm II
rose, the more insecure Falkenhayn felt. If this was an attempt to extract
some sign of favour from the Kaiser, in the short term it worked. On
both 13 and 24 August, the Kaiser assured him: ‘we will stick together to
the end of the war’.12
On 16 August Bethmann Hollweg sent an urgent telegram to the Kaiser,
backing a request of Hindenburg’s for reinforcements and sharply criticizing
the way Falkenhayn was still concentrating on the Western Front. It was
becoming ever clearer, Bethmann Hollweg argued, that the best hope was
for a separate peace with Russia. The Kaiser, never the most tactful of men,
promptly passed the telegram to Falkenhayn. The Chancellor kept up the
pressure. Reinforced by a detailed brief from Lieutenant-Colonel Max
Bauer on Falkenhayn’s mistakes, he renewed his attack via the War Minister
on 23 August. The response of both Lyncker and Wild showed how keen
both were to follow, rather than lead, the Kaiser’s opinion. Falkenhayn, they
stated, was irreplaceable and enjoyed the full confidence of the army.13
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

106 T h e A n v i l 1915–16

Meanwhile, by coincidence or design, Rupprecht was also attacking

Falkenhayn. Perhaps emboldened by his recent promotion to Field Marshal
in both the Bavarian and German armies, on 21 August he sent off a formal
letter to Lyncker for him to pass on to the Kaiser. He stated that Falkenhayn
had lost the trust of the army and must go. He also requested a month’s leave
on medical grounds. His doctor was concerned about his heart and stom-
ach, and Rupprecht was having trouble sleeping. So far in two years of war
he had taken no leave whatsoever. The reply came in two parts. First, he was
told that a reorganization of commands might be imminent and he might
be needed. Rupprecht withdrew his leave request at once. Second, Lyncker
wrote back, non-committally, that no decision had been made but that it
was important to bear in mind Falkenhayn’s good service in 1914.
Further developments on both scores followed rapidly. Late in the even-
ing of 22 August a telegram arrived from OHL ordering Rupprecht to hand
over command of Sixth Army to Colonel-General von Falkenhausen. He
was being promoted to take over a group of four armies: Sixth in Artois;
First and Second on the Somme; and Seventh from Noyon down nearly
to Reims. Rupprecht was to be no longer a mere spectator of the struggle
on the Somme, but in charge of the defence. ‘Taking over command of
this army group brings me the gravest concerns’, he noted, ‘since with it
comes responsibility for the fighting on the Somme front: an extremely
thankless task.’14
On 27 August neutral Romania, against Falkenhayn’s predictions, finally
joined the danse macabre by declaring war on Austria-Hungary. The news
‘hit German headquarters like a bomb’. Here was something else Falkenhayn
had got wrong. Did this mean defeat for the Central Powers and the end of
the war? The Kaiser thought so, at first. Some of his advisers were less pes-
simistic. With the court divided, Admiral von Müller, head of the Imperial
Naval Cabinet, suggested calling for Hindenburg and asking his opinion. If
Hindenburg came, Falkenhayn would have to go. Everyone knew that.
Valentini agreed with Müller. Eventually, even Lyncker and Plessen silently
sided with the cabal.
If Falkenhayn ever had a chance to save his job by projecting optimism
and showing his grip of the new Romanian development at his regular mid-
day briefing to the Kaiser on 28 August, he was too tired and sick to take it.
Indeed, his lacklustre performance increased Wilhelm’s dissatisfaction with
him. Plessen and Lyncker saw their chance. They suggested changing
Falkenhayn for Hindenburg and Ludendorff. According to their accounts,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

E a r ly Days on t h e Som m e107

the Kaiser took some persuading. Lyncker probably brought up Rupprecht’s

letter to prove that the army no longer trusted Falkenhayn. Finally they
managed to win Wilhelm around. By lunchtime, the affair was settled:
Hindenburg and Ludendorff were summoned to imperial headquarters. As
anticipated, Falkenhayn submitted his resignation when he heard the news.
Within twenty-four hours, Falkenhayn had handed over to Hindenburg
and left.15
When Rupprecht heard the news, he was ‘radiant’.16 He wrote to his father
that ‘the change from Falkenhayn to Hindenburg is to be very much wel-
comed and it is only to be regretted that it did not happen before, since
Falkenhayn messed everything up’.17 The task he now faced was a big one,
but at least he could have some confidence in his immediate superiors at last.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

11
Rupprecht the General

B efore we examine Rupprecht’s part in the second half of the Battle of

the Somme, it is worth pausing to address two important aspects of his
military career. First, this chapter will explore his experience of daily life in
wartime. Then, it will examine how he fitted into the command system of
the German army. How did he relate to his chiefs of staff, his superiors, and
his subordinates? It will thus begin to address the two stereotypes about the
German army which were noted in the Introduction: that the General Staff
ran matters and commanding generals were little more than decorative
figureheads; and that Auftragstaktik, or ‘mission command’, was the defining
principle of the whole system, lent the army a unique flexibility, and con-
tributed to excellent battlefield performance.1
On an average day, Rupprecht rose and was dressing by seven o’clock in
the morning. A naturally fastidious man, he knew the importance of dress-
ing the part and setting a good example, so he always made sure he was
immaculately turned out. Rather taller than average for the time at five foot
ten inches, he was a naturally active man who ate simply and did physical
exercises daily to keep his figure trim. His bearing was erect and martial.
After breakfast, the first wave of reports and messages would arrive between
eight and nine o’clock. If nothing urgent cropped up, he would then head
out on horseback with one of his adjutants for his daily ride.
Unless a major battle was underway, the first important event of the day
was the noon briefing. By that time Rupprecht’s staff would have collated
situation reports from the front and any incoming messages from Oberste
Heeresleitung (OHL). The chief of staff and department heads would each
make their report. Then action was discussed and any decisions necessary
were made. Lunch followed at one o’clock. Rupprecht might spend the
afternoon on paperwork, or drive to visit some of his units. There he
met commanders, inspected troops, and awarded medals. It was widely
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e Ge n e r a l109

accepted that it was important for generals to see and be seen by their men.
Sir Douglas Haig, likewise, spent most afternoons on such visits. Rupprecht
was not so assiduous as Haig, however. He was no natural orator and the
common touch did not come easily to him. Consequently, early on Krafft
had to push him to get out and about. Over time, however, Rupprecht grew
accustomed to this aspect of his job. In November 1915, for instance, he
spent several days reviewing various regiments, making a ‘very good, brief
address’ to the assembled troops, and then making a point of speaking to
many of the officers and men individually or in small groups. These visits
also enabled him to cast an eye over his soldiers and their leaders.2
By seven o’clock, Rupprecht would be back at headquarters for the even-
ing briefing, with dinner served at eight. He did not generally eat in the
officers’ mess but privately, with selected members of his staff and any guests.
Food was generally plain, talk of the war discouraged, and after eating the
time was spent chatting or playing cards. Occasionally they played Boccia, a
form of indoor bowls. Entertaining distinguished visitors was an important
part of Rupprecht’s role. The famous Swedish explorer Sven Hedin has left
a vivid, if rather gushing, portrait of Rupprecht and his hospitality in
November 1914:
He is one of those rare men whom all love and admire. . . . In the German
army he is looked upon as a very eminent general—a born strategist and a
thoroughly schooled soldier. As regards appearance, manner and speech, he is
fascinating and congenial in the highest degree, neither regal nor humble, but
without artifice and modest like an ordinary mortal. When one knows that he
has recently experienced the greatest private sorrow which could befall him,
one fancies, perhaps, that one detects a trace thereof in his features—an air of
sadness—but otherwise he does not betray, by a look or a sigh, how deeply he
grieves over the death of the little prince of thirteen, the darling of all Bavaria.
When the country and the empire are in danger, all private sorrows must be
put aside! The Crown Prince has no time to grieve or to think of the void and
bereavement which he will feel on his victorious return to Munich. He lives
for and with his army, and is like a father to each and all of his soldiers. He
devotes all his power of mind, all his physical strength, all his time, to the one
great object which dominates all else in the minds of the whole German army.
Crown Prince Rupprecht walks in with brisk and easy stride, stretches out his
hands towards us and gives us a truly cordial welcome.

The twenty guests that night included Hedin, a duke of Mecklenburg who
was also an explorer, Prince Löwenstein, and the Kaiser. They began eating
at eight and left the table once the Kaiser had finished his cigar at half past
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

110 T h e A n v i l 1915–16

ten. In May 1918 Rupprecht hosted four such banquets within a little over
a week. Guests might include other royals, senior generals, politicians, bankers,
and industrialists from Germany, her allies, or indeed neutral states. To
entertain VIPs in the appropriate manner was an important part of his job.3
Circumstances could alter this routine, of course. Rupprecht might have
to travel to OHL for meetings, or host conferences at his headquarters.
Major battles obviously could become all-consuming as telephones jangled,
reports arrived, and situation maps were updated. On 25 August 1914,
Rupprecht did not leave the operations room for a minute all day. As both
he and his staff grew more experienced, however, he learnt to stand back a
bit. After all, there were limits to how much real-time control was possible
with the communications available. After the war, Kuhl described Rupprecht’s
way of working. He had:
a strikingly good grasp and deep understanding of operational situations. He
was extremely hard-working, and had all the precise battle maps in his office
and kept them up to date, he worked through all messages, letters, orders and
reports in detail and was always perfectly informed. Consequently it was easy
to brief him and to propose a decision. He would quickly grasp the situation
and had himself already considered the conclusion. He was never nervy; even
in the most difficult combat situations he allowed us in the staff the time to
collate and inspect the messages and never pressed us or disrupted the process
by interfering. He did not chase around after every message but waited until
the situation had become clear enough that one could get a sufficient over-
view and a definite course of action could be decided on. He always made a
bold decision. He was highly respected by his entire staff. Relationships within
the staff and between him and his staff were the best imaginable.
It is impossible to be sure that Kuhl consistently felt so bullish about
Rupprecht during the war itself, but nor is it clear what he had to gain from
over-praising him in the 1920s.4
The size of Rupprecht’s staff grew rapidly as the war went on. His imme-
diate command group in August 1914 comprised fifteen officers. By September
1916 it was twenty-three strong and in late 1918 it had grown to forty-five
officers within a broader headquarters over 1,000 strong. In 1914 it had
been possible to squeeze headquarters into a school. By April 1917 it had
expanded so much that half the old town of Mons was closed off to accom-
modate it, with departments filling offices throughout the southern and
­eastern parts of the town. The operations centre occupied 4, Avenue de
Frère Orban. Officers were billeted around the Grand-Place while Rupprecht
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e Ge n e r a l111

himself slept at the Chateau Hardenpont, 2 kilometres away on the outskirts

of the town.5
Life for the members of Rupprecht’s headquarters revolved around long
hours of paperwork and meals in the mess. Officers paid 1.20 marks per day
for breakfast, lunch, and dinner. These were served from seven a.m., one
p.m., and eight p.m. respectively. Until August 1915 wine, schnapps, and
cigars were free, but later they had to pay for their own. Hard as the work
could be, there was always time to celebrate special occasions. Sometimes
this might take place formally. In 1916, for instance, Rupprecht hosted a
party for his officers on Christmas Eve before appearing as guest of honour
in the Senior Officers’ Mess on Christmas Day itself. The Junior Officers’
Mess took their turn to lead the celebrations on New Year’s Eve. At times
the parties became more boisterous. On 26 January 1915 a ‘lavish dinner’
thrown by the Bavarian officers of Sixth Army headquarters for their
Prussian colleagues to celebrate the Kaiser’s birthday went on until half past
one in the morning, despite a spirited argument over how to deal with
Austria and the partition of Poland. Staff officers were afforded other com-
forts denied to their front-line compatriots.While Xylander was convalescing
from an operation on his leg, his wife came to stay for several weeks to nurse
him.6 Some found other ways to enjoy female company. One evening in
Douai the Turkish liaison officer attached to Rupprecht’s headquarters,
Zekki Pasha, entertained a young Frenchwoman for an hour or two. Losing
track of time, she left after curfew. On her way home, three times she was
stopped by patrols. Each time they refused to allow her to proceed and
escorted her back to his quarters. Eventually, she decided that she might as
well accept her manifest destiny and spend the night at his billet. Rupprecht
had slightly more discreet arrangements. These included visiting a young
Belgian who had previously been under the protection of a wealthy Douai
businessman. Creeping out of his house at night to visit her was hardly dig-
nified, however, and eventually the story reached the French newspapers.7
Rupprecht also kept up some of his other old hobbies, such as his interest
in art and architecture. He carved out time where he could to visit collec-
tions and buildings of interest. One day in October 1916, for instance, on his
way back from a visit to Seventh Army, he stopped off to tour the town of
Laon, the Prince of Monaco’s home at Marchais and the medieval castle at
Coucy-le-Chateau: ‘a unique sight, the proudest castle I’ve ever seen. . . . It
would be a great shame, if the castle became a victim of this war.’ Similarly,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

112 T h e A n v i l 1915–16

Figure 11.1.  Rupprecht and his son Albrecht

on 27 March 1917, Rupprecht visited Courtrai and Bruges, exploring

Flemish decoration and architecture of the Middle Ages. Less cerebral
pursuits, such as shooting hare or pheasant, were popular over the winter.
When headquarters moved to Mons in spring 1917, a large area outside the
town was set aside for hunting. Finally, Rupprecht’s son Albrecht generally
came to stay during his school holidays. Xylander described him as ‘a really
nice and well brought up young man, who speaks like a true Bavarian’. For
Christmas 1915, Rupprecht gave Albrecht a bicycle. The fact that even
when things were quiet, however, Rupprecht could spend no more than a
couple of hours with him each day highlights the pressures upon the
Crown Prince.8
The best way to escape the stresses of war was to go home. For the first
two years of the war, Rupprecht did not take any leave at all. From late 1916
on, however, he made a total of five trips to Bavaria, each lasting on average
between ten days and a fortnight. These trips were not all holiday. In Munich
he often had to spend time meeting politicians, for example, and much of
his February 1918 trip was filled by official duties connected to his parents’
Golden Wedding anniversary. These trips gave him a break from the war,
however, and allowed him to spend further time with his son. In season
Rupprecht would go hunting in the Berchtesgaden Alps. Rupprecht was
fortunate in this regard: ordinary soldiers might get no more than a week or
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e Ge n e r a l113

ten days of leave once a year and Kuhl took none at all for four years, seeing
his family only if he happened to be in transit through Berlin.
In summary, Rupprecht took the lead on ceremonial aspects of the job
while his staff carried the main burden of administration and paperwork.
Each army possessed a large and complex administrative apparatus. It needed
to feed, clothe, train, and supply hundreds of thousands of men. Someone
had to keep accounts and maintain military discipline, collate intelligence,
prepare briefing notes, and draft orders. The lessons generated by operations
must be collected, doctrine adjusted where necessary, and new techniques
disseminated around the army. All these were jobs obviously best suited for
trained staff officers to carry out. Equally, when it came to representing his
command in negotiations with other formations or OHL, or to making the
operational decisions which might make the difference between victory
and defeat, and would certainly affect many lives, then charisma, and the
habit of authority that the royal general possessed, might be more useful.
When it came to decision-making, however, quite where the responsibil-
ities of the army commander stopped and those of the chief of staff began
is not easy to untangle. The two were expected to work together as a team
but there was no set template of how they should do so. It was up to the
individuals concerned to achieve their own equilibrium. Rupprecht had a
different relationship with each of his three chiefs of staff. Indeed, each rela-
tionship itself evolved over time as the two partners developed and external
stimuli changed.
The appointment of Konrad Krafft von Dellmensingen as his first chief
of staff came as a pleasant surprise to Rupprecht. He was delighted to have
a Bavarian officer to work with and knew Krafft of old. He respected his
professional ability and the two men seem to have forged and maintained a
close bond from the start. To begin with, Rupprecht seems to have deferred
to Krafft. Krafft led most of the early strategy discussions with OHL, for
instance. On occasion, he even appeared to boss Rupprecht about, asking
him to go and hurry up a laggard corps on 27 September 1914, for example.
Rupprecht had too much confidence in his own judgement to be a cipher,
though. When he felt the need, he was happy to criticize the handling of
his troops, as we saw with the Life Guards at Badonviller (14 August).
Rupprecht was soon intervening in planning. On 18 November he argued
that the strength available for a proposed offensive was inadequate. This
happened ‘without me suggesting it to him’, Krafft noted. By January 1915,
Krafft was beginning sometimes to defer to Rupprecht. He gave up the idea
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

114 T h e A n v i l 1915–16

of taking Carency Wood after Rupprecht argued that the terrain was too
difficult, for instance. Rupprecht had served his apprenticeship. His partner-
ship with Krafft seems to have become one of military equals.9
Lambsdorff ’s appointment to succeed Krafft ironically helped Rupprecht
grow as a general. Although first impressions were favourable, Rupprecht
soon began to wonder whether he could trust him. During the dispute over
Lochow’s independent command in June 1915, the chief of staff seemed to
be on Falkenhayn’s side rather than Rupprecht’s. When the French attacked
in September, Lambsdorff seemed incapable of making a decision. By the
middle of October, Rupprecht’s patience was wearing thin:
He wants to have control of everything, briefs me very poorly and was not
always very well informed himself since he will not delegate to his colleagues. . . .
His main activity was almost every day to visit one headquarters or another,
inviting himself to breakfast [regardless of the needs of the situation]. . . . I did
not trust Lambsdorff in a tight spot. A cold, smooth courtier, but not a real man.

Lambsdorff even began to try to cut Rupprecht out of decision-making. He

showed him orders only at the last minute before they had to go out, or not
at all. When Rupprecht found out that Lambsdorff had sought to conceal
the loss of the important Gießler Heights, he was furious. Xylander reck-
oned that he had never seen him so angry. Rupprecht immediately wrote to
the Chief of the Kaiser’s Military Cabinet asking for Lambsdorff ’s relief and
within the week he was moved sideways and out.10
Lack of trust in Lambsdorff forced Rupprecht to become more self-
reliant. When Hermann von Kuhl became Rupprecht’s third and last chief
of staff, therefore, Kuhl found himself working with an experienced general.
Kuhl was very thoroughly trained—he had been a protegé of Schlieffen’s—
and had spent the war so far serving as chief of staff to generals von Kluck
and Fabeck. The former, especially, had a reputation for being tricky but
Kuhl had managed him skilfully. Falkenhayn warned him that he would
need to be diplomatic in his new post:
Crown Prince of Bavaria [is] rather difficult, defensive about Bavarian rights,
one minute exulting to the heavens, the next gloomy as death. Likewise he
swings one minute from being unable to hold his position and having to give
ground, to everything going brilliantly the next. His previous chief of staff,
Graf Lambsdorff, and he could not get along.

Kuhl reached Sixth Army headquarters on 1 December 1915. He was

welcomed with a gala dinner of Bavarian sausages and the two men rapidly
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e Ge n e r a l115

grew to trust one another. They shared a view of offensive tactics and
together disagreed with Falkenhayn’s insistence that the British were
planning to attack Sixth Army in early 1916. Rupprecht was fully capable of
holding his own with his chief of staff as something close to military equals.
As we shall see, during the second half of the war both Rupprecht and Kuhl
were capable of taking the lead and proposing new ideas for how to pro-
ceed. Throughout the battles of 1916 and 1917 it is hard to see the join
between the two men.11
Rupprecht’s influence as a general was defined by his relationships, not
only with his chief of staff, but also with his superiors and subordinates.
Again, there was no agreed definition of the responsibilities and power of an
army or Army Group commander. Rupprecht’s role fluctuated in import-
ance depending on how much latitude OHL was prepared to allow him and
how far he in turn was willing to leave his subordinates to run their own
shows. The traditional view of the German army has been that, by virtue of
its tradition of Auftragstaktik, it has always been comfortable with a high
level of decentralized command. However true this may be at the tactical
level, the rest of this chapter will argue that operationally it is mistaken.
Above divisional level, delegation was the exception, not the rule.
Rupprecht’s boss was in effect the Chief of the General Staff, to whom
the Kaiser tended to delegate day-to-day command of the German army.
Traditionally, Moltke the Younger has been seen as a laissez-faire holder of
this post. Through a combination of poor communications, illness, and
incompetence, he passed decision-making power down to army commanders
who, seeing themselves as some kind of semi-feudal paladins of the Kaiser,
were more than happy to exercise their own authority. In reality, Moltke was
consistent only in his inconsistency. Under him, OHL veered between
trying to pull every string and abdicating all responsibility. As we saw in
Chapter 2, the decision to let Rupprecht attack on 20 August 1914 was an
example of the latter. Between 23 and 26 August OHL followed up with
attempts at close central control, every day ordering him to attack in a
different direction. Including Moltke, six different OHL officers worked
liaison with Sixth Army at some point that month. Lack of consistency in
personnel and intention generated destructive uncertainty which left deci-
sion-makers at all levels unsure of both their powers and responsibilities.12
This shambles could not, and did not, last. Neither did the promises of a
free hand which Falkenhayn made to Rupprecht once he took over as Chief
of the General Staff. His insistence that Sixth Army engage piecemeal as its
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

116 T h e A n v i l 1915–16

units arrived on the Somme in September 1914 left Rupprecht no room for
manoeuvre. In October, Falkenhayn would not allow Rupprecht to fight
the Flanders battle as he saw fit. He insisted that Sixth Army wait for the
new Fourth Army to arrive before advancing: ‘even if the Crown Prince of
Bavaria stands on his head, he may not attack!’ Trust between Falkenhayn
and Rupprecht rapidly evaporated and each began actively to undermine
the other. In this environment, OHL was not keen to delegate. In any case,
the changed realities of trench fighting, with nearly fixed battle-lines and
telephones everywhere, permitted a degree of top-down control unimagin-
able during mobile warfare. When the French attacked in Artois in spring
1915, Falkenhayn tried to bypass Rupprecht by installing Lochow and telling
him to report direct to OHL. Interestingly, Rupprecht does not seem to
have complained about interference from OHL during the Third Battle of
Artois in the same way, but that may be just because Artois was only a side-
show and Falkenhayn’s attention was elsewhere.13
When Rupprecht moved to command his Army Group in August 1916
and Hindenburg and Ludendorff took over at OHL, roles were recast.
The Army Group was intended to be a high-level command which con-
centrated solely on tactics, operations, and related personnel matters.
Administrative, financial, and legal matters remained the responsibility of
each army. The Army Group was to take over the job of managing the pro-
cess of roulement by which formations were rotated in and out of the line.
From now on, Army Groups would be free to transfer formations from
army to army without asking OHL’s permission each time. They should be
sufficiently large to generate reserves internally without having to request
outside reinforcement. At their initial meeting at Cambrai on 8 September
1916, Ludendorff and Hindenburg reacted against Falkenhayn’s record of
interference and promised Rupprecht that they would leave him to run his
Army Group. At least to begin with, they resisted the temptation to meddle
as Falkenhayn had. After that first meeting, Rupprecht did not even see
Hindenburg again for eight months. Once Ludendorff had checked and
agreed Rupprecht’s scheme of roulement, ‘for the whole length of the
immense and bloody Battle of the Somme’, Rupprecht later reminisced,
‘I was given a free hand’.This seems to have been part of a broader programme
of allowing subordinate commanders greater latitude, as we shall see.14
Did Rupprecht make the same mistake of over-controlling those beneath
him in the chain of command? Relations with his subordinates were not
always easy, but on the whole he tried to leave them plenty of scope for
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e Ge n e r a l117

exercising their own initiative. During the mobile warfare of 1914, Rupprecht
had trouble with those under him. Some of this was the result of personality
clashes. It was perhaps tactless to put Rupprecht in charge of General von
Heeringen in the first few weeks of the war. Heeringen was a highly experi-
enced general and former War Minister, nineteen years older than Rupprecht.
More importantly, though, since neither had worked together before and all
arrangements were being improvised, coordination remained poor. This ad
hoc Army Group was soon broken up.
Communications proved another problem. As we saw above, the infra-
structure did not exist for Sixth Army headquarters to keep in regular, much
less continuous, touch with its units in mobile warfare. By the time of their
attack on 20 August 1914, Rupprecht still had no established link to at least
one of his corps. This gave ambitious generals such as Ludwig von Gebsattel
scope to follow their own star, rather than the orders they had been given.
The defeat of Gebsattel’s attack on 25 August, undertaken despite instruc-
tions to remain on the defensive and guard the right flank, caused problems
for Rupprecht’s larger operations. Even where there was no active insubor-
dination of this kind, it took time for all levels of command to adjust to the
fact that battlefield communications were subject to even greater lags and
friction than they had imagined. One way to get around this was to decen-
tralize command, but this was a lesson not learnt until the fighting on the
Somme in late September 1914.15
Once both sides settled into their trenches, the dynamic and rhythm of
the war changed. Options closed down as the front froze, especially for the
side on the strategic defensive. While Sixth Army did its best to ensure its
troops kept up to date with the latest defence doctrine and organization, it
needed to spend less time making operational decisions and more managing
the flow of manpower and matériel which the new warfare devoured in ever
greater quantities. Even if the enemy attacked, Sixth Army generally did little
beyond allocating reserves where they were most required and reviewing
performance to weed out tired and incompetent commanders and learn
lessons for next time. Most of the time, the threat required little more.
On occasion, however, a major decision did become necessary. A good
example is the counter-attack of VII Corps at Neuve Chapelle on 12 March
1915. Rupprecht set the objective and provided the resources required but
left the detailed planning to the corps commander. As we saw, he regretted
not having been firmer with Claer but did not alter his practice much
thereafter. On the whole, although Rupprecht always offered advice and
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

118 T h e A n v i l 1915–16

retained the right of veto, he tended to leave corps (and, later, armies) to
fight their own battles. Rupprecht’s headquarters was more often occupied
with managing the overall effort than with minute-by-minute decision-
making. Most of the time, the Entente attacks were not threatening enough
to force more active intervention. One exception, the crisis of the Second
Battle of Artois, proves the rule: such was the French pressure by 11 May
that Rupprecht bypassed corps command and directly ordered one of the
divisional commanders to clear the lines of communication to the village of
Carency. This kind of intervention two rungs down the chain of command,
while much simplified by the well-developed communications networks of
trench warfare, remained rare. As we have seen, more often Rupprecht’s
interventions involved asking OHL for reinforcements, directing them
where they were most needed, and approving or suggesting counter-attacks.
The nature of defensive trench warfare left little scope for much else. The
conduct of the Battle of the Somme generally unfolded along similar lines,
as Chapter 12 shows.
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

PaRt III
Holding the line
1916–17
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

12
Rupprecht Takes Command

R upprecht wasted no time in getting briefed on his new command and

the battle underway. Overall, although the Somme battle was putting
the German army under considerable strain and losses were heavy, things
appeared to be going tolerably well. German casualties so far totalled 195,172
men. The majority of the German divisions committed to action north of
the river had lost at least half their infantry, and some had lost considerably
more. These figures were too high for comfort but there was no cause for
immediate panic. Germany still had some 1,400,000 men in reserve if
required. Further, high as German casualties were, British losses were esti-
mated to be five times as many. This was later revealed to be an exaggeration:
the British had ‘only’ lost 190,000 men and the French 80,000. The balance
of attrition was much less favourable than thought. Rupprecht, though, had
no way of knowing this at the time.1
Manpower, however, was only one dimension of war, 1916 style. Machines
were equally important. The bravest of soldiers could achieve nothing with-
out the tools of battle. German industry was ramping up extremely rapidly.
For example, monthly output of field artillery ammunition rose 60 per
cent between the second half of 1915 and summer 1916. The trouble was
that consumption had doubled and the German army had begun to eat into
its stocks. Ludendorff vowed to increase supply again to 300 trainloads per
month and re-establish equilibrium. Rupprecht, he promised, would receive
70 per cent of the output.2
At the operational level, in his four armies Rupprecht commanded forty-
five divisions and some 700,000 men. Sixth Army, occupying the line from
the Belgian border down to just south of Arras, of course, he knew well. It
gave no cause for concern. The thirteen divisions which made it up were
in good hands under Falkenhausen: ‘despite his 73 years he is still very
flexible and youthful, a handsome man with a sharply aquiline nose and
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

122 Holdi ng t h e Li n e 1916 –17

Figure 12.1.  Rupprecht and his staff

bright eyes’. In any case, the British were unlikely to have the resources to
attack here. Things were also likely to remain quiet on the front of Seventh
Army between Noyon and Reims. The commander of Seventh Army,
Heeringen, was quickly promoted away to avoid further clashes.
The main enemy pressure continued to be between the rivers Ancre and
Somme against Below’s First Army. Here, twenty-one German divisions,
eleven in the front line and ten in reserve, were being attacked by thirty-two
British and four French ones. In the six weeks since his success of 14 July,
Haig had been trying to secure the observation posts and jumping-off
positions he needed for his next major drive. A series of small-scale but
intense battles for places such as Pozières, Mouquet Farm, and High and
Delville Woods eventually added up to an advance of a thousand metres
or so on an 8-kilometre front for Rawlinson’s Fourth Army. The cost was
82,000 British soldiers.3
The enemy was most menacing in three sectors. First, his drive on Mouquet
Farm threatened to cut off the salient at Thiepval. Second, British pressure
around Delville Wood and Ginchy threatened the high ground there and
at High Wood. Third, the French were driving eastward into the German
second position on the right bank of the Somme. Below was confident that
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d123

he could hold the first and last of these three sectors. In the centre, however,
he was more concerned and was planning a large operation to recapture
Longueval and Delville Wood on 31 August. The morale of his army, Below
reported, was mixed. His men were unconcerned by the Entente advantage
in manpower, but felt keenly that they were inferior in ammunition. Since
enemy gunfire had destroyed fixed fortifications and barbed wire, leaving
the defenders to fight from improvised positions in shell holes, they relied
especially on protective barrages, which needed shells. Below seemed pleased
with the new command arrangements. He and Rupprecht had always got
on well but Below had found it difficult working for Gallwitz, whom he
accused of funnelling resources First Army needed to his own Second Army.
Gallwitz himself felt that he had been passed over for command of the
Army Group and was prepared to be difficult with Rupprecht about it.
As  we shall see, he succeeded in causing considerable irritation as the
battle went on.
The military situation of Second Army reassured Rupprecht, however.
Intelligence counted twenty-two French divisions opposite Gallwitz’s eleven.
Although Second Army was low on reserves, it had plenty of guns and
ammunition to hand. One difference between the two armies, though, was
that under Below almost all the heavy artillery was parcelled out to the
command of divisions. In Gallwitz’s army, on the other hand, the majority
remained under corps control. This more centralized approach ran the
risk, in Rupprecht’s opinion, of being slower to respond when needed in
a hurry.4
The first test of the new Army Group and its commander was not long
in coming. On 3 September eight British and four French divisions renewed
the assault from Thiepval to the Somme. The defenders fought off rather
clumsy British efforts at Thiepval, Mouquet Farm, and High Wood yet
again. Further east, Rawlinson’s men were a little more successful. After
nearly a week of vicious fighting they finally cleared Ginchy, Delville Wood,
and Guillemont, or what little was left of them. The French, however, gave
the Germans a real scare. Fayolle’s troops captured over 4,000 Germans and
drove up to 3 kilometres into the German lines along the banks of the river.
This Anglo-French attack on First Army was ‘the strongest yet’ and Rupprecht
moved reserves up to support Below’s left. Gallwitz protested noisily that he
needed reinforcement, too.5
Gallwitz was proved right almost at once. At four p.m. on 4 September
General Micheler’s French Tenth Army attacked across the Santerre Plateau,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

124 Holdi ng t h e Li n e 1916 –17

aiming for the Somme crossings upstream of Péronne. By nine o’clock that
evening, Second Army reported, the whole front line had been lost and
there was a risk its guns might be overrun. Reserves would have to be sent
to Gallwitz after all, however great Below’s need. Rupprecht noted:
Unfortunately our men on the Somme front are on the whole simply not the
best and those who take their places certainly won’t be any better.The veteran
officers and other ranks become fewer every day and the replacements, however
plentiful, have not had the benefit of the same military instruction and training.
Physically also, they are mainly of low quality. . . . The situation is definitely
extremely strained and dangerous for us.

The next day brought more heavy combat and further Entente gains, but
also showed that the situation south of the Somme was not as poor as first
feared. French progress in the south had been exaggerated. Much of the
front line remained in German hands. Nonetheless, Oberste Heeresleitung
(OHL) designated the situation ‘very serious’ and ordered six divisions,
artillery, and air squadrons to be stripped from other sectors of the front and
rushed to the Somme as quickly as the congested railways would allow.6
Rupprecht was impressed by the reaction of the new OHL to its first
Western Front test: ‘the change at OHL makes itself very pleasantly felt. What
they arrange is effective and they take care to provide the best possible
resources to their projects.’ Ludendorff asked Rupprecht for the schedule of
reliefs for the front-line troops and also how many replacements he needed.
Falkenhayn had never been so professional. Managing a sustainable system
of roulement to replace worn-out divisions with fresh ones now became
thought so crucial that it sometimes took priority over other operational
requirements. For example, Rupprecht refused to commit reserves to an
attack on Delville Wood on 31 August because that would involve disrupt-
ing the relief schedule. He was worried about the resilience of his units: ‘the
ability of our men to resist is unfortunately not what it once was, and I can
see with concern the time coming when divisions which have already been
committed once at Verdun and on the Somme must be put back in to
where the heat of battle is highest’.7
The question of roulement came up at a conference of Western Front
commanders held at Cambrai on 8 September. Present were Hindenburg,
Ludendorff, Rupprecht, Crown Prince Wilhelm, Duke Albrecht of
Württemberg, Fritz von Below, Gallwitz, and their chiefs of staff. In general,
Rupprecht found himself impressed with the two men now running the army.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d125

In his diary he recorded his impressions with a typically nineteenth-century

emphasis on physiognomy:
Hindenburg creates a good impression. His forehead is strikingly broad, his
eyes are blue and create an open and kind effect. His bearing is composed and
determined.He loves hunting and is a joker but very approachable . . . Ludendorff
has a well-defined, smart head. His face is very pale and I had the impression
that he ought to take better care of himself. What he said was all spot on.

At the conference it was announced that eight new divisions were to be

raised, five for the east and three for the west, of which two were intended
to reinforce Army Group Crown Prince Rupprecht. More detailed discus-
sion followed. The Second Army chief of staff, Colonel Bernhard Bronsart
von Schellendorff, proposed speeding up reliefs by rotating divisions in and
out of the line, rather than corps. He argued that imitating the British
Expeditionary Force (BEF) and leaving corps headquarters in place at the
front would generate greater continuity in command. There were obvious
drawbacks, however. First, command cohesion would suffer: divisions would
take time to get used to their new corps and vice versa. Second, corps played
a vital role in coordinating artillery which it would be a mistake to dilute.
Third, corps commanders were less likely to push strangers as hard as they
could those they knew well. Lastly, rotating divisions rather than corps would
make the army commander’s job harder. Below’s chief of staff proposed
instead that part of the staff of the corps being relieved should stay in situ for
a few days to ease the transition. Rupprecht preferred this solution.8
As Alexander Watson has pointed out, the Germans suffered more casualties
at the start of September than in any other ten-day period on the Somme
in absolute terms, with the exception only of the first ten days in July. As a
proportion of overall strength, early September was probably the bloodiest
period of the whole campaign. First and Second armies lost 13 per cent of
their strength, compared to 10 per cent casualties between 1 and 10 July and
considerably higher than the average ‘normal’ rate on the Somme of between
6 and 8 per cent every ten days. Concerns were now beginning to grow
about both the quality and quantity of manpower. OHL began to toy with
the idea of withdrawing to the line Arras–Laon–Verdun to shorten the line
and free up reserves. At the Cambrai conference of 8 September, when he
warned that if losses carried on at the present rate they would be forced to
make peace before being bled dry, he found German Crown Prince Wilhelm
agreeing with him.9
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

126 Holdi ng t h e Li n e 1916 –17

In the middle of September, the pressure ratcheted up another notch. By

now, Foch had given up trying to arrange a tactically synchronized British
and French attack. The two armies seemed never to be ready to attack sim-
ultaneously. The British official historian blamed the French for not trying
hard enough to support their ally and repeatedly requesting postponements
to operations. He did not mention that British requests for delay had become
fairly frequent, too. The two allies were, however, capable of offering each
other a measure of support at the operational level. They could stage assaults
close enough together to generate some overall momentum. Some historians
see this as a deliberate policy by Foch. Given the emphasis usually placed on
simultaneous attacks, however, one suspects this was a militarily second-best
method which had the political advantage of tactfully minimizing inter-allied
recrimination.10
Either way, on 12 September the French struck again. Fayolle’s Sixth
Army tried to overrun the German third position and outflank Péronne
from the north. Five divisions drove eastwards towards Bouchavesnes and
the Bapaume–Péronne road. The attack was extremely successful. According
to the official history, German protective artillery fire came down too late
and missed the attackers. Consequently, the French were able to overrun the
Saxon defenders in their dugouts. In Rupprecht’s version, the Saxons simply
ran away. Within half an hour Bouchavesnes had fallen to the French, who
went on to seize nearly 7 kilometres of the German third position and open
a gap 1,500 metres wide. To many on both sides, it seemed that the elusive
breakthrough might at last have been made. Fayolle, much to his surprise,
found himself enthusiastically embraced by the normally stolid Joffre.
The next German defensive line remained only sketchy and could not be
relied upon to offer much obstacle. First Army began preparations to recap-
ture Bouchavesnes, although it would take time to bring up reserves and
they could not launch a counter-attack for several days. When the counter-
attack finally took place, on 20 September, it failed. By then, however,
French efforts to exploit the breakthrough had also ground to a halt. The
gap had simply been too narrow and any attempt to push on had been
chopped up by fire from the flanks. Moreover, the problems of moving up
timely supplies and reinforcements, especially in heavy rain and mud,
remained as yet insurmountable.11
In the meantime, after a three-day bombardment, on 15 September the
British and French launched the Battle of Flers-Courcelette. This, in Haig’s
mind, would be the climax of the campaign, an assault designed to capitalize
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d127

on all the hard-won progress of July and August. He hoped finally to break
through the last German defences and into open country. To shatter the
deadlock and return to mobile warfare would redeem all the frustrations
and casualties of the last two months. One tool to help smash a hole was, for
the first time in history, the tank. The story of the Battle of Flers-Courcelette
is often told primarily through the prism of tanks. In particular, the narrative
has tended to be dominated by debates about whether, in the first place,
their use was a premature mistake; and secondarily about whether doling
them out in twos and threes all along the line was the correct tactic, or they
might have been more effectively employed en masse. As we shall see, there
was more to the battle than tanks.12
At about eight a.m. on 15 September 1916 the front from Thiepval to
Combles burst into life as the British Reserve and Fourth Armies lurched
into action once more. Of the fifty tanks planned to participate, thirty-six
reached their starting positions, twenty-five got to the German front line,
and just six achieved their final objective. Inevitably, the British made mis-
takes with their handling of this new weapon. Fear of friendly fire, for
instance, led them to leave lanes in the barrage for the tanks to advance
down. Since many of the tanks failed to arrive, however, wide stretches of
the defences were left intact and able to inflict heavy loss on the assaulting
infantry. Harder to comprehend was the failure to find time for joint infantry–
tank training. Likewise, it is unclear why one general elected to ignore his
tank experts’ advice and sent four tanks into the unsuitable terrain of High
Wood. Nonetheless, where artillery preparation had been effective, the
creeping barrage worked well, and infantry followed up closely, sometimes
aided by tanks, good progress resulted. Within four hours, tank D-17,
‘Dinnaken’, led cheering British soldiers up the main street of Flers. ‘In
some places the appearance and fire of the tanks caused panic and terror’,
noted the German official historian, who also observed that the defenders’
resistance at Flers and Gueudecourt was as good as broken. Here, where the
British penetration was deepest, they had advanced some 2,000 metres.
High Wood, Martinpuich, and Courcelette also fell, and some 4,000 metres
of the German third position had been cleared.13
Rupprecht watched the fighting that afternoon from the upper window
of a building in the village of Bihucourt. At four p.m., he noted ‘the battle
was in full swing. The drumfire roared constantly like a powerful waterfall,
sometimes dying away a little but then growing once more in strength. . . . 
Judging by the smoke clouds, a counter attack to recapture the lost village
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

128 Holdi ng t h e Li n e 1916 –17

of Martinpuich was underway.’ At about the same time on the other side of
the line, it was becoming clear to Rawlinson that no decisive victory was to
be won that day. It would be pointless to commit further reserves without
moving guns up. He would have to prepare a fresh set-piece attack once
more from scratch. Again, over-ambitious objectives, beyond the range of
the guns, had been set. Just because Rawlinson was not ready to force
matters did not mean the battle was over, of course. Strong German counter-
attacks to regain lost ground were made throughout the afternoon and
evening. They failed. ‘Even by the standards of the Somme, 15 September
was an unusually intense day of combat’, according to the German official
historian. The majority of British battalions had lost 300–400 men each.14
In Rupprecht’s view, the fact that the defence around Flers had given way
in the first place, and the failure of the counter-attacks, could both largely
be blamed on exhaustion. The Bavarian troops holding the crucial sector
had been in the line for twenty-one days straight. They were slow to react
to defend Flers and some soldiers were so tired that they were falling asleep
even as they advanced across No Man’s Land under heavy fire. Basic errors
were repeated as a result. For example, Ginchy, which lay on the boundary
between two divisions, had fallen on 9 September because neither thought
to garrison the village properly. Less than a week later, Courcelette fell for
the same reason. ‘The whole staff of First Army, except for the Chief of Staff
Colonel [von] Loßberg, struck me as overworked and exhausted’, thought
Rupprecht. ‘The tension of recent weeks has brought some of the staff,
especially some of the over-strained Corps chiefs, close to breakdown.’15
Rupprecht also felt that the enemy had exploited a greater weight of
artillery fire. The Entente, he reckoned, fired four times as many shells as the
Germans. One thing he remained unimpressed by was British tactical skill:
The enemy brought up armoured vehicles for his successful attack against 3rd
Bavarian Division in the area of Martinpuich yesterday which drove straight
across country at our trenches and fired on them with machine guns and
small-calibre cannon. Under the protection of these armoured cars the British
infantry pushed forward in thick bunches. The attackers seemed drunk; they
staggered slowly forwards, rifles under their arms, regardless of the gaps our
infantry and machine-gun-fire tore in their ranks, which were always filled up
with men from the rear.

Although the French made little progress on the east side of the bulge on
15 September, and reserves soon came up to contain the British, nonetheless
the situation of First Army was far from ideal. First, its front line had been
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d129

stretched by another 8 kilometres. Second, it seemed that the Entente were

trying to drive a wedge between First and Second armies. The counter-
attack at Bouchavesnes designed to forestall this would not be ready for
days and indeed some of the reserves intended for it were needed instead
to plug gaps between Flers and Courcelette. Third, the loss of Flers left
German positions in Combles extremely vulnerable and if Combles fell,
the Bouchavesnes attack would be impossible. Lastly, the Canadian cap-
ture  of Courcelette left just one communications trench running up to
Mouquet Farm. German positions in the Thiepval salient were becoming
dangerously exposed.16
If the situation north of the Somme was critical, on the other side of the
river Gallwitz, too, was feeling the pressure. The French Tenth Army ground
forwards on the Santerre Plateau between 15 and 18 September, forcing the
Germans to pull their heavy guns back to the right bank of the Somme.
Gallwitz directed a stream of warnings of impending disaster to Rupprecht
and Kuhl at Army Group headquarters. When his demands for reinforce-
ments were ignored, he wrote a formal letter complaining that First Army
was receiving preferential treatment. Relations with the Army Group went
from chilly to frosty. When Rupprecht and Kuhl visited Second Army
headquarters in Saint Quentin, no-one was waiting to meet them, to hang
up their hats and cloaks, or to direct them to Gallwitz’s office. The subse-
quent interview between Gallwitz and the Crown Prince was animated.
More damaging was the fact that both Kuhl and Rupprecht suspected that
Second Army was, at best, not keeping them properly informed and, indeed,
possibly downright lying. When Rupprecht asked why he never saw copies
of Second Army orders, chief of staff Bronsart claimed that everything was
done by telephone so there were no written orders to see. Rupprecht
was  not convinced. Kuhl went further: ‘Reports are biased . . . Gallwitz is
embittered and makes continual difficulties. Given his intelligence, I wouldn’t
have considered him such a small man. Bronsart is dishonest and Faupel [the
operations officer] is also a weasel. A completely crooked crew.’ Eventually,
in late October, Bronsart was transferred elsewhere.17
While Haig and Foch prepared a new push, the Germans finally launched
their counter-attack at Bouchavesnes on 20 September. Two important
points came out of this particular phase of the battle. First, the balance of
air power was shifting from the Entente to Germany. Throughout July and
August, British and French airmen had possessed general air superiority.
They were able to operate almost at will even over German lines. The Germans
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

130 Holdi ng t h e Li n e 1916 –17

responded by reinforcing their air force on the Somme. New squadrons

arrived, armed with the latest and best designs. Consequently, the intensity
of air combat increased as the Germans began to contest control of the sky
more fiercely. On 20 September, when elements of six German divisions
attacked either side of Bouchavesnes, air support was ample and the
German fliers were largely able to go about their work undisturbed. Second,
Bouchavesnes reminds us graphically how difficult it was, even with air
reconnaissance, for those up the chain of command fully to understand
what was happening on the First World War battlefield. The first reports
Rupprecht received were of enemy artillery silenced by gas; a defence
caught off balance; and an attack carrying all before it. In short, he thought
he had scored a stunning success and he regretted not having set deeper
objectives. He encouraged Below to push on. Only later in the day did he
find out that progress had not been quite so dramatic and that the advance
had in fact been stopped on the western edge of Bouchavesnes. Stubborn
resistance and a strong French counter-attack were apparently to blame.
The next day, the picture of the situation was revised once more. The attack,
it was now clear, had been a failure. Bouchavesnes had never been in German
hands at all. The attackers had finished back where they started, in  their
own  trenches. The preliminary bombardment had been insufficient. The
attacking corps commanders had rushed the execution of the operation
because they had not built enough cover for the assault troops and they did
not want to leave them exposed out in the open for too long. ‘Or maybe’,
Rupprecht wondered, ‘our infantry simply possesses no offensive power
any more.’18
The attack unleashed by the British and French on 25 September showed
what could be achieved if the two allies could ever coordinate their attacks
properly. The British Fourth and French Sixth armies conducted a major
set-piece assault simultaneously and side by side for the first time since 1 July,
and seized the largest slice of German-held territory taken so far. Seven
British divisions attacked Morval and other German strongpoints, aiming
eastwards for Le Transloy and the new German fourth line. A further corps
drove north to protect the left flank. Fayolle, with seven French divisions,
attacked north-east into the German defences in villages such as Sailly-
Saillisel and Rancourt and the huge Bois de St Pierre Vaast. With luck, he
might thus turn the flank of the formidably strong Mont St Quentin and
threaten Péronne town from the north. Reserves of both infantry and cavalry
stood by in case resistance suddenly collapsed and deeper exploitation became
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d131

possible. A further linked operation the next day, by the British Reserve
Army, would try once more to seize the bluff at Thiepval.
On 25 September the British did well. They captured Morval and
Lesboeufs. The next day they occupied Gueudecourt and Combles. The
French had a tougher time. Rancourt fell quickly but, caught between
machine-gun fire from Bois de St Pierre Vaast and artillery dug in on Mont
St Quentin, it proved hard to make progress over the open ground and losses
were heavy. Local actions continued over the next couple of days, and nowhere
along the Somme in the autumn of 1916 was ever completely quiet; but the
Battle of Morval had died away by the morning of 29 September. Any
idea that the Germans might counter-attack faded away when OHL told
Rupprecht that reserves were running short and in future his Army Group
would have to rely more on its own resources. Units in line would just have
to accept longer intervals before relief if necessary. Rupprecht thought this
unhelpful advice. It would work only if the intensity of enemy attacks slack-
ened, but that seemed unlikely. In the event, OHL was forced to send five
divisions from its reserves to help him withstand the impact of the battle,
not only of Morval, but also of Thiepval Ridge.19
General Sir Hubert Gough attacked on 26 September with four divisions
from Courcelette to Thiepval. His objective was to clear the crest of the
ridge, prevent German observation of Albert, and enable the British to see
down into the Ancre valley. Fighting was heavy but by early the next day
both Mouquet Farm and Thiepval at long last lay in British hands. Poor
weather and the usual problems of coordination rendered further progress
slow, however, and heavy trench fighting continued in and around the
Schwaben Redoubt well into October.
The battles of Morval and Thiepval Ridge, taken together, won the British
and French the whole German line from Thiepval to Bouchavesnes, a front
25 kilometres long, to a depth of about 1,500 metres. Christopher Duffy has
described the battles as ‘the supreme effort of the British on the Somme,
and the one that brought the Germans closest to collapse’. Certainly, they
showed a British army which had learnt much over the previous three months.
The attack of Major-General Ivor Maxse’s 18th Division on Thiepval, for
instance, is evidence of two major tactical improvements. First, the assaulting
troops had learnt the importance of crossing No Man’s Land fast to avoid
the German protective barrage. Second, everyone now knew that the Germans
were likely to counter-attack. Lieutenant-Colonel Frank Maxwell VC dis-
tinguished himself by rapidly organizing the mixed-up units which had
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

132 Holdi ng t h e Li n e 1916 –17

stormed Thiepval village into an effective defence. As William Philpott has

argued, ‘by the end of September 1916 British junior officers and other ranks
were no longer naive or casual in the military arts’. Indeed, in the words of
Robin Prior and Trevor Wilson, ‘in this grim, intense, close-quarter fighting,
the troops from the British New Army and the Dominions were quite capable
of outfighting the German defenders’. If it had ever been true that the
German soldier was man-for-man superior to his British counterpart, this
edge was now eroding.20
Rupprecht noted and lamented this, too: ‘while in the past, although our
infantry were fewer than the enemy, they were qualitatively superior, as a
consequence of the heavy casualties, especially in officers and NCOs, this
qualitative edge has declined’. Manpower problems seemed to be getting
worse. The British were rotating divisions through the front line every six
days but it was becoming harder to organize timely reliefs for German
front-line troops. Eight days under enemy artillery was enough to wear out
most units, even if there was no infantry fighting. Rupprecht, however, was
having trouble getting Ludendorff to agree even to a fourteen-day relief
cycle. Further, morale was suffering as a result of fighting bloody battles for
land which in itself had no obvious value and which was ultimately lost.The
same was becoming clear the other side of the wire, too: British intelligence
officers ‘have never seen the moral[e] of the Germans so low as at present’.
First and Second Armies between them had lost 135,000 men. September
was by far the Germans’ bloodiest month on the Somme.21
Where the strain was even more marked, however, was in matériel. German
infantry as yet had no answer to the tank, although they soon began experi-
menting with trench mortars and 37 mm naval guns in an anti-tank role.
In the air, the German air force, as we have seen, was back on level terms by
the end of the month. While the British and French lost 123 aeroplanes
in September, the Germans lost just twenty-seven. Nevertheless, Germany
remained outnumbered by at least three to two in aircraft and could not
ease up. German gunners, noted Rupprecht, ‘were on a par in quality with
the British, but not with the French, and in terms of numbers they have not
been able to keep up with the ever rising number of enemy batteries’. The
crews were becoming overstretched and exhausted, in action night and day.
The field artillery fired over four million rounds in September, up from one
and a half million the previous month. Heavy howitzers expended 907,000
shells, compared with 643,000 in August. Inevitably, tired crews sometimes
performed poorly. Infantry–artillery cooperation remained less than ideal,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t ta k e s Com m a n d133

with protective barrages coming down too late and targets of opportunity
being missed. The guns themselves were wearing out and breaking down,
too. Half of the field guns and a third of the light howitzers had to be
replaced during the month.22
So far, Rupprecht and his men had clung on, but the strain was telling.
Would October be the month that saw the German army break?23
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

13
Autumn on the Somme

T he Somme was simultaneously slaughterhouse and war school, where

three armies struggled to adapt to the protean challenges they faced.
This can be seen most clearly in the to-and-fro between offensive and
defensive tactics. We have already seen, for instance, how by late September
the British Expeditionary Force (BEF) had lost its July naivety and began to
learn how to exploit the predictabilities of the German defence. Once the
attackers knew that the Germans would advance to recapture any ground
lost, they could plan to consolidate their gains and dig in, catch the counter-
attack in the open, and inflict further losses. For their part, the Germans
realized they must eliminate this predictability. September saw the incoming
senior commanders propose a new means of fighting the defensive battle,
spread their ideas, and begin to build consensus for change. By the begin-
ning of October, their new methods were starting to bite and have effects
across the battlefield.1
The key issue was how rigid the defence should be. Falkenhayn, as we
have seen, had always been firm that his men should not give up an inch of
French soil voluntarily, contesting the front line ferociously and counter-
attacking to regain any ground lost. His aim was to make every Entente
advance as expensive as possible, both politically and strategically. If France
and Belgium discovered that they could only liberate their homelands by
destroying them, so much the better. Operationally, he considered this the
neatest way to maintain the integrity of the defensive line and prevent a
chain reaction of units being outflanked and withdrawing in turn. Retreats,
he also feared, might damage morale.
Ludendorff had different ideas, however. He outlined them at the Cambrai
conference of 8 September. His suggested approach was much more elastic.
He wanted the front trenches manned only thinly, mainly with machine-
guns. They would function as an outpost line to identify and disrupt enemy
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Au tum n on t h e Som m e135

Figure 13.1.  Hindenburg with Ludendorff

thrusts. The weakened attack would then break on the stronger defences of
the second and third German lines. Light machine-guns and trench mortars
would substitute firepower for manpower. Artillery should not act purely
defensively but should aggressively target enemy concentrations of guns and
infantry. Now that the enemy possessed artillery superiority and had evolved
effective drills for mopping up any trenches he overran, fixed positions and
underground shelters had become liabilities rather than assets. It was too
easy for troops to be overrun in their dugouts: better for them to stay mobile.
Further, the defenders were no longer automatically to counter-attack.
Only significant terrain was worth losing more men for.2
Soon it became clear that Ludendorff was not thinking about applying
this new flexibility merely at the tactical level. He was considering using it
operationally, too. On 12 September Oberste Heeresleitung (OHL) men-
tioned the idea of building a new defensive line from Arras to Laon via Saint
Quentin to enable Rupprecht’s Army Group to shorten its front by 50 kilo-
metres and free up reserves. This position eventually became known to the
Germans as the Siegfried Stellung. The British and French called it the
Hindenburg Line. We will discuss it further in Chapter 14.3
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

136 Holdi ng t h e Li n e 1916 –17

Debate on the new elastic defence went on throughout the month.

Second Army orders issued on 17 September still stuck to the old principles
but a consensus was already emerging in favour of Ludendorff ’s ideas. The
ground was fertile since, as we saw in Chapter  10, front-line troops had
already begun to ignore Falkenhayn’s official doctrine when it conflicted
with practicalities on the ground. Several details of the new approach
needed filling out, however. Who was to control any local counter-attacks,
for instance? Could it be left to the initiative of company, platoon, and even
section commanders?
On 27 September Kuhl summed up what had been learnt over the previ-
ous few months. The situation was now so different from what had gone
before, he wrote, that ‘only experience on the Somme is important. The
lessons of earlier battles must be corrected from now on.’ Operationally,
everything was about holding on: ‘on the Somme, victory will go to who-
ever lasts longest in matériel, munitions and manpower. There is nothing
else for it. We will prevent an enemy breakthrough, if we provide for timely
reliefs.’ Tactically, he identified five areas for improvement. First, units should
be deployed in greater depth, so divisional frontages should be reduced from
4 kilometres to under 3. Second, better cooperation between artillery and
infantry was desirable. Once an enemy assault was underway, the guns
should concentrate solely on the attacking infantry. Third, more fighter air-
craft were required, more contact patrols should be flown, and spotter crews
needed better training. Fourth, staffs were being stretched too thin. Divisions,
in particular, needed to have a second properly trained staff officer. Corps
commanders who knew their sectors well might be left in place even when
their men had been relieved to avoid losing the benefit of their experience.
Fifth, defences should be positioned on reverse slopes and eschew deep
dugouts, and the front line should be only thinly garrisoned and not auto-
matically held to the last man.The front line could best be held by counter-
attacks coming from deep. Counter-attacks were more effective when they
were quickly improvised locally (Gegenstoße) rather than deliberate set-
piece operations (Gegenangriffe).4
Tactics of this kind had already contributed to the breakdown of French
attacks on 25 September along the Bapaume–Péronne road.5 They were
far  from the only factor sapping momentum from the Entente advance
throughout October and into November, however. With the passing of the
autumn equinox, the weather changed for the worse. Mist and rain grounded
spotter planes and so weakened artillery preparation.They also made objectives
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Au tum n on t h e Som m e137

harder to spot, especially in the featureless wasteland that was being fought
over. Consequently, barrages tended to fall either short and hurt the attack-
ers, or too far ahead and miss the defenders. The ground, stripped bare of
vegetation and pock-marked with shell holes, was soon churned to a liquid
yellow-grey mud.This stuck to everything and made walking a slow-motion
torment. Guns which could normally be pulled by six horses now required
teams of twelve, if they could be moved at all. Stable firing positions were
increasingly hard to find. Moving food and ammunition across country had
rarely been easy. Now in places it became all but impossible. Where there
were roads, they were worn out after months of heavy traffic and began to
break up. A lorry passed every four seconds along the main French supply
road east from Amiens: the traffic was twice as heavy as along the famous
voie sacrée at Verdun. Where Foch’s staff had originally estimated they
would need three steamrollers to keep his roads open, by November forty-
nine were at work.The British supply network, hastily cobbled together and
patched up over the previous two years, reached breaking point. Attempts
to imitate the French and build a light railway network looked promising,
but this solution came too late to solve the supply problems in heavy artil-
lery munitions, roadstone, and timber.6
In the face of these major challenges, at times the tempo of French and
British operations slackened, granting the Germans crucial breathing space.
Still, it would be a mistake to see the October battles as mere postscripts to
the ‘real’ fighting of July–September. The Entente retained considerable
punch yet. Indeed, Rupprecht considered that two days in October (9 and 12)
saw the hardest fighting the Germans had yet faced. Even as late as 5 November,
he ranked the battle among the most savage they had experienced.7
The pressure of the September fighting had taken its toll on Rupprecht
and when October opened with fresh British and French attacks, he became
rather depressed. Kuhl was forced to intervene. On 2 October, for instance,
Rupprecht told his chief of staff that they could not hold out like this for
much longer and it would be better to pull back into a new line of defence
to the rear. Kuhl responded that there was no proper fortified position ready
to allow such a move; that trying to break contact with the enemy in the
middle of heavy fighting was an invitation to disaster; and that, in any case,
even if they did withdraw, they would only find themselves in the same
predicament again within a few days. The prince was still quite down two
days later. It was clear to him that Foch and Haig were now taking an attri-
tional approach, launching small-scale attacks to keep up the pressure in
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

138 Holdi ng t h e Li n e 1916 –17

between large set-piece operations. He worried that ‘the large number of

prisoners we have recently lost is an evil sign’. ‘But he can straighten himself
up again’, added Kuhl.8
The British and French continued to nibble away at the German defences.
They were trying to establish good jumping-off positions for another major
assault. Rawlinson’s Fourth Army’s objective remained Bapaume, while
Gough’s Reserve Army was still working to capture the crest of the Thiepval
ridge. With Mont St Quentin firmly in German hands, French Sixth Army
could drive only north-eastwards, on Sailly-Saillisel and Bois St Pierre Vaast.
South of the Somme, meanwhile, high ground around Lihons and Chaulnes
continued to block the French advance eastwards.
Entente plans for a joint attack had to be postponed when heavy rain set
in, and fighting degenerated into small-scale pushing and shoving very rem-
iniscent of the August struggles. An attempt to coordinate a large Anglo-
French operation on 7 October delivered only disappointing results for the
Entente and attempts to get the offensive moving again the next day yielded
little. In a vivid illustration of the difficulty of communication on the 1916
battlefield, Major Prager of Rupprecht’s staff happened to be watching the
action from an artillery observation post. The forward observer told him
that it would take at least an hour and a half to pass any message to the guns
defending Le Sars. Rupprecht and his men could feel some satisfaction,
nonetheless, that they had beaten off the ‘strongest attacks to date’ on the
Somme front. The weeks of fighting which now began in the twin villages
of Sailly-Saillisel, house by house and cellar by cellar, were characterized by
a ferocity fit for Stalingrad. It was, according to William Philpott, ‘the worst
fighting that [the French army] was to see in the whole Somme battle’.9
The Germans were suffering, too. 12 October in fact marked another
very heavy day of fighting. Five British divisions of Fourth Army went over
the top but, encountering an alert and ready defence with machine-guns
deployed and concealed in considerable depth, often out of range of the
British creeping barrage, they achieved little of value. Nor did further set-
piece offensives on 18 and 23 October. Fourth Army held a conference on
13 October to discuss improvements in tactics, but the weather and terrain
prevented implementation of any recommendations it made. The same day,
men of the German 15th Infantry Division mutinied and refused to enter
the line. Their corps commander claimed never to have seen such poor
troops. The problem was not limited to the full-scale attacks which make
the history books; nor yet the innumerable little patrols and raids which do
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Au tum n on t h e Som m e139

not. Rather, reported First Army, they faced a more relentless opponent: ‘our
worst enemy unfortunately remains the enemy superiority in ammunition,
especially in heavy calibres. These knock down our defences, cause most of
the casualties, disrupt traffic in the rear areas and deprive troops who are
supposed to be resting of true quiet.’10
On the other side, the Germans had rotated their forces, shifting out
exhausted divisions and artillery batteries and bringing in fresh ones. The
extensively fortified Butte de Warlencourt, a prehistoric burial mound
20 metres high, dominated the centre of the front.The British attacked it again
and again throughout October and November, each time without success.
As Gary Sheffield has argued, ‘like Monte Cassino in a later war, the Butte
exercised a powerful and malevolent grip on the minds of the British
­soldiers crouched in its shadow’. Unlike Cassino, it never did fall. Despite
the impossible mud and painful evidence that the offensive had stalled, Haig
insisted on a final British attack on 5 November. The XIV Corps com-
mander, Lord Cavan, lodged a rare formal protest, but the operation went
ahead nonetheless. It captured but a little ground of limited significance.11
In the centre, therefore, the British and French were stuck firm. On the
Santerre Plateau on the left bank of the River Somme, however, the attackers
had greater freedom of movement and resupply and reinforcement were
slightly easier than in the devastated centre but even here, after some pro-
gress in heavy fighting on 6 and 10 October, by the second half of the
month French momentum was fading. Indeed, by the end of October the
situation of the German army on the Somme felt much more comfortable
than it had only a month earlier. German casualties in October dropped to
85,000 from 135,000 in September. Fresh divisions had been committed to
relieve those worn out by the fighting and a proper system of roulement was
in place. As the tempo of operations dribbled away, divisions were beginning
to endure in line for longer than the fourteen-day target. One Bavarian
division managed to last three weeks before it was relieved. A French offen-
sive at Verdun on 24 October, although successful in recapturing Forts
Douaumont and Vaux, could not be sustained for long enough to disrupt
the German reliefs rota. In any case, the Somme was supposed to be relieving
pressure at Verdun, rather than vice versa.
Stronger artillery increased German confidence. On the main battlefront
there were now four times as many heavy artillery batteries as there had
been on 1 July. A new Army Artillery Reserve had been established to
replace worn-out guns more promptly. Infantry and artillery were liaising
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

140 Holdi ng t h e Li n e 1916 –17

more effectively and increased ammunition supply enabled deeper defen-

sive barrages to be fired. More ammunition was being manufactured and
supplied and it was being distributed more efficiently. OHL delegated to
Rupprecht the power to decide where to send ammunition, so he could
direct shells where they were needed in a more responsive and timely
­fashion. In the air, the Germans were still outnumbered, although not by so
much.They had, however, mastered new machines and famous units such as
Oswald Boelcke’s Jagdstaffel ( Jasta) 2 led the way in developing new tactics
to fly and fight in large formations, rather than as individual machines. The
Entente had lost air superiority. Although Boelcke himself was killed on
28 October, by then his most famous pupil, Manfred von Richtofen, had
already begun his lethal career.12
On the British left flank, Gough’s Reserve Army was rechristened Fifth
Army at the end of October. The name changed but the mission stayed the
same. This remained the elimination of the Thiepval Salient. In what was
later called the Battle of the Ancre Heights, Gough’s men spent October
and early November launching a series of limited-objective assaults designed
to nibble away at the German hold on Thiepval ridge. At last by 21 October
they had cleared the Stuff and Schwaben Redoubts and reached the crest of
the ridge. It had taken a month to advance 750 metres. German intelligence
had been monitoring the build-up of British strength here closely for some
time, fearing a more ambitious scheme was being prepared. Both Rupprecht
and Ludendorff suggested evacuating the salient, but Below was reluctant to
give up the good observation posts he held there. He felt that his defences
were strong and he should hold there as long as possible.13
Finally, on 13 November, after a week-long bombardment, the British
attacked with five divisions. The focus was on the far side of the River
Ancre, around the villages of Serre and Beaumont Hamel, where the BEF
had suffered so badly on the morning of 1 July. This time the German
wire and fixed defences were systematically destroyed. A creeping barrage,
­incorporating machine-gun fire, covered the infantry advance closely. At least,
that was the case where the attack went well. Inevitably, that did not happen
everywhere. Tanks were unable to have much impact: the ground was too
boggy. Beaumont Hamel and St Pierre Divion fell to the British, but other
objectives of the original attack, such as Serre, remained firmly in German
hands. Even another follow-up attack on 18 November could not prise them
free. Casualties were heavy: battalions of the 51st (Highland) Division lost up
to half their strength. Nevertheless, the Battle of the Ancre (13–19 November)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Au tum n on t h e Som m e141

was considered successful enough that Haig could, at last, declare victory
and formally suspend the offensive.14
On the other flank, by 12 November the whole of Saillisel was finally in
French hands. It was, however, evident that operations were winding down
at last. The first snowfall on 18 November gave a final sign that winter was
at hand and that the time for major operations was past for another year.
By that time, Rupprecht had already decided he could finally get away on
leave for the first time in over two years. He left for Munich on the evening
of 16 November.
On the face of it, the German army had done a good job of frustrating
the Entente armies in the Battle of the Somme. After 141 days of fierce
fighting, it had allowed the French and British to advance about 10 kilometres
at most. Two of the attackers’ main geographic objectives, the towns of
Bapaume and Péronne, both remained firmly in German hands. Casualty
statistics are not straightforward for the Somme. British casualties probably
totalled 419,654, with the French losing half as many: 204,253 men. Most
estimates for the Germans range from a conservative 429,209 to a frankly
implausible 680,000. About 500,000 seems roughly right.15
With casualties so high, and results apparently so meagre, was the Entente
right to fight and then persevere with the battle? Debate continues about
this, so, before we leave the Somme, we might briefly review some of the
arguments and explore what light Rupprecht’s evidence sheds on them.
Sir Douglas Haig was quick to justify the Battle of the Somme in his official
despatch of December 1916. No-one was more aware than he that his early
hopes for a breakthrough and a return to mobile warfare had been dashed.
So Haig offered not just one, but three objectives which he argued the
campaign had achieved. These were: first, to relieve Verdun; second, to tie
down the German army in the west and prevent it being transferred to fight
in the east; and, third, to wear down the enemy. ‘Any one of these three is
in  itself sufficient to justify the outcome of the Somme battle’, he said.
‘[Together] they have brought us a long step forward towards the final
­victory of the Allied cause.’ How do these three justifications stack up?16
The weakest of the three is Verdun. While it is true that the Germans
suspended active offensive operations at Verdun on 12 July, it had been obvi-
ous to everyone for months that they were going nowhere and would have
to give up anyway. Rupprecht had privately written it off in his diary as
early as 20 March. Long before 1 July it was clear that the British were not
going to do what Falkenhayn had so painstakingly planned for them.When
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

142 Holdi ng t h e Li n e 1916 –17

the attack on the Somme came, and proved much heavier than Falkenhayn
had anticipated, this served as final proof not only that there was still life in
the French army but also that Haig was not going to play neatly into
Falkenhayn’s hands. The Verdun business had long been bankrupt: the
­decision of 12 July merely shut down the shop. In any case, if the Somme
were seriously intended as an offensive to relieve Verdun, it was hardly
a very  timely one. Over four months had passed since the fighting at
Verdun began.17
Did the Somme prevent the Germans from transferring troops from the
Western Front to other theatres? No: possibly as many as fifteen divisions
were transferred from west to east during the campaign, although nine of
them were exhausted and in any case some six fresh divisions went back the
other way in exchange. If we step back from the statistics, however, the
­larger picture is equivocal. On one hand, the Central Powers remained
­capable throughout 1916 of fighting off multiple opponents on widely
­separated fronts. Indeed, from August onwards, Germany could take on the
new commitment of Romania and crush her army in just four months.
On the other hand, though, the Entente strategy of concentric offensives
was actually beginning to operate rather well. OHL sent divisions east
because it had to. Germany was having to juggle multiple threats and begin-
ning to find doing so a considerable strain. We will never know how much
strength Germany could have moved east if the Somme, or some other very
similar campaign, had not been fought, but politically Haig and Foch had
little choice if they were not to let down their allies.18
If the verdict on the second justification seems evenly balanced, the third
appears more clear-cut. Whether attrition was always in Haig’s mind as a
back-up plan or put forward as a rationalization after the fact, the Somme
certainly wore down the German army. Every time a German division
spent a tour in line on the Somme it lost, on average, about 3,000 men or
about a third of its infantry. Losses on this scale inevitably depressed morale
and the quality of the troops suffered. In October Rupprecht wrote to his
father, King Ludwig III, that ‘sadly it’s becoming ever clearer that our troops
lack attacking power since almost all our attacks, even the limited ones, fail.
The troops are not what they were in the spring.’ Alexander Watson has
demonstrated the psychological damage suffered by the German army. This
comes through strongly in Rupprecht’s papers. The Somme did not break
the back of the German army, but it weakened it considerably.To expect any
battle at this stage of the war to do much more is probably unrealistic.19
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Au tum n on t h e Som m e143

Some have argued that the Somme was a necessary learning step for the
Entente forces. The tactical lessons the BEF learnt there, or the refinements
Foch made to his operational method, contributed to final victory in 1918.
We shall explore the changes made over the winter of 1916/17 in more
detail in Chapter 14, but this is true, in a sense. The only way to learn how
to fight is to do it. They were horribly expensive lessons, however. Worse,
Rupprecht’s view of how British combat effectiveness developed is mixed.
Throughout the campaign he saw them as dangerously brave.They were less
proficient and menacing opponents than the French, however. They often
relied on sheer weight of numbers, of artillery and men, to blunder for-
wards. As the campaign developed, he accepted that the qualitative gap
between German and British infantry was closing, although he attributed
that more to his men getting worse than to the British getting better. By
November 1916, he actually noted signs of British tactics becoming clum-
sier once more, with a return to the old-fashioned close formations which
had proved so vulnerable earlier in the war.20
Finally, some historians have argued that the Germans became so keen to
avoid a repeat of the Somme that they decided to pull back into the
Hindenburg Line in spring 1917 and wait for the U-boats to win the war
instead. In fact, there was no strategic master plan. The Hindenburg Line
was originally constructed as a contingency measure, and the decision to
retreat into it was taken only shortly before implementation. The German
army was under no illusions that the retreat would prevent another Somme-
style offensive from the enemy: it would just delay one for a few weeks and
give the defenders a better chance of withstanding it, when it came.We shall
discuss this decision further in Chapter 14.21
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

14
Scorched Earth

B y the end of 1916 it was at last clear to all that this war could end only
when one side or the other toppled into the abyss. As Hindenburg
wrote to the Chancellor that autumn, the outcome ‘would decide the
­survival or destruction of the German people’. Only if Germany mobilized
all her resources, he argued, could she prevail. Victory required a radical and
ruthless remobilization of society and industry to feed the military’s needs.
Such was Germany’s relative weakness in manpower, raw materials, industry,
and transport that a land offensive was unthinkable. Some, Rupprecht and
Kuhl among them, went further: they were beginning to doubt whether a
military way out of the nation’s difficulties existed. They kept their doubts
to themselves, however, for now.1
Hindenburg and Ludendorff remained optimistic. The military establish-
ment was not prepared to accept that victory was beyond reach. Oberste
Heeresleitung (OHL) was agitating for a renewal of unrestricted submarine
warfare. Earlier in the war Rupprecht had opposed this, not on legal or
moral grounds but because he was concerned that the risk of p­ rovoking
the United States outweighed the benefits the relatively small U-boat fleet
might achieve. Now that more U-boats were available, however, and with
American entry seeming more likely in any event, his attitude changed.
OHL also recast the army to increase effectiveness. Over the ­winter of
1916/17 Hindenburg and Ludendorff built thirty-three new d­ ivisions, from
a mix of fresh recruits and existing units, bringing the total to 232. The
cavalry was reformed and infantry establishments were standardized. In
theory, each division now had three infantry regiments, each of three bat-
talions; fifty-four heavy and 108 light machine-guns; and thirty-six field
artillery pieces. The distinction between active, reserve, Landwehr,
and  Ersatz formations was abolished. In practice, these ideals were not
always achieved. Nonetheless, in total, the army deployed about 9,300 field,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h145

and nearly 5,000 heavy, artillery guns, together with 2,271 aeroplanes and
150 balloons.2
The German army also had to reconsider its battlefield approach. As we
saw, it had started the Battle of the Somme somewhat complacent. 1915 had
shown that it was perfectly capable of containing Entente attacks. Along
much of the British Expeditionary Force’s (BEF’s) front on 1 July, this com-
placency proved justified. Defensive firepower had shredded the British
ranks. Any break-ins the British managed were cut off by artillery and soon
pinched out.Where the British and French had scored successes, the weight
of their artillery seemed to have made the difference. The French, in par-
ticular, had concentrated their guns and systematically destroyed German
strongpoints one by one. Next, a thick and effective barrage kept the
defenders in their dugouts while the assault troops overran the shattered
trenches and seized their objectives. As fighting on the Somme continued it
became ever clearer that this was not a battle of genius and manoeuvre, but
of matériel and attrition. The Germans called it Materialschlacht.
This called for new tactics. Evidently, troops in the front line often
achieved least and suffered most. In response, several corps and divisional
commanders began to downgrade the forward crust defence and hold more
of their strength rearward, whatever Falkenhayn’s orders said, as we saw in
Chapter  10. Other changes followed, too. To improve air support, for
instance, each corps was given a specialist air commander in the middle of
August. Further, First Army experimented with delegating control of all bar
the very heaviest artillery batteries right down to divisional level. Response
times improved as a result.3
These lessons were among those digested by the new team: Rupprecht,
Kuhl, Ludendorff, and Hindenburg. Their first priority was to increase the
flow of food, ammunition, and matériel to the troops. Once they had taken
steps to improve that, they began to re-examine everything else with fresh
eyes. In this case, it is possible to follow the learning process closely. They
discussed new tactics at Cambrai on 8 September and began to codify their
ideas in a circular of 25 September. Kuhl developed them further in his
memo of 27 September, which we discussed in detail in Chapter 13. OHL
then called in after-action reports from headquarters at every level from
regiment to Army Group that had fought on the Somme. The archives in
Munich preserve at least sixty-eight of these reports which, together with
lessons gathered from the defeats suffered at French hands at Verdun in the
autumn, provided the raw material from which OHL pulled together a
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

146 Holdi ng t h e Li n e 1916 –17

major revision of defensive doctrine over the winter of 1916/17. A team

within the Operations department worked on a slew of memoranda cover-
ing topics such as the construction of trench systems, the use of mortars, and
effective liaison between spotter ’planes and their artillery.4
These documents were together known as ‘Collected Instructions for
Trench Warfare’. The keystone Part  8, ‘Principles for Command of the
Defensive Battle in Trench Warfare’, was released on 1 December 1916. With
attrition now beginning to bite, the objective ceased being to hold every
inch of ground regardless of its value. Instead it became to cause the enemy
maximum losses while minimizing one’s own. Four core ideas were to work
together to achieve this. First, the defender must remain active and not
allow the enemy the initiative. Instead, he must keep the other side off bal-
ance by aggressive patrols, raids, and local attacks. Second, the defence must
be built around machines rather than manpower.Third, ground must not be
held unthinkingly, regardless of its tactical value. It was acceptable to pull
back a little if necessary in the face of a major enemy attack. Any ground
yielded should be immediately counter-attacked and regained. If that proved
impossible, however, and only a full-scale set-piece operation seemed likely
to regain it, the importance of the terrain should be weighed up before any
decision to proceed. Counter-attacks were resource-intensive and should
no longer constitute an automatic response. Lastly, the defence should be
deployed in depth to drain the attacker’s momentum and leave him vulner-
able to counter-attack. Under the new regulations the distance between
trench lines in each defensive position was to extend from 50–100 metres to
150–300. Similarly, the depth between first and second positions was
increased from a kilometre or 2 up to somewhere between 4 and 10. Some
debate ensued about who should control any counter-attacking reserves.
Should they stay under the control of their own division or be commanded
by the man leading the front-line garrison? The latter, it was eventually
decided.The same commander also won control of all the heavy guns in his
sector. To increase the elasticity of the defence, the commander on the spot
was to be given greater scope to use his initiative. He would only be permit-
ted to do so in two very specific circumstances, however: he might choose
whether to pull back temporarily if under heavy attack and whether to
recapture lost ground. Otherwise, OHL reserved decision-making power to
itself. This was, therefore, far from a general delegation of authority.5
The aim of codifying doctrine in this manner, in one set of documents,
was to increase uniformity of practice. Previously, each individual army had
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h147

kept separate files on the methods they expected their formations to follow.
Now the ‘Collected Instructions for Trench Warfare’ offered a single ready
reference in case of need, a foundation for all subsequent debate and adapta-
tion, and a benchmark against which practice could be checked. Thus, the
‘Principles for Command of the Defensive Battle in Trench Warfare’ docu-
ment, for instance, was revised five times to incorporate new lessons before
the war was over.When First Army issued a report nearly seventy pages long
on the lessons of the Somme, Rupprecht told his officers to read it in con-
junction with the ‘Defensive Battle’ document. Comparing the two would
throw into relief differences in approach between the two, particularly as
regards elastic defence. Kuhl took care to highlight and explain the changes
made in a subsequent edition. More practically still, in January 1917 OHL
ordered subordinate headquarters to present detailed schemes of defence,
following the new regulations. Staff officers scrutinized the maps and went
through the calculations to check compliance.6
In November 1916 Hindenburg issued another significant memorandum,
‘Command in War and the General Staff ’. In it, he warned that Germany’s
enemies were starting to surpass her in some respects as her pre-war edge in
training, discipline, and drill was eroded. It was noticeable, he complained,
‘that we have in some respects learned more slowly than our enemy’. The
solution lay in improved training. Rupprecht picked up on this, establishing
a school at Solesmes to train divisional commanders and staff officers.
The first course began in February 1917. The new ‘Collected Instructions
for Trench Warfare’ were at the centre of the curriculum. Demand was so
strong that Rupprecht soon set up a second school and in March Crown
Prince Wilhelm established one at Sedan for his Army Group. This provi-
sion of new training for senior officers built on attempts to improve that of
officers and non-commissioned officers (NCOs) at company and platoon
level. These had been underway since October 1916 with the establishment
of ‘Field War Schools’ in each Army Group. These lasted five to six weeks
and trained 100 officers and 100 NCOs each time. Specialist schools, for
instance for stormtroop tactics and artillery techniques, were also expanded
and improved. In addition, each division was to spend three weeks of
uninterrupted training over the winter of 1916/17. All in all, the Somme
proved ‘the high school of the German army for the defensive battle in
­position warfare’.7
All these changes would, it was hoped, improve the quality of the army;
but quantity remained a problem. At best, Ludendorff and Hindenburg
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

148 Holdi ng t h e Li n e 1916 –17

Figure 14.1.  Rupprecht inspecting his troops

would be able to count on a reserve of twenty divisions in the west in

spring 1917. The BEF alone would have as many and the French another
seventy-two. A meeting of army chiefs of staff on 2 January 1917 was almost
unanimous that a major breakthrough operation was out of the question.
Offensively, the best that could be achieved was a series of limited-objective
local assaults. These might hurt the enemy but could not defeat him. There
was no alternative to assuming the strategic defensive and ceding the initia-
tive to the Entente. Kuhl put together a detailed memorandum along these
lines. The question now, therefore, became how best to meet the enemy
onslaught intelligence was predicting for the beginning of March.8
Two possibilities presented themselves. Either the army could stand and
fight where it was, much as it had throughout the previous two years. Or it
could trade space for time and pull back some 30 kilometres into the
­position known to the Germans as the Siegfried Stellung and to the British
as the Hindenburg Line. This was not a new idea. OHL had already been
thinking about it since early September 1916. They calculated that a retreat
to the Arras–Laon line would shorten the front by some 50 kilometres and
free up about ten divisions. Construction began in autumn 1916.
Ideas about how best to use the Hindenburg Line evolved over time. To
begin with, and indeed well into 1917, it was seen as no more than a safety
net which might come in handy if enemy pressure, especially in the Somme
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h149

sector, became intolerable. When morale at OHL, which had been high
after the capture of Bucharest and defensive successes on the Somme, was
knocked by General Charles Mangin’s victory at Verdun on 15 December,
orders went out to accelerate construction. Still, however, the Hindenburg
Line seems to have been viewed primarily as a contingency measure.9
In the course of January, views changed. To see the decision to retreat as
purely driven by a negative desire to avoid a renewal of the Battle of the
Somme is to over-simplify a more complex process. In fact, it was the result
of a combination of positive and negative considerations. First, the Kaiser’s
decision in favour of a resumption of unrestricted submarine warfare on
8 January 1917 seemed to open another possible path to victory. If U-boats
could throttle Britain and force her out of the war, all that was necessary on
land was to avoid defeat. Moving back into a strong defensive position could
obviously help. Ludendorff told Kuhl that once they had defeated the
­enemy’s next attack, the war would end in the summer. Neither Kuhl nor
Rupprecht was quite so optimistic, but they disagreed more with the timing
than the logic.10
The second factor was more negative. First Army had originally argued
against withdrawal. Below believed it would damage the morale of the men
who had fought so hard to hold the line on the Somme. More junior com-
manders, however, began to press for a retreat. Their troops, they suggested,
could not withstand another battle like the Somme. One First Army corps
commander, General Georg Fuchs, made a particular impression. He warned
Kuhl that the morale of his men was vulnerable and the condition of their
defences lamentable.The front line near the Ancre, for instance, consisted of
little more than shell holes.The dugouts were flooded and communications
trenches unusable. Better, he argued, to pull back. Convinced, Kuhl s­ uggested
withdrawing into the Hindenburg Line as soon as possible. Discussions
went back and forth between Army Group Crown Prince Rupprecht and
OHL. Finally, on 4 February 1917, the Kaiser signed off on Operation
ALBERICH, the code-name for the retreat into the Hindenburg Line and
the devastation of the evacuated zone. The commanders of the armies in
line thus found themselves overruled by a coalition of their subordinates
and seniors.
OHL hoped that pulling back into the Hindenburg Line would advance
four objectives. It would free up reserves by shortening the line.The strength
of the position itself would save further manpower by enabling each ­division
to hold a broader sector. The Hindenburg Line was based on multiple
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

150 Holdi ng t h e Li n e 1916 –17

defensive positions, echeloned in depth and wherever possible sited on

reverse slopes to prevent enemy observation. Thick belts of barbed wire
offered further protection. Large numbers of machine-gun strongpoints
dotted the defensive zone. Next, withdrawing would buy time. Before the
enemy was able to attack, he would have to establish new lines of commu-
nication forward to the new front line. Then he would have to make all the
time-consuming preparations for a new offensive again, this time from
scratch and on new ground. The evacuated area would be harrowed, with
villages burnt, bridges blown, and rails and roads torn up, to make the
­enemy’s job harder. German and Russian retreats in Poland earlier in the war
had demonstrated how effective a scorched-earth policy could be. Third,
every day bought on the Western Front might give the U-boats another
twenty-four hours to throttle Great Britain. Lastly, if the enemy did eventu-
ally attack, the German army would be more likely to beat him off.11
Rupprecht’s attempts to exercise direct influence over decisions made
regarding the Hindenburg Line enjoyed little success. First, when the whole
idea had originally come up, Rupprecht had proposed a different course for
it, retreating further and incorporating cities, such as Lille and Cambrai,
which the Entente would be unwilling to attack and destroy, into the
defences. OHL rejected his suggestion on the grounds that it would require
too much manpower. Second, Rupprecht objected to orders to deport
local inhabitants, raze their homes, destroy their livelihoods, and demolish
the road and rail networks. Military necessity required only that fields of fire
in front of the new defences be cleared and he was concerned about both
the humanitarian issues involved and possible damage to Germany’s reputa-
tion. He was also concerned that discipline would suffer. At first, his protests
seem to have remained fairly quiet, largely because he considered the
scheme impractical. He could not believe that either the labour or explo-
sives would be available to carry out these orders. In the event, when
Operation ALBERICH was put into motion on 9 February 1917, within
three weeks 126,000 civilians deemed capable of working in the Reich were
shipped away without trouble. Some 12,000 children, mothers, and old
­people, ‘useless mouths’, were left in towns such as Noyon to be liberated—
and fed—by the enemy. Trainloads of useful machinery were also removed.
Only when the final demolitions were scheduled to start on 16 March and
it was becoming clearer that Ludendorff was in earnest about scorching the
French earth did Rupprecht contemplate resigning in protest. Fearful, how-
ever, that such a move would suggest disunity between Bavaria and Prussia,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h151

he signed the order. He sadly noted in his diary that he had watched the
villages burning as he drove back to his headquarters.12
Retreating in the face of an alert enemy is always a risky operation and
Rupprecht had to coordinate the movements of three different armies.
Security was extremely tight.The OHL Intelligence Staff made up ­elaborate
deception plans in the press and leaked dummy operations orders. German
aeroplanes repeatedly bombed the railway stations of Amiens and other
communications nodes in an effort to slow any enemy response. Although
preliminary limited withdrawals on the Somme front had gone well, nerves
grew tense as 16 March approached. In fact, good staffwork contributed to
a well-organized move. The British official historian Cyril Falls gave
Rupprecht and Kuhl generous credit for that. Patrols stayed behind to create
the illusion of normality while the main forces slipped away d­ uring the
night.The Germans leapfrogged back through a series of successive r­ earguard
positions until they were safely in the Hindenburg Line on 19 March.13
Enemy interference was minimal. Both the British and the French
seemed slow to detect German movement, slower to grasp what was under-
way, and slower still to follow up.‘It’s really amazing, how inactive the enemy
is being’, noted Rupprecht. ‘He still seems to have noticed only small parts
of our withdrawal and has not yet grasped the big picture. The French are
even more in the dark than the English.’ Entente troops and staffs had
become attuned to the rhythms of trench warfare and were perhaps not as
alert as they might have been. Also, British intelligence had been caught off
balance. Jim Beach has demonstrated that this marked a failure of both col-
lection and analysis. Spy networks in the German rear areas were unable to
provide sufficient information for a reliable picture to emerge and, deep
down, the British simply could not believe that the Germans would will-
ingly yield something they had fought so stubbornly to keep for two years.
A third factor was that Sir Douglas Haig was wary of over-extending his
forces and, while ordering his men to take any opportunities which pre-
sented themselves to inflict loss, warned against large-scale attacks in force
unless fully supported by artillery. Such support was hard to provide in
mobile warfare over broken ground. Fourth, conducting a pursuit over what
Falls described as ‘a zone of morasses, the joint product of shell-fire and
rain’ was undeniably difficult. Finally, both Haig and the French
­commander-in-chief, General Robert Nivelle, were focusing on their own
impending attacks, at Arras and along the Aisne respectively, rather than on
the space between the two.14
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

152 Holdi ng t h e Li n e 1916 –17

According to the German official historian, ALBERICH boosted, rather

than depressed, morale: it proved to the rank and file that their leaders knew
their business. There were two further positive outcomes from the retreat,
he argued. First, Entente offensive planning was gravely disrupted: the
whole area between Arras and Soissons became a no-go area for months.
The British official history corroborates this view. Second, the German
army now found itself in powerful defensive positions in which it would be
easier to withstand Entente attacks. He suggested that the German defensive
victories which followed, at Arras, on the Aisne, and in Champagne, owed
much to the withdrawal to the Hindenburg Line.We shall examine the first
of these battles in Chapter 15. It is worth noting first, however, that little
time was in fact bought. Within three weeks of the German withdrawal,
Rupprecht’s Army Group found itself entangled once more in a full-scale
attritional battle which in many ways resembled the Somme.15
Once the retreat to the Hindenburg Line was successfully completed,
OHL revised its estimate of enemy intentions. In the west 146 German
divisions now faced 183 enemy ones. The deployment of Entente reserves
seemed to suggest two, possibly connected, operations: a British effort
around Vimy and Arras and a major French push near Reims. From late
February, traffic analysis by signals intelligence suggested enemy reinforce-
ments were arriving at Arras, opposite Falkenhausen’s Sixth Army. German
intelligence soon picked up on work to improve road and rail links in the
area, as well as more air activity. Rupprecht secured reinforcements of artil-
lery, aircraft, and infantry for Sixth Army. By 22 March Falkenhausen
appeared to have all he needed to withstand a surprise attack. As the weight
of enemy artillery fire grew from 26 March, any remaining doubts evapor-
ated and within days Rupprecht was formally warning OHL to expect a
major offensive between Lens and Arras. In early April OHL gave him
authority over four divisions held in reserve behind Sixth Army.
On 4 April, a major bombardment began east and south of Arras. Overall,
320,000 rounds fell on Sixth Army between 29 March and 5 April, more
than double the amount fired the previous week. It was evident that
an attack could be expected in days, rather than weeks, but, even though
Sixth Army was outnumbered twelve divisions to nineteen, there seemed
little cause for real concern. An OHL liaison officer sent to inspect the
defensive preparations went away satisfied that Sixth Army was ready to
repel ­anything the British might manage to launch. The German forces
defending the Souchez and Vimy sectors, he noted, were especially well
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h153

p­ repared. Falkenhausen was keeping five reserve divisions well in the rear,
out of British artillery range, where they could rest and train undisturbed.
On 6 April both Ludendorff and Rupprecht ordered Sixth Army to move
these reserves up closer to the front line at once. The threat was now immi-
nent and grave.16
Before we explore the Entente offensives, we should briefly outline how
the British and French armies reacted to the lessons of the Somme and
hoped to change the way they fought in the new year. It was easy enough
to accept the need for change after a year which left neither army happy
with results. It was harder, however, to choose which lessons to learn.17
The French army found sifting the lessons of 1916 particularly difficult. It
was true that the methodical, artillery-dominated,‘scientific battle’ employed
on the Somme had enjoyed considerable tactical success and reduced French
casualties. An average infantry company which might have lost seven men
killed for every day of combat in 1914, and five in 1915, in 1916 lost just one.
However, the need always to follow the plan had prevented exploitation of
success. Worse, as an operational approach it clearly was not getting the job
done. It was hard to see how a method which had failed to liberate Péronne,
just 10 kilometres away, in four months could free France. Meanwhile, down
at Verdun General Robert Nivelle had secured impressive results on 24
October and 15 December with a very different approach. In some ways
returning to the tactics of 1915, French troops swept forwards in a single,
brusque onslaught, well supported by artillery and aiming for deep object-
ives including the enemy’s gun lines. In December 1916 Joffre signed off on
a 160-page ­memorandum on offensive methods for the following year. This
followed Nivelle’s method more closely than Foch’s.When Nivelle took over
as commander-in-chief on 17 December,‘scientific battle’ became discredited
and its practitioners, such as Foch and Fayolle, were sidelined. As we shall see
in Chapter 15, the new methods yielded only renewed disappointment. Major
reform of training and doctrine formulation would have to await the arrival
of Pétain as French commander-in-chief in the summer of 1917.18
The British on the Somme were starting from a much lower base than
the French. According to a German after-action report, in July 1916 British
infantry remained less dangerous than the French, even though they had
improved since Loos. The troops were not as well trained and they were
unsure of themselves tactically. British artillery fire was better directed than
before, but it still lacked the effectiveness of the French. We have already
seen that the BEF improvised and experimented frantically throughout the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

154 Holdi ng t h e Li n e 1916 –17

battle and that GHQ collated the most important lessons and distributed
them, especially via the doctrine manuals printed by the Army Printing and
Stationery Service (SS). Over the winter of 1916/17, however, GHQ under-
took a major revision of the BEF’s manuals.19
In February 1917 the new Director of Training, Arthur Solly-Flood,
wrote and published two new tactical manuals: SS 143, ‘Instructions for
the  Training of Platoons for Offensive Action’ and SS 144, ‘The Normal
Formation for the Attack’. These introduced two main principles. First, the
infantry platoon now replaced the company as the basic tactical unit.
Second, Solly-Flood tried to integrate new technology better. The platoon
was reorganized. Rather than the old four rifle sections, each platoon now
deployed specialist bombing, rifle, Lewis gun, and rifle grenade sections,
along similar lines to the French model. The platoon should form an ‘army
in miniature’ with its own integrated firepower and be able to fire and move
independently. Moreover, these pamphlets re-emphasized the need for pla-
toon and section commanders to act on their own initiative without wait-
ing for orders from above.This was a principle enshrined in pre-war practice,
but wartime experience had shown how vital it was on a battlefield where
communication was slow and difficult, at best. SS 144 also grappled with the
perennial tension between regimented step-by-step plans and a more lais-
sez-faire approach which might make it easier to exploit success. The com-
promise it proposed was that commanders should act in a manner appropriate
to the situations they found themselves in. This advice was sensible and
sought to prevent the micro-management that had been prevalent in many
formations. It did not, however, offer much guidance and so was not terribly
helpful, but then no attempt to reconcile two fundamentally opposed oper-
ational ideas at the tactical level was likely to be.20
The most important new contribution to the BEF’s operational thought
was SS 135, ‘Instructions for the Training of Divisions for Offensive Action’,
issued in December 1916.This came down firmly in favour of the methodical,
‘step-by-step’, approach. It stressed the importance of painstaking preparation
and of setting limited objectives within range of effective artillery support.
In essence, therefore, the British were now subscribing to what the French
called the ‘scientific’ method and moving away from deep-objective break-
through attempts. They were, in other words, signing up for the approach
the French were in the middle of jettisoning.
In some ways, then, the thinking of the BEF had made a major advance.
It had absorbed the tactical lessons of the Somme and was moving towards
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Scorch e d E a rt h155

finding repeatable solutions to the ‘break-in’ problem, largely by adopting

tactics reminiscent of those the French had developed in 1915 and 1916. At
the operational level, however, they remained behind. They had not yet
identified, much less solved, the problem with ‘scientific battle’: the slow
tempo which allowed the defender to maintain his equilibrium, reinforce
his defence, and dig in again.21
The British were not only learning new things, however. They were also
changing the way they learned. In February 1917 GHQ set up a new
Directorate of Training, under Solly-Flood, to coordinate and improve uni-
formity in training across the BEF. In addition to the new doctrine he pro-
duced himself, it was his job to reform and integrate the network of schools,
which had sprung up informally around the BEF from 1915 onwards, into
some kind of system. Eventually, in June 1917, SS 152 ‘Instructions for the
Training of the British Armies in France’ was distributed, laying out estab-
lishments and model curricula for all the schools. Learning, it was clear, was
now a serious business.22
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

15
The Battle of Arras

A t half past five in the morning of Easter Monday, amid driving rain and
flurries of sleet and snow, the British attack at Arras began. The inten-
tion was to break into the German defences, roll up the Hindenburg Line
from the north, and then push on Cambrai. Fourteen British divisions took
part in the first assault, a similar number to the first day of the Somme, but
the artillery on hand was much stronger. Where in 1916 Rawlinson had one
field gun every 20 metres and one heavy gun every 52, Generals Sir Henry
Horne (First Army) and Edmund Allenby (Third Army) now deployed one
field gun every 13 metres and a heavy piece every 25. Better ammunition,
and more of it, multiplied the effect.The shallowness of the German defences
up on Vimy Ridge allowed the attacking Canadian Corps to c­ oncentrate its
fire yet further. Proportionally, the Canadians unleashed twice the weight of
heavy artillery shell fired on the Somme, and thirty times that fired by the
French in May 1915.1
The main force of the attack fell on six front-line German divisions arrayed
in three ‘Gruppen’ (‘groups’): from north to south, Gruppe Souchez, Gruppe
Vimy, and Gruppe Arras. Each ‘group’ equated to, and was built around, a
traditional corps. The Germans were outnumbered by some three to one
in artillery and their guns were closely targeted, and much neutralized, by
British counter-battery fire both during the preparatory bombardment and
on 9 April.2
The German defenders were caught off balance by the timing of the
attack and were soon overwhelmed along much of the front. The wire had
been well cut and their trenches were badly damaged. Following closely
behind a creeping barrage, the attackers washed over the front line with
relative ease. Often, the garrisons never made it out of their dugouts. Up on
Vimy Ridge, the Germans lost the crest and were pushed back into their
third line on the eastern downslope. By eleven o’clock, a few companies of
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as157

defenders still held out in a strongpoint on Hill 145, but most of the ridge
otherwise was firmly in Canadian hands. (Hill 145, where the Canadian
National Memorial now stands, was marked as Hill 140 on French and
German maps and referred to as such in Chapter 8.) Further south, between
Bailleul and the River Scarpe, 14th Bavarian Infantry Division collapsed
altogether. Troops from British 4th Division advanced 6 kilometres, further
than anyone had managed since trench warfare began. For a giddy moment,
they thought they had broken through and called for cavalry to exploit,
but their success was so unexpected that none were on hand in time. South
of the river, also, the Germans lost villages such as Neuville-Vitasse, together
with many of their guns, and were driven back almost to the hilltop village
of Monchy-le-Preux.
Any confidence, or even complacency, that Sixth Army was ready for any-
thing had been shattered. As ever, early reports were sketchy. Communications
had largely been cut by enemy shell-fire. One runner took three and a half
hours to cover the 2 kilometres from the strongpoint on Hill 145 down to
regimental headquarters. When news did get through, it was universally
bad. At half past four Gruppe Arras reported despairingly: ‘Situation poor,
casualties very heavy, much artillery lost. . . . Ammunition and reinforcements
crucial. Number and weight of shell-fire much stronger than Battle of the
Somme. We have destroyed fifteen tanks but there are many more still there.’
The worst news was that all attempts to counter-attack had collapsed.
The reserve divisions, whose job it was to seal off any enemy penetrations
and drive the attackers back, were nowhere to be seen. Falkenhausen and his
staff had ignored repeated instructions to close them up to the front and all
bar one of the five reserve divisions remained east of Douai, a full day’s
march from the battlefield. Although Rupprecht tried to hurry them along,
little could be done now. Even the ready reserves of the garrison divisions
arrived too late and were too weak to do anything more than plug gaps.
The tired and disordered remnants of troops who escaped being overrun
that morning had no chance of recapturing the ground they had lost. By
nightfall, reinforcements were on their way. Rupprecht sent Sixth Army
two divisions at lunchtime. That evening, he added another five. These new
formations, he made clear, were to be moved up tight behind the front at
once, without worrying about possible billeting problems. These reserves,
at least, would be available if required.3
The question, of course, was whether they would arrive in time to hold
the British, or whether Haig’s men would be able to exploit their stunning
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

158 Holdi ng t h e Li n e 1916 –17

successes first. Rupprecht remained sure that this time his line would hold.
He was less convinced, however, how much that would achieve. He noted
that ‘the situation is undoubtedly serious, certainly as serious as that in First
Army at the beginning of the Battle of the Somme’. Although the Germans
had more reserves available this time, the terrain was more difficult and it
was very doubtful whether his men would be able to recapture Vimy Ridge.
He eyed the future with foreboding: ‘for the moment we can certainly
contain this push; but it’s questionable whether we’ll be able to withstand
further attacks supported by this weight of artillery. This poses a further
question: is it worthwhile for us to carry on fighting this war under these
circumstances?’4
Whatever the long-term outlook, in the short run the bad news kept
coming. Up on Vimy Ridge, the last defenders of Hill 145 were finally over-
come that evening. Over the next couple of days the British tried to exploit
their victory by pushing forward all along the line. They seized the high
ground of Monchy-le-Preux on 11 April, and the next day finally cleared the
defenders from their last footholds on the ridges of both Vimy and Lorette.
The British were not having it all their own way, however. An Australian
attack at Bullecourt turned to bloody disaster when they encountered the
fresh and highly trained Württembergers of 27th Division. The Australians
broke in to the German trenches but communications collapsed and confu-
sion followed when the Germans counter-attacked. About 1,200 Australians
were taken prisoner. Other British attacks were also repelled. Indeed, British
momentum was draining away. The attackers were tiring; there were no
fresh reserves or, at least, those which did exist were not released by GHQ;
coordination between units and arms was deteriorating; and all the time the
German defenders were improving their positions and getting stronger.
By the evening of 11 April Rupprecht was optimistic that the worst of the
crisis was now past. The idea of Sixth Army launching an attack to recapture
lost ground, which Rupprecht and Kuhl had resisted, was now abandoned.
The decision was taken once more to trade space for time. During the night
of 12/13 April, the Germans abandoned their exposed positions at the
foot of Vimy Ridge and pulled back 4–5 kilometres into more sustainable
defences in their third line, running from Méricourt to Oppy and Gavrelle.
The British followed up slowly. To crack the new line would require a full
set-piece assault, which would take the British Expeditionary Force (BEF)
time to prepare. Fighting over the next few days became increasingly scrappy
and small scale as the intensity of operations died away.5
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as159

Figure 15.1.  The front-line trenches of the Hindenburg Line (Siegfried Stellung)
near Bullecourt

The British lost 13,000 killed, wounded, and missing in the first phase
of the Battle of Arras, a quarter of the losses of 1 July 1916. The normally
understated British official historian described 9 April as ‘one of the great
days of the war. It witnessed the most formidable and at the same time
most successful British offensive hitherto launched.’ The advance averaged
6 kilometres on an 18-kilometre front. The Germans lost 233 guns and
23,000 men, of whom 16,000 were captured or just went missing. Seven
German divisions were so badly chewed up that they had to be relieved
without delay.6
Inquests into the disaster began at once. Why had the German army been
unable to contain an attack it had seen coming? With another huge French
offensive known to be imminent, there was an urgent need to check whether
the new defensive doctrine was flawed. Oberste Heeresleitung (OHL) began
a thorough debrief, interviewing officers who had taken part. Every level of
command involved, from regiment up to Army Group, wrote a report
on  their experiences and what had gone wrong. The independence and
objectivity of this process, however, was immediately prejudiced by a wave
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

160 Holdi ng t h e Li n e 1916 –17

of sackings, instituted by OHL over Rupprecht’s objections. Sixth Army’s

chief of staff, Major-General Karl von Nagel, was fired on 11 April. Loßberg
replaced him. Nagel’s assistant, Rudolf von Xylander, was also let go. Both
men were blamed for the failure to move reserves up in time. Falkenhausen,
the army commander, was badly shaken and offered his resignation to the
Kaiser. This was rejected, but a few days later an opportunity arose to
promote him out of the way when the Governor-General of Belgium died
in office. Falkenhausen replaced him in Brussels and Otto von Below took
over Sixth Army.7
These dismissals set the tone. It was now clear that OHL would be com-
fortable with reports which blamed the defeat of 9 April on the mistakes of
individuals, not on the new defensive system itself. Not surprisingly, most
of the after-action reports which came in took their cue from this. A pre-
liminary memorandum from Rupprecht on 12 April, for instance, suggested
a couple of reasons for failure. First, reinforcements of guns had not been
properly dug in and registered and so were of little use. Second, infantry
reserves were in the wrong place. A fuller report on 21 April discussed a
broader range of factors but re-emphasized these two as the most important.
Some points that subordinate formations made, which OHL might not like,
were not repeated on up the chain of command. For instance, suggestions
that the British might actually have planned and executed a sophisticated
combined arms attack rather well were suppressed. On the other hand, the
principles of defence in depth were, it was argued, shown to have worked,
even if in places on 9 April the terrain or state of the defensive works made
them impossible to apply.8
That the boss sometimes gets told what his subordinates think he wants
to hear is hardly news in any context, much less in the history of the German
army between 1914 and 1918. This instance, however, represents the first
time that we can see this process so clearly in action under Ludendorff
and Hindenburg. In the autumn of 1916 the thought processes of the
chain of command had briefly been characterized by a level of rigour
and  intellectual honesty which permitted the German army to tear up
established methods and bring in a largely new defensive doctrine. This
was possible partly because Ludendorff and Hindenburg possessed a measure
of self-confidence which Falkenhayn had long ago lost and partly pre-
cisely because their new approach overturned his ideas. The result was a
level of objectivity much greater than the chain of command had achieved
previously. The British success on 9 April, however, applied a new level of
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as161

pressure to the system, which reverted to its default setting: too much
truth hurts.
This becomes only clearer when we examine why the British thought
they had done as well as they had. The official historian identified three
main factors at work. Two of these, British artillery superiority and the slow
arrival of the defender’s reserves, corresponded closely with those identified by
the Germans. He also suggested a third factor, however. The German
defensive scheme, he argued, being new, was not only grafted onto forti-
fications not suited to it, but indeed was imperfectly understood by the
defenders and not properly implemented. He argued that the British
artillery plan, despite intelligence that German defensive methods had been
reformed, was designed to overwhelm an old-style defence. It only worked
because ‘the new German system had not made very much progress’. Instead,
‘contrary to Ludendorff ’s original design, the infantry had received orders
to hold the forward system at all costs, instead of regarding it as a true
outpost system. Not only so, but it was held in such strength that when it
was overrun each division lost about one-third of its infantry.’9
After the war, the German official historians accepted that the new
principles were not properly applied. Partly this was because they were hard
to implement. The terrain, most obviously up on Vimy Ridge, offered little
depth: the eastern face was so steep that losing the crest meant losing the
whole thing. Partly it was the result of a failure of higher command, from
army level up to and including OHL, to understand how imminent an
enemy offensive was. Preparations had not been made with sufficient thor-
oughness or urgency. Too few shelters had been built behind the front lines,
so the garrison was left forward, where it proved highly vulnerable to British
combined arms tactics. Building the Hindenburg Line had diverted resources
which might have helped construct Sixth Army’s defence in depth. The
success of ALBERICH promoted a sense of security which proved all too
false. But partly, also, it was based in a failure to understand and apply the
new doctrine. Although, as we have seen, the after-action analyses did not
emphasize this, there is nonetheless evidence to that effect. Even where the
ridge was not a problem, units had been ordered to defend well forward and
to hold their positions to the last, rather than to roll with the punches.
A good example is the 8th Bavarian Infantry Regiment, holding a sector of
flat ground just north of the River Scarpe. One battalion was being held right
back in the outskirts of Douai, some 22 kilometres and at least six hours
away. Of the remaining eight companies, six and a half were in the front line.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

162 Holdi ng t h e Li n e 1916 –17

Nine hundred metres of the second line ended up being held by one
non-commissioned officer (NCO) and twelve soldiers. This sector was
one of those which collapsed most disastrously on 9 April.10
The significance of the British assault on 9 April must not be exaggerated.
Tactically, it showed that the BEF, given sufficient preparation and resources,
could break in to a strong defensive position on a broad front, as it had failed
to do on 1 July 1916. The cost was also lower. This represented a measure of
progress.Vimy Ridge, with its views across the Douai Plain, the key to Artois,
constituted a prize in its own right. It was a prize, moreover, which had
eluded the French army twice in 1915. Impressive as the Canadian capture
of Vimy was, however, it is important to remember that it was a sideshow,
designed to secure the flank of the broader British attack, which was,
itself, a sideshow designed to draw attention away from Nivelle’s offensive
further south.11
Moreover, in one important respect the British seemed to have made
little headway. They remained as incapable as ever of turning tactical advan-
tage into operational success, much less strategic victory. Even on 9 April,
this had been a factor. We have already seen that the cavalry were not in
position to exploit success on the north bank of the Scarpe. More generally,
decisions made in planning reduced the chances of successful exploitation
on the first or second day. For instance, each corps was set up with three
divisions in the first wave of the assault and just one ready to leapfrog
through them and continue the advance to the final objective. Since the
final objectives were not reached, either the objective was too deep; or
a  single division on a three-division frontage was not enough; or both.
Another example arises with the handling of XVIII Corps, the three divi-
sions of which constituted the Third Army reserve. The choice of when and
where they were to be used belonged not to the army commander on the
spot, General Sir Edmund Allenby, as one might expect, but to Haig. No
units were to move without GHQ’s permission. Perhaps Haig did not fully
trust Allenby, who certainly lacked experience of command at this level.
Still, this was a strange arrangement, not least after Haig had criticized
something similar at Loos eighteen months before.12
The British made other old mistakes all over again. During the days
immediately after 9 April, in an attempt to maintain momentum and exploit
gains against an enemy whose weakness Allenby misjudged, they rushed
through the planning and execution of a cluster of narrow-front attacks. All
too often these achieved little, foundering on tired troops, poor intelligence,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as163

and insufficient coordination. The attack of the 9th Division on the Roeux
chemical works on 12 April was a good example. No-one thought to
synchronize this operation with the attack of the neighbouring division.
There was no time to arrange a smoke screen to cover the approach march
and German guns disrupted the assembly of the assault units. The troops
were cold, wet, and tired. They had received no hot food for days and were
unable to keep up with a creeping barrage which advanced unrealistically
fast. The British heavy artillery bombardment was inaccurate and ineffect-
ive. Ground gained was minimal and soon abandoned. As the official history
noted, ‘it is one thing to plan and organize when there is plenty of time,
but . . . in modern war it is often quite another matter when there is time
only for hurried improvisation, carried out by largely amateur staffs’.
Bashing away in this fashion had not achieved much on the Somme and it
yielded little in 1917, too. Haig recognized this on 15 April when, after rep-
resentations from some of Allenby’s division commanders, he ordered a halt
to such attacks and an operational pause.13
Abandoning the Arras offensive was not yet an option. The main purpose,
after all, had been and still was to prepare the way for the French offensive
which General Robert Nivelle had promised would win the war. This main
push began along the Chemin des Dames on 16 April. Ground was made
both here and in a subsidiary effort in Champagne. Some twenty-nine
French divisions took part, twice as many as the British used at Arras. OHL
was never really concerned, however, and the Germans were inflicting very
heavy losses on the French. By 20 April it was clear to Paris that the offen-
sive had stalled and breakthrough remained a dream. The French attacked
again on 4 May, this time with more limited objectives. They found resist-
ance no less ferocious. Over the next week the main operation was wound
down. Nivelle, after much politicking, was finally removed on 15 May. Haig
had been astute enough to suspect that the French were struggling and
that Nivelle’s position was under threat almost from the first. Nevertheless,
politically he could hardly suspend British operations until he was officially
told that French plans had changed.14
Rupprecht, of course, was unaware of any of the intrigues behind enemy
lines, although on 15 April he correctly judged that the crisis for Sixth Army
was past. On 21 April the Germans began to pick up signs of a renewed
British attack at Arras. When this assault duly unfolded on 23 April, it was
carried out in a very similar fashion to that of a fortnight earlier, under
a  thick smoke screen and heavy bombardment. Nine British divisions
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

164 Holdi ng t h e Li n e 1916 –17

took part. Although Loßberg described this as the heaviest attack he could

remember, British gains amounted to no more than a few hundred metres
for the loss of 8,000 men. The German corps commanders were able to cope
using only their own resources and required no extra help from army or
Army Group reserves. Three major factors contributed to the disappointing
outcome. First, the British never achieved the same artillery dominance that
they had on 9 April. German artillery had been reinforced while the British
had not. Counter-battery intelligence had been unable to plot the positions
of German batteries so thoroughly. On 9 April many of the German obser-
vation posts on Vimy Ridge had been overrun, leaving the guns blind. The
same did not recur on 23 April. Second, the British troops were tired, in
some cases having been in action repeatedly for a fortnight. Third, this time
the German reserves were quickly on hand. Already by midday they were
working to seal off and eliminate enemy break-ins. Another British attack
on 28 April with six divisions suffered from all the same drawbacks and
managed only to capture the village of Arleux.15
On 3 May Sixth Army found itself under full-scale attack once more,
as  fourteen British divisions from three different armies attacked over
22 kilometres from Bullecourt to Fresnoy. Once again, the Germans felt
under little threat. British command mistakes helped. A late change of time
for Zero Hour left plans in confusion and formations in disarray. The assault
waves found themselves silhouetted against a setting moon, offering easy
targets. When German infantry judged the assault imminent, they moved
forward into the shell-holes of No Man’s Land to escape the bombardment
and fight a mobile battle, disrupting the enemy until he was thrown back by
a counter-attack coming from deep. British units seemed unable to cope
with these counter-attacks, in some cases breaking and running. Overall,
progress was minimal. The only gain, the village of Fresnoy, was recaptured
by Bavarian troops a few days later. ‘We have every reason to be happy about
yesterday’s events’, wrote Rupprecht in his diary. Although Bullecourt
remained the focus of bloody combat for another fortnight and there were
occasional flare-ups of violence elsewhere along the Sixth Army front until
24 May, the Battle of Arras had lost any purpose it might ever have had by
nightfall on 3 May. Rupprecht noted that British attacks later that month
were merely repeating the pattern observed in the later Somme battles:
uncoordinated, poorly synchronized, small-scale, local operations. German
attention began switching to Flanders and on 30 May Sixth Army formally
reported that it expected no further major offensives in its sector.16
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as165

Nearly another quarter of a million men were killed, wounded, or

captured in the Battle of Arras. Again, for every soldier the Germans lost,
the British lost two. The original idea for the battle had been ‘to pin the
enemy . . . draw in his reserves, and thereby facilitate the task of the main
French attack’. None of these objectives was achieved. Even the normally
sympathetic British official historian could not resist a jibe: ‘as usual, per-
formance had lagged behind promise’. The British never looked like
breaking through to Cambrai. Even before it became clear that Nivelle had
failed, any idea of joining hands with the French seemed far-fetched. Arras
neither distracted the Germans away from the impending French attack
nor  sucked in their reserves before the main blow fell in the south. The
Germans knew Nivelle was coming and bolstered their defences accordingly.
Indeed, if anything Arras served to make the Germans even more alert. There
were plenty of fresh formations available to meet the double threat. Between
1 March and 20 April, the Germans sent more divisions to reinforce the
Chemin des Dames (eighteen) than to Arras (fourteen) and still held at least
another dozen in reserve. Short-lived and unsuccessful as the Nivelle offensive
was, it accounted for nearly twice as many German soldiers as Arras did.17
Even seen purely as an exercise in attrition, while the Entente of course
could absorb more casualties than the Central Powers, especially now that
the United States had joined the war, to be losing men so much faster than
Germany was hardly a palatable long-term approach. There was a danger
that, as the French General Fayolle had noted of his men on the Somme
the previous year, men of the Entente might become ‘fed up with getting
themselves killed, with no important success, no decision’. This was all the
more true when it seemed that at least some of the losses were rooted in a
failure to learn from previous mistakes. The British official historian, with
uncharacteristic bluntness, asked why the Battle of Arras became ‘pro-
foundly disappointing’ after the triumph of Easter Monday. He identified
three problems which impacted the British ability to fight effectively after
the first day. The first was an external factor, to some extent beyond British
control. The difficult terrain, and especially the problems faced building
roads across the shattered battlefield, constrained supply and slowed the
tempo of operations, allowing the Germans time to prepare their defences.
Second, commanders erred in repeatedly sending tired troops into improvised
local attacks when the situation called for a pause to prepare a new large-scale
full-dress offensive. The third factor was more fundamental, however. The
underlying problem was that, as Cyril Falls put it, ‘the atmosphere of siege
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

166 Holdi ng t h e Li n e 1916 –17

warfare hung about the British army when it passed to open or semi-open
warfare, and it was by no means only the rank and file or junior officers
who were still affected by it’. This came through in many ways. Staffs would
plan thoroughly for the first day, but fail to consider what might happen
thereafter, becoming intellectually paralysed by the inevitable uncertainties
involved. Too often, instead of the chain of command remaining flexible, it
became rigid. With communications inevitably poor, the centre tried to
exert greater control itself rather than to delegate authority to the man on
the spot. Artillery might be devastating to begin with but, once deprived of
thorough intelligence and struggling with the problems of redeployment
forwards, the guns rapidly lost effectiveness. Infantry performed well
when working to the script, but improvised less impressively, at least until
it had the opportunity to learn on the job. In summary, the army could
be quite effective on the first day of an offensive but became hesitant and
unsure when asked to go off-script. That, Falls implied, was largely a failure
of training.18
To Falls’ three we can add three more. First, as Jim Beach has shown, Haig
and his intelligence staff made the mistake, for neither the first nor the last
time in the war, of under-estimating available enemy reserves and reading
too much into strategic reports of uncertain value. They assumed that
anecdotal reports of German home-front discontent would translate into
exploitable battlefield weakness. Second, the new German elastic defence,
when properly used, was sufficient to contain and ultimately defeat British
attacks. The contrast between outcomes on 9 April, when the defence was
shallow, and 23 April, 28 April, and 3–4 May, when it was arrayed in depth
with reserves ready to counter-attack promptly, demonstrates the truth of
this at the tactical level. But it was true operationally, too. While the British
were able to overrun the first and second German positions, the next line
of defence stood firm and prevented a breakthrough. Third, the Germans
proved quicker to learn and adapt than the British. The tactical tweaks the
BEF had introduced over the winter of 1916/17 could overwhelm an
old-style German defence, as 9 April showed. When the Germans began to
defend more skilfully in depth, however, the British were lost once more.
As Rupprecht reported to OHL on 13 May, elastic defence had now worked
well in three major battles. It was not leading to increased casualties, as had
been feared. According to Loßberg, the troops liked the new approach.19
The Germans worked hard at learning and disseminating lessons as they
went along. For instance, in early May Rupprecht held a conference where
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le of A r r as167

the General Officer Commanding (GOC) of the 3rd Bavarian Infantry

Division ran through recent lessons of the fighting in which he had taken
part. A practical demonstration of how to operate a mobile elastic defence
followed on the training ground, with every formation in the Army Group
represented.20 OHL was continually collating data on enemy attack methods
and comparing them with German tactics.21
That is not to say, however, that everything was for the best in the best
of  all possible German worlds. In particular, there were signs of growing
rigidity within the German command set-up. On 1 May, for instance, Sixth
Army sought to increase its tactical control at the expense of its subordinates’
initiative. Below ordered generals commanding divisions and corps to check
with Army headquarters before undertaking any retreat.22 Less than two
weeks later, Ludendorff started micro-managing the deployment of a div-
ision in Second Army and even ordered that henceforth no division was to
be moved without his agreement. ‘Such a reduction of my powers is defin-
itely too much’, complained Rupprecht in his diary. ‘What’s the point of
having army groups, if every decision must be approved by OHL? Ludendorff
cannot oversee every detail on every front. If he wanted to do so he inevitably
would neglect some other and more important missions. The paperwork
generated by [his] inquiries and answers to them has in any case already
grown a lot.’23 By the spring of 1917, the relaxation in command Ludendorff
and Hindenburg sponsored the previous autumn was fading away. Central
control was being reasserted and the flow of information upwards, as we
have seen, was losing its purity. The German army was reverting to type
and the weaknesses which would lose it the war were coming once more
to the fore.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

16
The Battle for Flanders
Summer 1917

T he double failure of the spring smashed British and French hopes.

With Nivelle’s methods discredited and the morale of the French army
vulnerable, they would obviously have to change approach. On 4–5 May
statesmen and soldiers meeting in Paris resolved to carry out a series of
­low-risk, limited-objective attacks. With luck, these might achieve local
successes and over time build up to exhaust Germany’s power of resistance.
The Germans needed little more than common sense to guess this would
be the case. Nonetheless, it was confirmed by one of their highly placed
agents in Paris.1
Where would the next offensive strike? Back in April Oberste
Heeresleitung (OHL) had already begun to worry about the threat to
Flanders. If the British Expeditionary Force (BEF) drove east and swept the
Germans from the higher ground overlooking the Ypres Salient, it might be
able to threaten vital railway nodes at Roulers and Courtrai, undermining
the whole German position in Belgium and threatening the U-boat bases
along the coast. Worryingly, rail activity in the British rear seemed to be
picking up so Kuhl visited the German Fourth Army, commanded by
General Friedrich Sixt von Armin, which was responsible for the defence of
Belgium. After his tour, Kuhl highlighted the weakness of the defences in
the mini-salient south of Ypres around the villages of Wytschaete and
Messines. The Germans had held the high ground here since autumn 1914,
not least because the ridge commanded excellent fields of fire towards the
British lines.To cover No Man’s Land, however, the defenders had to occupy
trenches on the westward, forward slope. Such positions, once relatively
sensible, were now highly vulnerable to the artillery-intensive methods of
the BEF, as the battle for Vimy Ridge had shown. Further, the terrain of the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le for F la n de rs: Sum m e r 1917169

ridge itself militated against a defence in depth, as it had at Vimy. If the

Germans lost the crest, the next suitable position to defend was the Flanders
Line, some 10 kilometres back. To avoid this danger, Kuhl suggested with-
drawal to an intermediate Third, or Warneton, Line, but Sixt and his com-
manders all argued for holding in place. Rupprecht and Kuhl were not
prepared to overrule the men on the spot. Instead, they sent a division of
infantry and some gun batteries as reinforcements.
Throughout the second half of May it became more obvious almost
every day that an attack on Messines was approaching. British artillery more
than doubled in strength between March and June.The Royal Flying Corps
was getting more active in the air and almost every night the British sent
groups of infantry off across No Man’s Land to raid the German trenches.
Equally, British forces further south seemed to be thinning out, making
operations there less likely. Radio intercepts showed that Canadian troops
around Loos were being sent on leave. Relative quiet in the rest of the Ypres
Salient suggested that any attack on Messines would be a local attack, rather
than part of a larger operation, at least for now. By early June it seemed
imminent.2
The Germans were as prepared as they could be. The garrison had been
reinforced and no fewer than four defensive positions had been built, one
behind the other. Concrete pillboxes and blockhouses protected command
posts and machine-gun nests. Reserves stood nearby. The British artillery
bombardment, three and a half million shells between 26 May and 6 June,
flattened the villages of Messines and Wytschaete, cut barbed wire, and neu-
tralized German batteries. About a quarter of German field guns, and nearly
half of the heavy artillery, was put out of action by counter-battery fire.The
front line was reduced to a wasteland of craters a kilometre deep. Only one
artillery observation post survived. Two of four German divisions garrison-
ing the front under this bombardment had to be relieved even before the
enemy infantry assault began.3
For all the warnings the defenders had received, when the attack came it
still took them by almost complete surprise, because the British used a
weapon the Germans had overlooked: the mine. Under-estimating the British
ability to tunnel deep, the defenders discounted the possibility of large-
scale mining against the Messines Ridge. At ten minutes past four on the
morning of 7 June 1917 they were proved catastrophically wrong. Half a
million kilograms of explosive, in nineteen different mines, went off together
underneath the middle of the German front lines. The German official
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

170 Holdi ng t h e Li n e 1916 –17

history quotes an observer: ‘we saw nineteen giant roses with bright red
petals, or immense mushrooms, rise slowly and majestically out of the
ground. They then broke apart with a dull roar. Immediately afterwards
bright multi-coloured pillars of fire and smoke shot skyward, carrying earth
and debris up with them.’ The earthquake was felt 30 kilometres away; the
roar could be heard in London.The 3rd Bavarian Infantry Division, defend-
ing Messines itself, effectively lost all its three forward battalions in the mine
explosions. Only three officers and thirty men survived.4
Even before the roar of the explosions had faded away, the guns of British
Second Army opened up again. The artillery plan was the most sophisti-
cated yet. It incorporated three main belts of fire, 650 metres deep. Two-
thirds of the field artillery delivered a creeping barrage which rolled forwards
ahead of the assaulting infantry at a rate of 100 metres every two minutes.
Machine-guns also took part. Further ahead, targets of importance such as
command posts and machine-gun nests were singled out and bombarded
until overrun. All the while, counter-battery missions neutralized enemy
artillery. The infantry of nine British divisions advanced to the attack, sup-
ported by seventy-two of the new Mark IV tanks.
The resistance they faced was, on the whole, feeble.Those defenders who
had not been killed or wounded by the bombardment or mine blasts were
badly shaken. In many cases they either gave themselves up or headed for
the rear. Clouds of smoke and dust blocked both defensive fire and effective
command and control. The latter had been further complicated by a deci-
sion to split responsibility for the defence between two ad hoc battlegroups.
The British rolled easily over the front line and, by four hours after Zero
Hour, had seized control of almost the entire German second line along the
crest of the ridge, including the ruins of Wytschaete and Messines. Here
they planned to dig in, consolidate, and, with the benefit of uninterrupted
observation over the German rear, meet and defeat any counter-attacks.
At around three p.m. a second phase would follow, with fresh troops passing
through to continue the attack.Their objective lay down the east side of the
ridge at the Oostaverne Line (known to the Germans as the Sehnenstellung)
which ran across the base of the salient.
As the British troops dug in along the crest they found themselves silhou-
etted against the skyline and uncomfortably tightly packed into a small area.
Losses to machine-gun and artillery fire were heavy. Nonetheless, the British
plan seemed to be working well. German reserves turned out once more to
be too far back to counter-attack effectively. Some had to march 16 kilometres
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le for F la n de rs: Sum m e r 1917171

before they reached the battlefield. Worse, they were unfamiliar with the
ground and exposed to British shelling as they approached. Conflicting
orders from the two battlegroup commanders injected further confusion.
The reserve divisions ended up being committed piecemeal and late.
Counter-attacks which were planned for the morning did not take place
until mid-afternoon. By then, the British were easily strong enough to
beat them back into the Oostaverne Line, although the BEF’s second-wave
attack then met little success in turn.5
Rupprecht could not achieve much on 7 June. This was a short, sharp
corps-level defensive battle which even the men on the spot had difficulty
keeping control of. In his diary, he queried some aspects of the handling of
the battle. Why did the reserves arrive late? Why had Fourth Army not
rotated divisions out of the line sooner, if they were worn out? He doubted
his men were strong enough to recapture the Messines Ridge. Could they
hold on to the Oostaverne Line, though? Rupprecht felt that, as soon as
the British moved guns up on the high ground, this position, too, would
become untenable.
What to do next? The corps commander at Messines, General Maximilian
von Laffert, had also been shaken by the events of 7 June. His first reaction
was to retreat over the River Lys. This risked losing much of the artillery,
though, and Fourth Army forbade it. Sixt proposed staying put and seeing
what turned up. Rupprecht favoured a retreat of some 10 kilometres into
the Flandern-Stellung (Flanders Line), while Kuhl was more conservative.
He expected a major follow-up offensive at Ypres and was keen to protect
the southern flank of their positions there, so did not want to pull back too
far. He again suggested a compromise, pulling back into a new position built
around the old Third (Warneton) Line and along the banks of the Ypres–
Comines Canal. Eventually, after repeated pressure for a decision from
Ludendorff, on 10 June Kuhl told Sixt to withdraw to this intermediate
position that evening.The British took over the evacuated Oostaverne Line
but were in no position further to exploit their success. By 14 June, major
operations in the Battle of Messines ceased.6
Messines was, without doubt, a BEF victory. The British achieved their
objective of seizing the ridge. They could now push ahead with their plans
for a further drive out of the Ypres Salient in July. They took over 7,000
prisoners and inflicted some 25,000 casualties on the Germans. They owed
their success, the German official history later concluded, to five main
­factors. First, the British were superior in both manpower and matériel,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

172 Holdi ng t h e Li n e 1916 –17

especially in artillery where they outnumbered the Germans by more than

three to one. Second, the mines had a tremendous impact, destroying
defences and demoralizing defenders. Third, the German forward slope
positions were flawed and vulnerable. Fourth, the British had been able to
enfilade the defenders from both flanks. Lastly, the counter-attack divisions
arrived too late.
The German command, however, need not get too depressed by the
Battle of Messines. At least one of these five factors was unrepeatable.
Mining on such a scale would not be feasible again. Further, as Rupprecht
noted, ‘the British are great at painstaking planning of attacks, but never
manage to exploit their initial success’.The British remained unable to earn
maximum return on the huge investment of resources involved in battles
such as Messines. British casualties were also around 25,000 men. How long
would they be able to carry on with such a costly way of war, when the
returns were so slight? The main lesson Rupprecht drew from Messines was
that his defence must be more elastic and mobile, not only to pull back from
positions that were vulnerable to superior enemy artillery, but also to exploit
the German army’s perceived comparative advantage in open warfare.7
That the Battle of Messines would be followed by a further British attack
in Flanders was partly a matter of common sense. Unrest in France, both at
home and in the army, made a major French effort unlikely. The British
could not logistically support an attack across the old Somme battlefield and
the area devastated by ALBERICH. They had been fought to a standstill in
Artois. Only Flanders remained. German intelligence estimated the most
likely effort would combine a drive up onto the ridges surrounding Ypres
with a push along the coast, perhaps supported by amphibious landings.
It  was clearly desirable that Germany have the ablest possible command
team in Flanders to meet this onslaught. Messines had shown that Sixt and
his chief of staff, Stapff, were not the best. Loßberg, the acknowledged
defence expert in the German army, was currently chief of staff to Sixth
Army around Arras, under the tough and experienced Otto von Below.
Kuhl and Ludendorff discussed the problem on 12 June.They felt that, since
Loßberg and Sixt would not collaborate well, just swapping over the chiefs
of staff would not work. Instead, Kuhl proposed exchanging the entire staffs
of Fourth and Sixth Armies. Ludendorff agreed. So did Loßberg. Rupprecht
approved the idea and forwarded it to Hindenburg. When the Field Marshal
took the proposal to the Kaiser, however, Ludendorff told Kuhl, something
seemed to go wrong. Stapff and Loßberg were ordered to change places
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le for F la n de rs: Sum m e r 1917173

without anything further being mentioned about the army commanders,

much less the rest of the staffs. Two days later, Kuhl and Loßberg tried again
to get Sixt moved. This time, Ludendorff told the full story. There had been
no mistake on the first occasion. Instead, the Kaiser had vetoed any change
of army commander. This was his form of revenge. Six months previously,
Rupprecht had got two corps commanders sacked for poor performance
after the Somme, despite both men having served at the Kaiser’s court as
imperial adjutants. Wilhelm II remained piqued. Kuhl’s diary entry finishes
on an incredulous note at this pettiness: ‘Does he not realize the seriousness
of the situation?’8
Hard evidence that the British were planning an offensive in Flanders
multiplied as the summer continued. The Germans watched as enemy
­engineers built roads, railways, and camps for the new troops arriving.
Diversionary operations around Arras fooled no-one. British artillery began
to build up in the salient and BEF units replaced Belgian troops in the
bridgehead over the Yser at Nieuport, which Sixt attacked on 10 July in a
successful attempt to disrupt plans for an attack along the coast.
On 11 July the Royal Flying Corps launched an offensive to win control
of the air over Flanders. The result was some of the largest aerial battles of
history so far. On 26 July, for example, a battle developed over Polygon
Wood in which no fewer than ninety-four fighters eventually became
embroiled. By then, the pace of the British concentration was picking up.
German intelligence identified nearly twice as many batteries opposite
Fourth Army as a fortnight previously. In response, Rupprecht built up
Fourth Army’s reserves of both manpower and matériel. Net, Sixt received
eight divisions in May, five in June, and a further four in July. Fourth Army
expanded to seventeen divisions in the front line and thirteen and a half in
reserve, representing nearly half the strength of Rupprecht’s Army Group.
Loßberg began to build new defensive positions on the Passchendaele
Ridge and Gheluvelt Plateau, so that the British would now have to fight
through between five and seven lines of defence in a zone more than
12 kilometres deep.9
On 15 July the British guns began to register their targets with ranging
shots. Within ten days it was clear to Rupprecht that a full-scale artillery
battle was underway. Despite poor weather, German guns could not always
keep up with the pace of action. Thirty per cent of the heavy guns in
Gruppe Wytschaete had broken down or were put out of action by enemy
fire. Rupprecht noted that the bombardment was heavier than during the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

174 Holdi ng t h e Li n e 1916 –17

worst days on the Somme. Indeed, where in July 1916 the British had fired
1.7 million rounds from about 1,400 guns, a year later at Ypres a little over
3,000 barrels let loose nearly 4.3 million shells. Three times as many shells
hit each kilometre of front as on the Somme. One single German battery
reported being targeted by no fewer than a thousand rounds. Although the
bombardment did a good job of destroying the German front line on the
left and centre of the planned attack, little damage was done on the right,
on the Gheluvelt Plateau, where the terrain was more difficult, or to the
German rearward defences. The British commander at Ypres, Sir Hubert
Gough, was once again setting objectives too deep for the weight of fire he
had available.10
On 31 July nine British and two French divisions launched the main
attack, jumping off at ten minutes to five in the morning. On the left, the
French made their objectives and more.Their mission, however, was only to
guard the flank of Gough’s Fifth Army. The main British attack had mixed
results. On the left and in the centre, Gough’s men advanced about
2,700 metres and captured their second objectives. Stiffening resistance pre-
vented them reaching the third, as did a breakdown in coordination with
the artillery. On the right, up on the Gheluvelt Plateau, the British were
much less successful. The German artillery here had escaped the worst of
the counter-battery fire and was able to target the British infantrymen, who
were advancing too slowly to keep up with the creeping barrage. Soft ground
and the need to clear the many copses and woods in their way caused delay.
The defenders had plenty of time to block this increasingly uncoordinated
assault. They restricted the British to a gain of about 900 metres.11
Fifth Army remained less than halfway to its day one objectives, had suf-
fered heavily, and had badly under-estimated German powers of resistance
and counter-attack. Nevertheless, relative to some of the operations which
had gone before and, indeed, some of those which came later, it is possible
to see 31 July as a success of a kind. Robin Prior and Trevor Wilson have
pointed out that overrunning a couple of defence lines, capturing 45 square
kilometres of enemy territory, and clearing away German observation posts
overlooking the town of Ypres were all solid achievements, especially com-
pared to the 6 square kilometres taken on the first day of the Somme.
British casualties, at 31,850 men, were heavy in absolute terms but unsur-
prising for an attack of this scale and nature, and again looked light relative
to 1 July 1916. In any case, the Germans had suffered heavily too. Fourth
Army lost about 30,000 men in the last ten days of July. The failure to clear
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le for F la n de rs: Sum m e r 1917175

the Gheluvelt Plateau, however, left the Germans able to enfilade the flank
of any attack further north and would haunt subsequent British operations
for weeks.12
The action of 31 July, now known as the Battle of Pilckem Ridge, caused
little concern within the German command. Fourth Army had everywhere
been able to contain the attack with its own resources. On the first day of
every previous battle examined so far, the defenders’ almost immediate
reaction was to call for reinforcements.This time, Fourth Army felt no need.
Rupprecht looked on with great calm:
I was looking forward to the attack with the greatest equanimity, since we had
never been so strong on a front under attack and our reserves had never been
so rehearsed in their roles as we were that day. . . . Considering the strength of
the enemy attack and the fact that experience tells us that the first blow is the
most dangerous, we have plenty to be happy about with the events of today’s
fighting.
Ludendorff alone seems to have fretted. He telephoned Kuhl many times for
updates during the day. This may tell us more about the stress he was feeling
than about the battle. By 2 August, however, he was more confident and ask-
ing Fourth Army to give up divisions for the sake of operations in Russia.13
There was no room for complacency, however.The usual process of after-
action review went ahead. Some intriguing points emerged. Rupprecht
blamed the French success on the collapse of a Saxon division and poor
leadership by the corps commander, General Martin Chales de Beaulieu.
The most noteworthy feature was the poor quality of the British infantry.
They were over-reliant on artillery support and went to ground as soon as
they came under even the slightest machine-gun or rifle fire. Another old
issue came up again: who was to control counter-attacks by the rapid
response (Eingreif ) divisions? Was it to be the commander of the division
itself, or the man in charge of the defence forward? It is striking that this
obvious and important question, which we saw come up before at Arras,
had not been resolved even after some six months of practical experience of
the new defensive methods.14
Rain started to fall late on 31 July and did not stop for three days and
nights. August continued wetter than average, with only three dry days, and
the salient rapidly became a series of huge swamps. The poor conditions
inevitably made operations more difficult for the attacker. It became espe-
cially hard to move guns and shells. Even for the defenders, though, the
weather caused major problems. Shell holes and trenches filled with water.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

176 Holdi ng t h e Li n e 1916 –17

Troops were left out in the open, huddling under scraps of corrugated iron
and wood in a desperate attempt to stay dry. Rupprecht worried that, in
these conditions, with sleep difficult, units would wear out and need relief
more rapidly. This concern proved exaggerated, but it was not groundless:
on the Somme, the average division had sustained 4,000 casualties during a
fortnight’s stint in the front line, whereas in early August 1917 German for-
mations were having to be replaced after losing only half as many. The mud
made reliefs themselves tricky. Heavy British use of gas made moving food
and supplies up to the front even more difficult. Bearers slipped and slid
through the slime as they carried supplies forward in the dark and found it
hard to fit their respirators in time. One Bavarian division alone had 1,200
men gassed. On balance, though, Kuhl felt the rain was more help than
hindrance: ‘We’re lucky, the weather is very bad, constant rain, in Flanders
the soil is without foundations, so one can only proceed on the roads. Thus
we have time to regroup, carry out reliefs, etc.’15
Regardless of the poor weather, General Gough was eager to push his
men on. In addition to a number of smaller actions, he launched a series of
major attacks over the next few weeks, all of them disastrous, all of them
suggesting that he had learnt none of the lessons of previous battles.
­Ill-prepared and poorly coordinated attacks threw tired units against unsup-
pressed defences. The results were depressingly predictable. On 10 August,
two British divisions again made no significant progress on the Gheluvelt
Plateau but lost 2,200 men. The Battle of Langemarck on 16 August saw a
broad-front assault by nine British divisions, plus the French again. Again,
the Germans lost ground in the north but their defences in the centre and
south held firm. On the difficult terrain of the Gheluvelt Plateau, the British
made their third assault on alert defenders and chalked up their third failure.
Altogether, the British suffered 15,000 casualties.The Germans lost St Julien
to a special operation involving tanks on 19 August but larger operations,
such as those on 22 and 27 August, made little progress. From now on,
Gough’s troops were capable only of sporadic and disjointed narrow-front
assaults, five of which were launched in the early days of the next month.
By then Haig had finally stepped in and forbidden exactly that kind of
penny-packet attack. He had also transferred responsibility for capturing the
Gheluvelt Plateau from Gough to Plumer, so it is hard to understand what
Gough was trying to prove.
The results of the August fighting were, for the British, extremely dis-
appointing. There had been no breakthrough, much less any chance of
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e Bat t le for F la n de rs: Sum m e r 1917177

linking up with an amphibious force. Indeed, the final objectives set for
31 July mostly still remained in German hands. By 28 August, Gough had
lost 68,000 men with little to show in return. Within the British ranks,
the official history noted, ‘discontent was general’ at the poor decisions
taken and worse conditions endured. To Rupprecht, British troops were
definitely inferior to their German opponents: they would not advance
under fire and were offering almost no resistance to counter-attacks.
Even officers were surrendering freely. British bravery seemed to have
declined since Arras. A Fourth Army conference reviewed German defen-
sive tactics since 31 July. Unsurprisingly given his success, Sixt decided
that there was no need to change anything. Even through the prism of
attrition, the Germans had little as yet to worry about. Although the rate
of ammunition consumption was worrying, Fourth Army losses between
21 July and 31 August were 58,794, almost 10,000 fewer than the British
had suffered. Early worries that the pace of rotation of divisions in
the front line would need to increase, putting unsustainable pressure on
reserves, were proving unfounded. The average German division was man-
aging to stay in line for twenty-two days or even longer. However grim
the conditions the front-line soldiers were having to endure, British pres-
sure was plainly insufficient. Indeed, Rupprecht was amazed by the slow
pace of British operations and keen that his Army Group should hold
out without begging for aid from OHL. Kuhl estimated that they could do
so until the beginning of October. A heavy attack from the supposedly
exhausted French at Verdun on 20 August only made this more important.
In the course of August, eighteen divisions were rotated in and out of
Fourth Army but there was no need yet to reinforce the defenders at Ypres
in net terms.16
Indeed, as the British push petered out in early September, Ludendorff
felt able to transfer three divisions and twenty-five artillery batteries away
from Fourth Army to support operations elsewhere. By the middle of
September morale among the defenders was extremely high. Rupprecht
and the German high command were convinced that the enemy offensive
at Ypres had fizzled out. They were almost ready to declare victory. When
heavy artillery fire opened up on the southern half of the salient on
15 September, it was at first written off as being just for show. When the
bombardment intensified two days later, however, it became clear that there
was no room yet to relax.To the front-line commanders on the spot, another
major British assault seemed imminent.17
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

17
The Battle for Flanders
To Passchendaele

O n the morning of 20 September, Kuhl arrived back at headquarters

from an overnight visit to Oberste Heeresleitung (OHL). He was
astonished to find a full-scale battle underway. The campaign had kicked
back into life. What became known as the Battle of the Menin Road Ridge
began at twenty to six that morning when nine British divisions attacked
on a front 9 kilometres wide, supported by some three times the weight of
fire laid down on 31 July. German after-action reports later stressed the
sophistication of the attackers’ fire-plan but expressed pride that machine-
guns, deployed in depth, had halted the British assault short of the artillery
positions. The Germans, however, were merely stopping the British from
something they had no intention of doing in the first place. General
Sir  Herbert Plumer had taken over the main British effort and made
two changes to British Expeditionary Force (BEF) assault methods. First,
he adopted looser infantry formations to avoid the thick concentrations of
troops which had been so vulnerable to enemy fire, for instance at Messines.
Second, he chose not to set objectives deep in the rear. Experience had
shown that doing so risked the attackers losing momentum and outrunning
their artillery support, leaving tired troops vulnerable to counter-attacks.
Instead, he opted for shallow objectives, less than 1,500 metres deep and well
within his own artillery range. He aimed to overrun the thin enemy front-
line garrison, seize a manageable bite of the enemy defences and rapidly
consolidate what he had grabbed, then defeat the inevitable counter-attacks.
This is in fact what happened on 20 September. The whole p­ rocess would
then be repeated a few days later, with another operation following soon
after, and so on. The British intention, therefore, was to grind, rather than
burst, through the German defences, chewing up defenders as they went.1
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le179

Plumer’s next operation, the Battle of Polygon Wood on 26 September,

reinforced the effectiveness of the new British methods. Plumer had trouble
moving artillery up to new positions so he reduced the scale and ambition
of his operation, narrowing the frontage to 8,000 metres and setting even
shallower objectives, not much more than 1,000 metres deep. The British
broadly achieved these, clearing most of Zonnebeke and Polygon Wood.
Again, German counter-attacks developed too late in the day and broke on
the new positions the attackers had speedily dug themselves into.2
The Germans seemed powerless to prevent the relentless British march
forward and they were getting rattled. In late September, for the first time
in the campaign, OHL decided to reinforce Fourth Army with three fresh
divisions. Less could be done about ammunition consumption, which
remained worryingly high. Even on relatively quiet days, Fourth Army
was firing off more train-loads of shells than the whole Army Group was
receiving daily. During days of battle such as 20 September, they might use
more than twice as many.3
With hindsight, we can see that the Germans had less to worry about than
they thought. It is hard to interpret the Battles of Menin Road Ridge and
Polygon Wood as significant British successes except relative to the disasters
which had gone before. The casualty figures, as always, are heartbreaking.
The two armies together lost 75,000 men in the September fighting, about
half each. Since 31 July the British had lost 86,000 men and had advanced
6 kilometres towards the German rail junctions still over 30 kilometres distant.
How many more men must fall to get that far?4
At the time, however, German commanders were deeply worried that
there was something wrong with their tactics. In a flurry of meetings and
memoranda, they frantically tried to diagnose what was going wrong.There
are two different narratives of how the Germans responded to the threat of
Plumer’s ‘bite-and-hold’ approach.The traditional version was set out in the
official history and has been repeated frequently since. It is flawed but worth
outlining before we look at this in a more fruitful way.
According to the German official history, the Battles of the Menin Road
Ridge and Polygon Wood underlined once more the fundamental flaw
of  elastic defence. Relying on counter-attacks could work, if they were
mounted at the right time. To do so, however, was nearly impossible on
the often devastated, and always lethal, First World War battlefield, where
communication and movement were so difficult. As Rupprecht pointed
out, if reserves were stationed close behind the front line, they were not
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

180 Holdi ng t h e Li n e 1916 –17

only vulnerable to artillery fire, but would also often find themselves in
action before the enemy’s main axis of thrust had become clear. If they were
kept too far back, on the other hand, they arrived too late, when the enemy
had already had a chance to dig in.5 Loßberg proposed reverting to the
old style of holding the front line much more firmly to make an enemy
break-in harder. If ground was lost nonetheless, Eingreif divisions should be
held back to take part in a well-organized set-piece offensive at a later date,
rather than frittered away in improvised attacks at short notice. These ideas
were formalized in orders issued by OHL on 28 September and by Fourth
Army two days later.6
Loßberg’s new tactics failed completely, however, at the Battle of
Broodseinde on 4 October.After no special artillery preparation, twelve British
divisions attacked. One of Rupprecht’s staff officers, Major Prager, described
this as ‘surely the toughest battle so far’, with the British heavy artillery
fire being particularly ‘smothering’. The German front-line garrison was
suppressed and swamped. For example, the 4th Guards Division had
increased the proportion of their strength in the front line as instructed.
When the attack began, the forward garrison was taken by surprise, blinded
by thick clouds of smoke and dust, and overrun. Thickening up the forward
defence in this way, the Guards reported, only increased casualties. The
Germans lost another kilometre of ground and the village of Broodseinde.
For a while, indeed, Rupprecht thought his line was broken. The Eingreif
divisions had to be thrown in to prop up the buckling line and were no
more available for set-piece counter-attacks than they had been under
the previous scheme. The Germans had at last been pushed off most of the
Gheluvelt Plateau, exposing the southern flank of the Passchendaele Ridge.
The British lost 20,000 men.
The next day Rupprecht advised Fourth Army to end the experiment
with holding the front line rigidly and move back to an elastic defence in
depth. On 7 October this policy, the third in two weeks, was written into
orders from both OHL and Fourth Army. From now on, the front line was
to constitute little more than a screen. A small force, based in shell holes
rather than trench lines, would cover a Forward Zone (‘Vorfeld’), between
500 and 1,000 metres deep. This stretched back to the Main Line of
Resistance (MLR; in German, ‘Hauptwiderstandslinie’). In the event of a
major enemy offensive, the forward garrison was to pull back across the
Forward Zone and into the MLR, where the battle proper would be fought,
supported by reserves moving up from deep.7
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le181

According to the German official history, therefore, the move away from
a ‘modern’ elastic defence in depth was largely the idea of one man: Fritz
von Loßberg.When Broodseinde showed that his rigid forward defence did
nothing to help stop the British either, the German army quickly wrote
it off as an aberration and reverted to an elastic defence. This traditional
narrative very neatly suited the German military writers who constructed it.
First, it blamed a single individual, Loßberg, and deflected criticism from
OHL and the rest of the General Staff. Second, it painted a rational and
systematic picture of the German way of working. Most of the historians
concerned, of course, had been members of the General Staff, and many
of OHL, during the war and liked to think of themselves as the brains of
the  army. Third, it reinforced the case for elastic defence, which was an
important tenet of German military thought between the wars.8
This traditional narrative considerably over-simplified what had happened,
however. More importantly, it deflected attention away from the real prob-
lems the Germans were facing in the autumn of 1917. First, although it is
impossible to untangle exactly who first advocated the switch back to a
more forward approach, it was not all Loßberg’s fault. OHL was heavily
involved. Indeed, OHL issued orders calling for a forward defence on
28 September, two days before Loßberg spoke at Courtrai and issued similar
orders to Fourth Army. Second, whatever the orders, it is far from clear that
every front-line unit was able to adjust their tactics in time for that to be a
major factor in defeat at Broodseinde. For every formation which followed
the new instructions, like the Guards, there were others which complained
that it was impossible to follow all the twists and turns of new doctrine. The
119th Division, for instance, was at the front for sixty-seven straight days
from 11 August to 18 October. It pointed out that new orders incorporating
the latest lessons learnt were of only limited use with no opportunity to
train. Third, there were simpler explanations for the problems the Germans
were facing. These spoke far more to the operational level of war than the
tactical. The cumulative effect of attrition was beginning to make itself felt.
Although a net six divisions reinforced Fourth Army in the first ten days of
October, quality was slipping. Rupprecht noted on 5 October that the only
replacement divisions available were either inexperienced and in need of
further training, or those whose confidence had been shaken by repeated
defeats. Even formerly good divisions had been weakened by absorbing
large numbers of recruits. Poor leadership was a problem, too. Rupprecht
felt that even Loßberg temporarily lost his nerve at one point during the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

182 Holdi ng t h e Li n e 1916 –17

Battle of Broodseinde. The commander of Gruppe Ypern, General Alfred

Graf zu Dohna-Schlobitten, had more experience at the imperial court
than he had of the Western Front, let alone commanding at this level.
Rupprecht regarded him as ‘a military non-entity’ and he was sacked from
the hottest seat on the Western Front the day after Broodseinde, after three
defeats in less than a month. The most decisive factor on 4 October,
however, was, as so often during the war, the guns. As Rupprecht noted
in  his  diary the next day, even though Fourth Army fired off a record
thirty train-loads of ammunition, British firepower was overwhelming. The
Germans had no answer.9
The response to Plumer’s ‘bite-and-hold’ operations of September and
early October teaches us two interesting broader lessons about the German
army. First, the German General Staff sought tactical solutions to what was
in fact an operational challenge. This was not because they were stupid, but
because they knew they had no operational solution to offer. Ludendorff,
for instance, as he warned of the dangers of high enemy tempo and the
problems this was causing on the railways, explicitly spoke of the need to
use tactics to reduce attrition. The tendency of the German military to seek
military solutions to political problems and to attempt to offset operational
weakness with tactical brilliance is a recurring theme in its history from
Schlieffen to Stalingrad. Second, it highlights the intellectual arrogance of
its commanders. Men such as Ludendorff were convinced not only that a
single solution to their difficulties existed but also that they could find it.
This had two important consequences. First, it blinded them to the
possibility that in fact there might be no panacea, and that different situ-
ations might require different responses. Second, it meant that doctrine
formulation became increasingly centralized and dogmatic. While this did
not altogether preclude bottom-up innovation, it did restrict the initiative
of subordinate commanders and render the Germans predictable to their
enemies. Although neither Rupprecht nor Kuhl was comfortable with this
tendency, it proved difficult to overcome. On 2 October Rupprecht visited
Fourth Army and told Sixt to avoid over schematic approaches and over-
centralization. He was to allow his subordinates more latitude for initiative
so they could operate less predictably. When Rupprecht got back to his
headquarters at Mons, he was infuriated to find there a copy of a highly
detailed and schematic order just issued by Loßberg, setting out an artillery
defence plan which even laid down the deployment of individual batteries.
Two days later, Rupprecht again told Fourth Army to allow the corps
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le183

commanders more scope to use their own initiative and to avoid employing
the Eingreif divisions in too formulaic and predictable a manner. Kuhl, too,
was worried: ‘the new tactics which Loßberg invented were much too
­schematic. That is in general Loßberg’s big mistake. We must leave more
freedom.’ ‘Loßberg’, Kuhl noted, ‘is a great organizer, but not much of a
tactician or strategist.’ Rupprecht agreed, noting in his diary that the new
tactics introduced on 7 October, which relied on close coordination between
infantry and artillery as the forward garrison deliberately gave ground in the
face of an enemy attack, displayed Ludendorff ’s lack of understanding of the
difficulty of communications on the modern battlefield: ‘it would be better
if he did not bother himself with intervening too much in the details of
tactical methods’, he remarked. Further, the continuous stream of tactical
instructions was proving bewildering: ‘There is no cure-all. A pattern is
harmful. The situation must be dealt with sometimes one way, sometimes
another.’ This tension between central and local solutions and control was
not new in 1917, although we see it particularly clearly around the Battle of
Broodseinde. We saw it at work on the Somme and before, and we shall see
it again in 1918, for it was never resolved.10
Four days of heavy and persistent rain followed the Battle of Broodseinde.
Both Gough and Plumer proposed closing down the Flanders campaign
but  Haig was determined to capture the Passchendaele Ridge. The next
attack was set for 9 October and would involve eleven French and British
divisions. In normal conditions, to launch another major attack after just
five days would be ambitious. In the mud and wet it proved calamitous.
Even to move troops and guns up proved extremely challenging. The night
before Zero, some of the assault battalions took nearly ten hours to march
4 kilometres. Divisions which had enjoyed support from ninety guns on
20 September could now call on just twenty-five. The barrage was weak or
invisible and ‘no previous attack organized by the Second Army in the war
had such an unfavourable start’, wrote the official historian. In many places,
the main attack did not even reach the German first line, much less pene-
trate it. Most of the defending divisions did not even have to call on their
reserves. Total British casualties are unknown but heavy: just three of the
assault divisions alone lost nearly 7,000 men between them.
Although the Battle of Poelcappelle was a clear German tactical victory,
the operational strain on the defenders was worsening. Rupprecht was
growing increasingly worried. The first ten days of October were the
bloodiest single period of the whole Flanders campaign for Fourth Army.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

184 Holdi ng t h e Li n e 1916 –17

It lost 29,000 men, or 17 per cent of its combat strength. Only September
1916 on the Somme had seen a higher casualty rate for any of Rupprecht’s
armies. The average battalion in late July had been 713 men strong. By
11  October, it was down to 621. Rupprecht noted that enemy artillery
superiority was so stifling that nothing the Germans did seemed to help.
‘Our most effective ally’, he suggested, ‘would be rain.’ It was still just possible
to find enough replacements but if the French renewed offensive operations
further south, OHL would be unable to help further and Rupprecht’s Army
Group would have to stand alone. It might be necessary to pull the line
back to dodge enemy attacks. Morale was also suffering, both in the trenches
and at headquarters. To Kuhl, both Loßberg and Sixt seemed depressed and
borderline unfit.11
Plumer remained committed to a high tempo of operations. The British
decision to attack again on 12 October, however, was founded not in any
understanding of the level of worry Rupprecht and his men were suffering,
but in much vaguer intelligence about German home-front morale and a
misunderstanding of the situation on the ground. By the time Plumer and
Gough realized that the Battle of Poelcappelle had not been the success they
had first imagined and that Passchendaele remained both further away and
more heavily defended than they thought, it was too late. The British enjoyed
even less success on 12 October, in the First Battle of Passchendaele, than
they had three days previously. Poor weather, deep mud, and insufficient
artillery support all contributed to disaster. In the rare places progress was
made, it was measured in metres. Although Rupprecht believed that the new
Forward Zone tactics had worked well, it is clear that he was misinformed
about the extent of their use and over-estimated their impact. German tactics
were not the decisive factor. Not only did different German corps try to
employ different methods; even divisions within the corps went about things
in a variety of ways. In the centre of the British attack, Gruppe Ypres in
­theory evacuated their deep Forward Zone when the enemy attacked, while
on either wing Gruppe Dixmuide and Gruppe Wytschaete relied on a more
traditional rigid forward defence. Even within Gruppe Ypres, divisions such
as the 195th Infantry Division were unable to use the new tactics properly.
Nonetheless, 13,000 British troops became casualties for no significant gain.
Between 11 and 20 October, on the other hand, Fourth Army lost fewer than
10,000 men, a third of the casualties in the previous ten days.12
Such was the nature of 1917 communications, though, that the initial
reports reaching Rupprecht on 12 October were worrying. The Passchendaele
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le185

Ridge, it seemed, was under serious threat. The Crown Prince moved two
divisions up from reserve to restore the situation. By the next day it was
clear that the situation was under control and Rupprecht felt free to go off
hunting. Congestion on the railways was now his most immediate concern.
The army was running some 900 trains daily and the network was so
overloaded that no fresh reinforcements could reach him before 18 October.
Still, the front-line troops were adequately rested, reserves were relatively
plentiful, and there was no need for worry, in the short term at least. Observers
such as the chief of staff of Gruppe Wytschaete, Lieutenant-Colonel Albrecht
von Thaer, who had been very downbeat in later September, were pleased
that things now seemed to be going well. Ludendorff was actually upbeat,
telling Rupprecht that within two weeks they would have won the war.
Rupprecht could not understand or share his optimism. The British, he felt,
were quite capable of hammering away for another four to six weeks. After
all, they had only suspended the Somme campaign when the weather made
it impossible to carry on.13
A few days of relative quiet followed. Kuhl and the staff took advantage
of the lull to begin planning for 1918, as we shall see in Chapter 19. The
British restricted themselves to small-scale operations designed to keep the
defenders busy and prepare for the next big push. The first stage of this
unfolded on 26 October, with a main attack launched by two Canadian
divisions aiming at the village of Passchendaele on its ridge. On the first day
they made only 450 metres, about half of what they had hoped. Rupprecht
noted another defensive victory. Although, as he also observed, ‘the state of
the ground defies all description’, on 30 October another Canadian push
saw them drive another 800 metres and reach the outskirts of Passchendaele
itself. Five of the six Canadian battalions which spearheaded the assault
suffered 50 per cent casualties.
Reinforcements and artillery ammunition were becoming causes for con-
cern in the German ranks again. By 29 October Rupprecht’s Army Group
had no reserves left and even OHL could not be relied on to provide more
than another two divisions, at best. Consumption of ammunition was run-
ning so high that Rupprecht worried he would run out within eight days.
Finally, on 6 November the Canadians cleared the village of Passchendaele.
The German defenders were blasted out of their positions by British guns.
Yet again, communications broke down. It took three hours for counter-
attack orders to reach the reserves, by which time it was too late to have any
hope of success. Although, as Rupprecht said, ‘the loss of Passchendaele is
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

186 Holdi ng t h e Li n e 1916 –17

painful for us’, lack of reserves due to transport delays meant the idea of a
full-scale attack to recapture the ridge had to be abandoned. On 10 November
the Canadians launched a further operation to consolidate their hold. This
proved the last day of major fighting in the Second Battle of Passchendaele.
The Canadian Corps lost nearly 13,000 men between 26 October and
11 November.14
On 12 November Fourth Army reported that there were no signs of
preparations for a further attack. Rupprecht could begin to hope that the
Battle for Flanders was over. This was confirmed three days later when
intelligence noticed French troops being replaced in line by British and
Belgians. OHL had already begun to redeploy forces away from Flanders
and in all thirteen divisions were pulled out of the line for rest and retraining
during November.15
The fighting in Flanders of the summer and autumn of 1917 devoured
manpower and matériel at a ferocious rate. Hundreds of thousands of
men fought on both sides and battle casualties probably totalled nearly
half a million. Almost every division in the BEF fought at Ypres in 1917,
as did nearly half the German divisions in the west. German casualties,
at  the worst estimate, were 217,000, British losses either 238,000 or
244,000, while the French lost about 8,500 men. The German Fourth
Army fired 18 million shells or nearly 2,000 train-loads of ammunition;
the British used perhaps six times as much. The emotional impact of
the  horrendous conditions cannot be over-estimated. Even Ludendorff
realized that, for the soldiers fighting on the Passchendaele Ridge in
the last stages of the campaign, ‘it was no longer life at all. It was mere
unspeakable suffering.’ More soberly, the Bavarian history suggested that
‘in probably none of the big defensive battles did the soldier suffer as
much as in Flanders. Nowhere offered so little protection from enemy
fire and the weather.’16
The British official history argued that ‘no great victory had been won; it
was . . . an ordinary, not an annihilating victory’. Clearly, the most ambitious
hopes of Haig and those around him were unfulfilled. No breakthrough
took place. The German army did not collapse. The rail junctions at Roulers
and Thourout remained firmly in German hands. So did the Belgian coast.
The British official historian claimed, however, that Rupprecht’s diary
proved that the German army in Flanders was broken and contemplating a
major retreat. At a time when it was politically vital to show shaky allies
that Great Britain remained committed to the war, the BEF had kept the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le187

initiative, working and bleeding hard. More directly, the British had, he
argued, distracted the Germans and so ensured the security of the wobbly
French army. The British had advanced as much as 8 kilometres and cap-
tured the crest of the high ground encircling Ypres. This reduced German
ability to shell into the salient. And the damage done to the German army
both compelled Ludendorff to gamble on the success of offensives early in
1918 and weakened the forces he had at his disposal for those attacks, thus
ensuring they failed.17
From the German perspective, things look rather different. Army Group
Crown Prince Rupprecht’s War Diary records the battle as ‘a big defeat
for the enemy, a major victory for us’. The official history agrees, describing
Flanders as a ‘major German victory’ born of outstanding defensive
organization by Fourth Army. The British offensive had been halted on the
third of Loßberg’s five lines. True, the German army had been worn down,
but in some ways the impact had actually been rather less than that of pre-
vious battles. The absolute number of German divisions sucked into the
fighting was down by a quarter from the Somme, for instance (71 versus
95.5) and the proportion of the army in the west involved fell to under half
from 70 per cent. The impact on other fronts was minimal: seven German divi-
sions left the Western Front in July 1917 and were not replaced until the end
of November, after the Third Battle of Ypres had been closed down. German
casualties in Flanders in the six months from June to November were roughly
the same as had been suffered in the Anglo-French attacks of April–May, less
than half those on the Somme and less than in the first six months at Verdun.
As Alexander Watson has demonstrated, ‘although by the end of the Ypres
offensive the German army’s confidence had been shaken, the bulk of the
force was not yet ready to collapse’, not least because it appeared to German
intelligence that the British were experiencing similar problems. Incidents
of indiscipline were on the rise, but more often involved men being
transferred from Russia than soldiers who had been through the Flanders
campaign. Soldiers’ complaints were about lack of leave and poor conditions
on badly organized troop transports, rather than a rejection of the war or
their duty. While it is true that Rupprecht discussed withdrawal in his diary,
he did so only in the context of a temporary pull back to unbalance the
British before counter-attacking them.The collapse of Russia and imminent
arrival of a large American contingent were at least as important as drivers
of Ludendorff ’s decision to attack the next spring as his wish to move from
defence to offence.18
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

188 Holdi ng t h e Li n e 1916 –17

It seems perverse to argue that a campaign which achieved none of its

strategic objectives was any kind of British victory and few modern historians
have dared to do so. Although, as Gary Sheffield argues, the offensive did
allow the Entente to retain the initiative on the Western Front, the impact
on German operations elsewhere was extremely limited. Further, J. P. Harris
has pointed out that the worst of the French army’s problems had in fact
been resolved well before 31 July. A wave of mutinies had briefly gripped
the French army in May but was largely resolved by the end of June. French
troops proved effective in the attack at Ypres itself, at Verdun in August and
at La Malmaison in October 1917. The lag between the Battle of Messines
and the beginning of the main attack, however unavoidable, significantly
undercut at least one of the strategic rationales proposed for launching the
Ypres offensive. The decision to continue attacking, given the scale of the
problems involved, the high cost, and scanty returns on the blood and
treasure invested, is harder to defend. As Jim Beach has argued, weak
understanding of the state of the German army played a part but was
not the determining factor: GHQ Intelligence tended to follow Haig’s
­optimism, rather than lead it. Poor intelligence, primarily strategic, merely
lent weight to a campaign which already had its own political and tactical
momentum.19
Did that momentum need to extend as far as the capture of Passchendaele
village, with all the horror that entailed? The British salient thus created was
highly vulnerable and had to be rapidly evacuated when the Germans went
on the offensive in spring 1918. Robin Prior and Trevor Wilson have argued
that it was quite possible to suspend the campaign at any point. One option
might have been to pull back to the Pilckem Ridge, won in the early days
of the attack, although the morale and political costs of doing so would have
been high. Nonetheless, the military logic, once the British had started to
clear the high ground overlooking Ypres, of pushing on to finish the job is
clear. Indeed, Rupprecht saw it, too: ‘it can be said of Passchendaele that it
forms the keystone of the entire high ground east of Ypres and that is why
the enemy fought so hard to possess it’, he noted on 16 November. If you
look back at Ypres from Passchendaele, then walk around to the back of the
church and look out over the plain stretching away to the east below you, it
is easy to see what Rupprecht meant.20
Even if it made sense at some level to push on to Passchendaele, however,
that does not mean that the tactics used to do so were sensible. Prior
and Wilson argue that the British had failed to understand and employ the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

T h e Bat t le for F la n de rs: To Passch e n da e le189

‘bite-and-hold’ tactics which offered a tactical solution. They have a point.

Certainly in October Plumer was rushing his attacks, not allowing enough
time for adequate artillery preparation. He was out of touch with conditions
on the ground. J. P. Harris, on the other hand, has suggested that in fact, for
a frontal attack into the teeth of alert defenders, British losses were relatively
low and bespoke increased efficiency of tactical method, at least compared
to, say, the Somme. The casualty returns were probably, however, reduced as
much by the low numbers of attackers getting into action as by skill in
avoiding enemy bullets and shell fire.21
The central point about the Entente in late 1917, though, is a chilling and
dispiriting one, which David Stevenson has identified.The Allies confronted
a terrible dilemma. Fighting on as they had at Third Ypres ‘promised to
cripple the British army as surely as it had crippled the French and Italian
ones’. On the other hand, the methods which had proven successful at
Messines and La Malmaison would be prohibitively expensive in lives,
matériel, and destruction of the land they were supposed to be liberating.
Doing nothing was hardly an option, but doing anything seemed to invite
disaster, even if successful. Neither the Entente in general, nor the BEF in
particular, seemed any closer to finding a way to win the war on the battle-
field than they had been in the three previous years of fighting. Both sides
continued to experiment with new tactics to break the Western Front
impasse, though, as we shall see when we explore the Battle of Cambrai in
Chapter 18.22
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

18
Cambrai

A lthough winter was near and further sustained operations were unlikely,
Rupprecht never expected the end of the Flanders campaign to mark
the end of the fighting season. Local enemy attacks remained likely and
indeed on 19 November Second Army reported that they were picking up
indications of an assault involving tanks at Havrincourt, a previously quiet
sector of the Hindenburg Line 15 kilometres south-west of Cambrai.
Rupprecht and Kuhl reinforced the division holding the line there and
moved reserve battalions closer up to the line.1
At quarter past seven the next morning, 20 November 1917, six divisions
of the British Third Army under Lieutenant-General Sir Julian Byng
attacked on a 9-kilometre front stretching from the Canal du Nord, just
west of Havrincourt, down to Banteux. There had been no preliminary
bombardment. Instead, 378 tanks cut gaps in the wire for the infantry to
rush through. Over a thousand British guns laid down an effective lifting
barrage and intense counter-battery fire. The Germans had three divisions
holding the front, with a fourth in reserve at Cambrai.2
The first reports to reach Rupprecht caused little real concern. He heard
that the defenders had been caught by surprise and the front line overrun
in places.The tanks had proved surprisingly effective, moving quickly across
the dry and firm ground, and ploughing up the hundred-metre-deep belts
of wire in front of the German lines. They had demoralized some of the
defending infantry, too. ‘I never expected tanks would be able to break
through so fast’, wrote Rupprecht. Nonetheless, in many places the enemy
had been repelled or had gained only toeholds in the German lines. Only
in one spot had he driven beyond the front line. Soon, however, the news
grew more serious. Both the front and support lines had been lost. Brave
German resistance at Flesquières, coupled with timely counter-attacks by
the reserve division, had absorbed much of the sting of the British attack,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ca m br a i191

but the unfinished reserve line remained under threat and 126 guns had
been overrun. The Germans would have to evacuate Flesquières that night
before it got completely cut off. The British had crossed the St Quentin
Canal at Masnières and Indian cavalry had been spotted approaching the
outskirts of Cambrai. Ludendorff accused Rupprecht of getting distracted
by the situation in Flanders, a charge Rupprecht refuted in his published
diary, although the manuscript version admitted Ludendorff was right.
Clearly, a powerful counter-attack would be required to restore the situ-
ation and Rupprecht rushed reinforcements to Second Army. Six divisions
began arriving the next day and in the course of 21 November a new
defensive line was stitched together to contain further British thrusts.3
The dramatic British success of 20 November was seized on as great news
in Britain. At last, it seemed, a clear-cut victory had been won. The bells of
churches across the country, including St Paul’s Cathedral, rang out, the first
time they had done so in this manner since the Boer War. King George V
congratulated Sir Douglas Haig and promoted Byng in the field. First
reports emphasized the role played by massed armour in securing this vic-
tory. This theme was picked up by cheerleaders for tanks between the wars,
by men as influential as Winston Churchill and Basil Liddell Hart. More
recently, modern historians have questioned their importance. For instance,
Bryn Hammond’s excellent book on the battle argues convincingly that it
was no complete victory, much less the ‘first great tank battle’ and a victory
for the tank. In his view, what success the British enjoyed on 20 November
was based not on any single weapon, but on the convergence at last of a
whole range of strategic, political, scientific, technological, and tactical
­factors, allowing an integrated combined-arms approach.4
German views of why the British succeeded on 20 November, however,
put more stress on armour once more. The key issue, according to after-
action reports and the diaries of Rupprecht and Kuhl, was that tanks had
enabled the attackers to achieve surprise. German reports on tanks should be
treated with a little suspicion, however. The tank was, after all, a convenient
scapegoat to explain away failure, distracting attention from defensive short-
comings such as insufficient artillery support. Neither Kuhl nor Rupprecht
had shown much concern about armour before. Indeed, even this time they
seemed comfortable that they had discovered an effective counter-measure,
using lorry-mounted anti-aircraft guns in a mobile anti-tank role.5
Haig had originally intended to suspend operations quickly if progress
stalled. His troops found themselves in a valley, however, overlooked by the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

192 Holdi ng t h e Li n e 1916 –17

high ground of Bourlon Wood. Staying where they were would be difficult
and painful. Either they must go on and take Bourlon, or they would have
to pull back to the next defensible position along the Flesquières Ridge.
As in the final stages of the Third Battle of Ypres, tactical considerations
were permitted to warp the operational design. Bourlon Ridge proved
as magnetic as Passchendaele. Several days of bitter fighting rolled to and
fro in the wood and outskirts of the village. The longer the fighting went
on, the less palatable a retreat became. British momentum slowed as logis-
tics came close to breakdown, and attacks became increasingly s­pasmodic
and uncoordinated. The German line held. Remarkably, as late as
25  November the British were still hoping for a breakthrough and sent
cavalry up to exploit. Left standing on an exposed hillside while they
awaited their chance, the cavalrymen achieved little except to offer German
gunners ­target practice.6
German commanders had already begun planning a full-scale counter-
attack. As early as 21 November Rupprecht ordered Second Army to prepare
to pinch out the new British salient from north and east. Kuhl suggested
that the main attack take place on the left-hand, southern pincer. Presumably
his idea was to avoid the difficult ground of Bourlon Wood and a frontal
attack on the alert British troops already fighting there. Instead, he would
drive from east to west into the flank of the bulge, across open downland.
When Rupprecht and Kuhl visited Second Army the following morning,
Marwitz outlined a plan along exactly these lines. Kuhl was impressed by
the calm atmosphere in Second Army headquarters.7
Ludendorff, in contrast, remained flustered. Once more he could not
refrain from interfering. To Rupprecht he seemed ‘very nervous. On the
telephone he enquired into a thousand details which had a very disruptive
effect.’ For instance, even as fighting raged, Oberste Heeresleitung (OHL)
was demanding to know at once how many batteries of which calibre guns
were firing on this target or that. Ludendorff was so worried by the
British success that he insisted Flanders be stripped of reserves to meet the
new threat, a step Rupprecht did not think necessary. Ludendorff also
involved himself closely in planning the counter-attack. On 27 November
he, Rupprecht, and Kuhl attended a meeting at Second Army headquarters
in Le Cateau to discuss the coming operation. Rupprecht, in particular, was
excited by the prospect of breaking in and rolling up the British front from
south to north. It offered ‘the best opportunity we’ve had for years’ to give
the British a real knock. A total of fourteen German divisions were involved.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ca m br a i193

On 30 November, after just one hour of artillery bombardment, two

corps were to launch a surprise attack into the vulnerable eastern flank of
the  British salient. As the infantry advanced, it would use infiltration or
‘stormtroop’ tactics, bypassing enemy strongpoints and leaving them for
subsequent waves to mop up. The attackers would infiltrate deep into enemy
lines and seize the village of Metz-en-Couture. Further north, a third corps
would drive south either side of Bourlon Wood. The task here was harder.
There was no chance of surprise because the enemy would be on the qui
vive after recent fighting. Consequently, the northern attack would be
delayed until two hours after the southern force had jumped off, in the
hope that resistance would be beginning to weaken by then. Once Metz-en-
Couture had fallen, the British in Flesquières and Havrincourt would be
isolated and they would have to evacuate the salient they had won on
20 November. At a minimum, Rupprecht’s aim was to recapture the sections
of Hindenburg Line lost the previous week.8
In the event, the attack in the south overran the thinly stretched British
front line. Desperate defence and a timely British counter-attack, carried
out with style by the Guards Division, eventually brought the Germans to
a halt, but not before some units had driven up to 6,000 metres into the
British positions. Metz remained out of reach, however, as did the Flesquières
Ridge. The standard of cooperation between German infantry and their
trench mortars was high. ‘Stormtroop’ tactics clearly had merit, although
Kuhl identified plenty of areas for improvement before next time. The fail-
ure of the northern attack, which substituted mass for tactical finesse, rein-
forced the point that surprise was a more effective weapon than weight of
numbers. Here, British artillery had shot the assault columns to pieces as
they crossed open ground and progress was slight.Vicious fighting on both
flanks over the next couple of days produced further heavy casualties but
little change to the overall situation. Ludendorff suspended offensive operations
during the evening of 1 December. Forty-eight hours later, Byng gained
Haig’s approval to withdraw from Bourlon Wood and Marcoing back to a
more defensible winter line along the forward slope of the Flesquières
Ridge. This manoeuvre was carried out over the next few days and the
fighting died away.9
By every measure, honours at Cambrai were even. Once the front had
re-stabilized by mid-December, the British had gained 20 square kilometres
of ground on the left, but lost 16 on the right. British casualties between
20  November and 8 December totalled 44,207, while the Germans lost
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

194 Holdi ng t h e Li n e 1916 –17

41,000 men, 14,000 of them in their counter-attack. Both sides could and
did claim a sort of victory. As Rupprecht noted, his men had scored the big-
gest offensive success over the British since the Second Battle of Ypres two
and a half years earlier.10
Neither side, however, had accomplished what they had hoped with their
attacks and the British had come close to defensive disaster on 30 November.
As a result, commanders on both sides had some explaining to do. Sir Douglas
Haig took full responsibility for the decision to carry on bashing away against
Bourlon. General Byng was rather less gallant, blaming the collapse of his
eastern flank on poor training, weak leadership, and low morale among his
junior officers and men. Although his version of events was upheld by a
Court of Enquiry early in the new year, historians subsequently have tended
to place more blame on the senior commanders, including Byng, who had
under-estimated the German capacity to counter-attack, had drawn up no
proper defensive plan, and had allowed themselves to be taken by surprise.
Haig’s chief of intelligence, Charteris, was replaced within a couple of weeks,
although, as Jim Beach has demonstrated, Cambrai was more pretext than
cause for the removal of an unpopular officer who had already lost the con-
fidence of many in both France and Whitehall. Further, the 30 November
attack, Beach points out, was inherently ‘difficult to anticipate because of the
relative haste with which it was improvised and because it was hard to dis-
tinguish between defensive reinforcement and offensive preparation’.11
The reasons for the failure of the German northern attack on
30 November were clear: lack of surprise, an alert enemy, and poor tac-
tics. Events on the southern wing were more interesting. According to
Ludendorff ’s memoirs, the counter-attack there was less successful than it
might have been because an otherwise good division, rather than pushing
on to its objectives, stopped to ransack an enemy supply depot. Rupprecht
blamed Ludendorff instead. At the 27 November conference, Ludendorff
had emphasized the importance of the Flesquières Ridge position, marking
up a map with charcoal crayon to show how it might be captured from the
south-east. According to Rupprecht, Marwitz replied ‘I will take special
notice of this sketch’ and subsequently, rather than driving deep for Metz-
en-Couture and bypassing enemy strength as Rupprecht’s plan envisaged,
directed a shallower advance towards Trescault which ran directly onto the
British defences. Ludendorff ’s explanation, which foreshadows disciplinary
problems encountered among hungry troops in the spring 1918 offensive,
seems convenient. When he wrote his memoirs he was an ex-general trying
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ca m br a i195

to shift the blame for defeat from his own decisions to a lack of home-front
support and supplies. That said, there is evidence that at least some troops
were off scavenging for food in British dugouts when the enemy hit back.
Similarly, if Rupprecht knew that Marwitz had shifted the axis of attack to
Trescault from Metz, why did he not tell him to revert to the original plan?
As at Neuve Chapelle in 1915, he seems to have felt constrained not to
interfere with his subordinate’s decisions. In any case, the Germans never
got close even to the tricky Flesquières Ridge position, much less Trescault
or Metz.
There are simpler explanations for the incomplete success of the German
southern wing, as the after-action reports themselves demonstrate. First, as
we have seen, resistance never completely collapsed; a defensive line was
eventually improvised; and some local British counter-attacks worked well.
Second, the German thrust was short of resources, and especially manpower.
Lack of reserves meant too much was asked of the assault divisions, which
inevitably soon became burnt out.There was no second wave ready to leap-
frog through and continue the assault, which prevented exploitation of early
success. Third, what reserves were to hand were often held back by corps
headquarters located too far to the rear to keep up with events, rather than
under the direct command of the men leading the attack. Fourth, staff work
on logistics was sometimes poor and caused confusion and delay. For
instance, five divisions were all trying to draw ammunition from the same
supply depot and three different divisions were routed through a single
crossroads, causing traffic jams so snarled that ammunition columns were
lucky to move 1 kilometre in an hour.12
The Battle of Cambrai, in the words of the British official historian, ‘had
little effect upon the general situation’. It made no difference to the stale-
mate on the Western Front and neither helped nor hindered the situation
across the Alps after the Italian defeat at Caporetto in October.The material
impact on both sides was approximately equal. If 20 November provided
any boost to British morale, the German counter-attack ten days later undid
it. Similarly, any damage to the defenders’ confidence caused by the British
assault was offset by the demonstration that the German army remained a
potent attacking force which could accomplish much, even without the
mechanical panoply deployed by the British. Politically, both sides could
claim at least some sort of a victory to end the year.13
The importance of Cambrai lies not in what it achieved, but in what it
promised for the future. On 20 November the British had, for the first time
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

196 Holdi ng t h e Li n e 1916 –17

on a large scale, successfully pulled together a range of technological, organ-

izational, and tactical advances, integrating artillery, armour, and airpower
with infantry to fight in a new and recognizably modern fashion. Equally
the Germans had, in places, showcased a style of warfare, first advocated as
early as 1915 but now being employed more widely, which became known
as ‘stormtroop’ tactics. Assaults were to use speed, infiltration, and close
cooperation between guns, infantry, and aircraft to overrun defences and
unlock the stalemate of the Western Front. The urgent need now, for both
sides, was to learn the lessons they would need to refine their methods ready
for the fighting of the year ahead. On 4 December, Rupprecht stressed the
importance of learning from the recent offensive. It was, after all, his Army
Group’s first large-scale attack. ‘The lessons learnt . . . are extremely valuable.
They must be integrated at once into the winter training of staffs and troops
and be incorporated into the design of further offensive operations. An
accelerated working out of lessons learnt is thus pressing.’ Commands at all
levels were to write up after-action reports. Rupprecht and Kuhl set out a
list of topics they wanted information about within days. Divisions were to
report on each arm, both independently and in combination, the preparation
and organization of the attack, staff work, communications, and mobility.
Gruppe and army headquarters were told to focus on preparations, organ-
ization and leadership, transport, secrecy, cooperation with neighbouring
formations, the dissemination of orders, intelligence and communications,
and supplies and munitions. They were to concentrate on hitting the most
significant points, rather than capturing every little detail, and not to worry
about style or presentation. Speed was essential because the German high
command had already decided that they would strike in the west in the
spring. Cambrai had shown that, if the German army was to be ready to
fight a mobile war, a winter of intensive analysis and training lay ahead.14
Kuhl summarized the reports he received in three memos he issued
­during December, making a series of recommendations. He highlighted the
importance of achieving surprise and the need for a follow-up wave to
maintain momentum and exploit success. Units must advance at all costs
and pay no heed to the situation on their flanks. Reserves should be at
the disposal of the commanders leading the assault, not held back under the
grip of senior commands. Decisions would need to be made quickly, so
the power to make them must be delegated to the man on the spot, not
jealously held onto higher up.‘In many cases we must return to the principles
which worked well at the beginning of the war’, wrote Kuhl, appealing to
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ca m br a i197

the past to show the way to the future.The memos also addressed a number
of detailed tactical issues which needed to be dealt with during winter
training. For instance, units should attack in small mobile assault squads, not
in the long skirmish lines laid down in the pre-1914 manuals.15
The lessons from these and other reports, for example from the Battle of
Caporetto and Haig’s despatch on his 1917 operations, were incorporated
into the new doctrine publication The Attack in Position Warfare, written
by Captain Hermann Geyer and published on 26 January 1918. This was
intended primarily to provide the operational framework within which the
stormtroop tactics laid out in a new edition of the Training Manual of Infantry
could work. Geyer put special stress on the need to maintain the attack’s
momentum with quick decision-making. Spearheads were to advance at
speed, leaving centres of resistance and threats to their flanks to be neutral-
ized by the troops coming up in support. Ludendorff shared Geyer’s view
and emphasized momentum in the series of planning conferences held
throughout the early months of 1918. These will be discussed further
in Chapter 19. According to Rupprecht, however, Ludendorff was under-
estimating the difficulties of maintaining momentum. Ludendorff seemed
to think that all that was required was good infantry–artillery cooperation
and smart traffic management behind the lines to keep reserves moving.
He thought divisions could continue in combat for several days at a time.
Rupprecht, on the other hand, believed that twenty-four hours was the
most they could manage. Rupprecht also had broader worries. ‘It is regret-
table that OHL is once more interfering in tactical details which should
better be left to individual commands to determine according to their
experience and requirements’, he complained. ‘I think that armies should
not tie themselves to the standardization laid out in the memorandum but
rather should conduct themselves as circumstances dictate.’16 1918 would
fully bear out Rupprecht’s concerns.
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

PaRt IV
Year of defeats 1918
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

19
Planning the Spring Offensives

I n October 1917, even before the fighting in Flanders had finished, the
German high command began turning its attention to the year ahead.
On paper, Germany had three options. She could settle for a negotiated
peace. She could stay on the defensive in France and Belgium, while elim-
inating other enemies such as Russia and Italy. Or, she could attack in the
west to drive Britain, France—or both—out of the war. In reality, her choice
was considerably more limited. First, as David Stevenson has shown, there
was no possible peace deal on the table. Public opinion in all the combatant
nations had hardened over time as the treasure and blood flowed. The pros-
pect of compromise was more distant than ever. No-one was more intransi-
gent than Ludendorff. In his mind, anything less than victory meant defeat,
and defeat probably meant revolution back home. His ambitions were
almost limitless, but discounting these, he was even at a minimum not pre-
pared to concede military control over Belgium. Neither Chancellor Georg
Michaelis, nor his successor, the Bavarian Georg von Hertling, was able or
willing to pursue a line independent of Ludendorff and Hindenburg. On
the other hand, for London, Paris, and probably Washington a full restor-
ation of Belgian sovereignty was the sine qua non for any agreement. There
was no common ground on this issue and consequently no basis for a settle-
ment. The German government rejected out of hand even President
Wilson’s ‘ostentatiously moderate’ peace proposals, the Fourteen Points of
January 1918. Second, Britain, France, and the United States were apparently
determined to continue the war to the end, Russia or no Russia on their
side. Ludendorff had to find a way to defeat them. All his senior generals
agreed that the only way to do so was by attacking. To sit on the defensive
would achieve nothing except give the Americans time to build up their
strength in Europe and shift the odds further against Germany.1
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

202 Y e a r of De f e ats 1918

The question, then, was: how, where, and when to attack? On 23 October,
his chief of operations, Major Wetzell, recommended a knock-out blow
against the British before the Americans could arrive in force, using thirty
divisions. Two days later Rupprecht and Kuhl submitted a memorandum
so  much less ambitious that Kuhl was ‘curious what OHL [Oberste
Heeresleitung] will say about it. Probably that we’re washed out. One must
tell the truth, though.’ They argued for only limited offensives in the short
term. Working on the assumption that they could count on no reinforce-
ments from the Russian front and that the British would continue their
offensive in Flanders, they argued that the best way to avoid the pain of
another attritional battle in the new year was to pull back deeper into
Belgium before launching an upper cut from Armentières into the flank
and rear of the advancing British Expeditionary Force (BEF) around Ypres.
Ludendorff asked for a detailed study of the forces that would be required
for such an attack, noting that events in other theatres would determine its
practicability. Rupprecht was worried that Ludendorff was concentrating on
the tactical issues involved and losing sight of the need to inflict a decisive
operational, or even strategic, defeat on the enemy. He also noted that much
would depend on the spring weather. If the River Lys stayed high and the
valley was marshy, the attack could not work.2
On 11 November, Ludendorff visited Rupprecht’s headquarters at Mons
to discuss offensive plans for 1918. Also present were Kuhl and a clutch of
staff officers from OHL and elsewhere, including Colonel Friedrich von
Schulenberg, chief of staff to the Prussian Crown Prince. The Bolshevik revo-
lution in Russia, underway now for a few days, had not yet delivered a cease-
fire on the Eastern Front but held out the prospect of freeing up troops and
artillery for the west, so Kuhl proposed a variant of his 25 October idea. Some
thirty-five divisions would strike north-west between the Lys and the La
Bassée Canal, cutting up through Bailleul and Hazebrouck and driving for the
Channel at Calais. To ease the task of slicing the BEF in two, it might be
necessary to pull back in front of Ypres and draw the British eastwards. The
British breastworks would be easy to break down and the Portuguese contin-
gent holding the line was especially vulnerable. The only drawback with this
plan was that the state of the ground would make it impossible before the
beginning of April at the earliest. Ludendorff insisted that any attack must
begin by the end of February or beginning of March at the latest.
Schulenberg then suggested attacks either side of Verdun, arguing that a
French defeat there would shatter her resistance for good. To Ludendorff,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Pla n n i ng t h e Spr i ng Of f e nsi v e s203

however, not only was the ground at Verdun difficult and enemy defences
strong, but an attack there would leave the British still free to fight in
Flanders. Thus the Germans might end up fighting two major battles when
they had only enough ammunition for one. In Ludendorff ’s view, it was the
British who were the engine of the Entente war effort and so the main
effort must be directed against them. The British also had the advantage of
being less formidable opponents than the French. If the Lys were ruled out,
the only viable sectors left were Arras and the area around the Somme near
St Quentin. Formidable enemy defences at Arras, not least on the blood-
soaked ridges of Vimy and Lorette, made an attack there unappealing. Further
south, the ground was more favourable and rail junctions such as Amiens
offered attractive targets. Crossing the wastelands created in 1916 and 1917,
however, would be tricky. Ludendorff ordered further studies, with five
operations to be developed in greater detail: Kuhl’s proposal for a drive on
Hazebrouck (Operation St GEORG, later plain GEORG); an attack on Arras
(MARS); St MICHAEL (MICHAEL), centred on Saint-Quentin; and twin
attacks north and east of Verdun (CASTOR and POLLUX).3
The Mons conference of 11 November shows the best and worst of
German military planning during the First World War. The German army
has often been accused of focusing so keenly on the tactical that it lost sight
of the political dimension of war. In fact, however, Ludendorff and his advisers
seem to have displayed a shrewd insight into Entente dynamics, fully appreci-
ating the depth of French resilience and the extent to which Britain was the
driving force of the alliance. The decision to target the British shows, indeed,
that OHL was building its military plans around this political understanding.
What Ludendorff proved unable to do, however, was to translate and sustain
that insight throughout the planning process. He failed to appreciate, for
instance, that attacking astride the join between the British and French armies
was more likely to push them together than to pull them apart.
Rupprecht put his name to the study which Kuhl sent to OHL on 20
November. This reiterated that, along the whole front of Army Group
Crown Prince Rupprecht, the sector west of Armentières remained the
most promising. A decisive attack should employ forty divisions and 400–500
batteries of heavy artillery and drive into the British rear to cut off and
crush the main body of the BEF. If available strength permitted only a
smaller effort, this would still be the best ground for a shallower diversion
involving twelve to fifteen divisions. Kuhl also briefly discussed other
options. With MARS, he worried about taking the high ground around
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

204 Y e a r of De f e ats 1918

Arras and calculated fifty divisions would be needed. MICHAEL, he argued,

brought several problems. First, the Germans would have to sustain the attack
south-westward across the devastated zone left by ALBERICH. Second, they
would then have to wheel ninety degrees right and head north-west to roll
up the British line, while holding off French counter-attacks from the south
along an ever-expanding front. At least fifty-five divisions would be required
to achieve an operational objective which remained vague, at best.4
The 20 November memorandum made no allowance for Russia drop-
ping out of the war. When she signed a cease-fire, Kuhl judged that the
British were unlikely now to attack in Flanders but would expect a German
offensive. He revised his plan. The best chance to achieve surprise and
reduce the chances of another Materialschlacht remained where British
defences were weakest, between Armentières and the La Bassée Canal, with
an attack in April. It would be important, however, to tie down enemy
reserves with a series of diversions up and down the Western Front first.
These should begin in early March and also aim to exacerbate frictions
within the Entente command. Local attacks around the Ypres Salient, north
of Lens, and at St Quentin would precede a large push at Cambrai aiming
to suck in British reserves. An offensive at Verdun would do the same for
the French. Kuhl was now proposing, in fact, not just a plan for a single
hammer blow, but a design for a whole campaign. By pulling together a
series of operations, he expected to deliver a cumulative effect much greater
than the sum of its parts. He was also keen to keep options open. If the
Germans failed to achieve surprise in one place, rather than hammer away
fruitlessly like the British, he wrote, they should reorganize and attack else-
where. Speed was essential.
In some respects Kuhl’s approach resembled that of Wetzell’s latest staff
study, dated 12 December, which also called for a series of operations.
According to this, Second and the newly formed Eighteenth Armies would
begin with an attack by twenty-two divisions at St Quentin. A couple of
weeks later, Fourth and Sixth Armies would drive on Hazebrouck. The
difference between the two plans lay partly in the different sectors they
chose for the main preliminary attack. More importantly, however, while
Kuhl and Rupprecht seemed still to believe that a single knock-out blow
was possible, Wetzell did not. He expected no single operation to be
­decisive on its own. In this, he was also out of step with Ludendorff, who
remained, as David Zabecki says, ‘fixed in his mind to a single great attack—
a Schlieffenesque Battle of Annihilation’. Ludendorff also remained committed
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Pla n n i ng t h e Spr i ng Of f e nsi v e s205

to attacking the British, while Wetzell believed that the French constituted
a more enticing target. Wetzell reiterated these points in a memo written
on Christmas Day.5
In theory, no final decision had yet been reached. At another planning
meeting on 27 December, all the Army Groups were told to continue plan-
ning for a range of options from the Swiss border to Flanders. Kuhl was
asked to pull together plans for six different attacks. Fourth Army was to
work on GEORG II, an attack on Mount Kemmel and Bailleul, while Sixth
Army studied GEORG I, a drive on Hazebrouck, and MARS, aimed at
Arras. Meanwhile, Second and Eighteenth Armies drew up MICHAEL with
objectives Bapaume, Péronne, and the Somme. In reality, however, Ludendorff
had already made up his mind. Much of the paper trail comprises contin-
gency plans or efforts to confuse and deceive. Ludendorff ’s intent was clear,
however, in the orders he sent out. For instance, on 19 December OHL
ordered Eighteenth Army to take over the front from St Quentin down to
the River Oise. General Oskar von Hutier was in command, with his artillery
under Colonel Georg Bruchmüller. This was the team responsible for stun-
ning victories at Riga in the autumn and their deployment here was taken by
British intelligence to indicate a possible attack. Similarly, on 29 December
OHL told Rupprecht that he was to have a new Seventeenth Army under
him. The commander, Otto von Below, the victor of Caporetto, brought a
reputation as a good attacking general. His chief of staff was Rupprecht’s old
comrade, Krafft von Dellmensingen. Rupprecht and Kuhl realized that
enemy intelligence would be alerted by Below’s presence and considered
sending him temporarily to Sixth Army before switching him back at the last
minute. This deception plan faded away, however, when Below objected to
any changes in the composition of his staff.6
Rupprecht finished his 1917 diary optimistically. It was true that the German
army had been unable to break through in 1914, and that the Entente had
failed to do so every year since. In spring 1918, however, the Germans would
be able not only to maintain their battlefield superiority over the enemy but
deploy reserves and artillery on a scale never before seen. The chances of
ending the war with a major blow were reasonable. Although Germany
would be in trouble if the offensive failed, Rupprecht was almost more
nervous that the British might imitate ALBERICH and pull back. While
1918 thus certainly held dangers, the situation for Germany was overall
better than it had been a year previously and he could hope that the new
year might bring victory and peace.7
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

206 Y e a r of De f e ats 1918

Planning continued in January under conditions of the utmost secrecy.

At Rupprecht’s headquarters, even members of the operations staff such as
Captain Wilhelm von Leeb, a future field marshal, were deliberately shut
out. Rupprecht was not happy with the plans the armies submitted. Sixth
Army’s ideas for GEORG I were too academic, especially because they
under-estimated the difficulties of re-taking the Messines Ridge. Without
clearing enemy observation posts from there, it would be impossible to
move supplies and reinforcements up to maintain momentum and seize
Mount Kemmel. The terrain for MARS, also, especially the high ground at
Vimy and Monchy-le-Preux, was very favourable to the defender. MICHAEL
presented fewer tactical problems but led in a less decisive direction. The
scale of forces necessary to turn any tactical success to strategic effect was
beyond what was ever likely to be available.8
On 18 January, after a couple of weeks of political distraction over Russia,
Ludendorff began a three-day tour of the front to discuss the offensive. He
followed this with a conference at Avesnes on 21 January. Again, he stressed
the importance of close artillery–infantry cooperation, as well as the need
to keep moving forward. Operationally, first-wave divisions must fight on
without relief for at least the whole of the first day of the assault. Ludendorff
also explained that OHL currently had sixty-three divisions on hand in the
west. By the end of March a further twenty-four would probably arrive
from Russia and Italy. This would create a reserve of eighty-five to ninety
divisions. Where to use them? He summarized his view of the options.
GEORG I was too dependent on the weather. If spring came late, it might
be May before it became feasible. MARS was tactically too hard. Therefore,
they would adopt a version of MICHAEL. To the objection that MICHAEL
lacked any defined objective, Ludendorff replied that ‘in Russia also we
always just pushed on to a nearby objective and then saw how we could
carry on’. Seventeenth Army would be slotted into the line between the
Sixth and Second Armies, occupying the sector between the River Scarpe
and Flesquières, whence it would launch MICHAEL I south-westwards
towards Bapaume. Second Army would drive due westwards on Péronne
(MICHAEL II), while the task of Eighteenth Army with MICHAEL III
was to cover the left flank. This last army, Ludendorff announced, was to be
transferred from Rupprecht’s Army Group to that of Crown Prince Wilhelm.
He made a point of asking for Rupprecht’s agreement to the change.
Rupprecht could not publicly object but told Kuhl that ‘now the war is lost’.
The decision, he thought, made no military sense. It would obviously be
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Pla n n i ng t h e Spr i ng Of f e nsi v e s207

better for the whole attack to be controlled by a single commander. He saw

it as nothing more than a political gesture designed to rehabilitate Wilhelm’s
military reputation after the failures at Verdun. In his memoirs Ludendorff
claimed that his motives were exclusively military: ‘though profoundly loyal
to my King, I am an independent man and no courtier’. In fact, this decision
would allow Ludendorff to maintain overall control as he could not if
the  operation was delegated to a single Army Group. Finally, Rupprecht
was  to  continue preparing for GEORG, which might be executed if
MICHAEL failed.9
Over the next few days OHL sent out formal orders codifying what had
been discussed at Avesnes. Rupprecht told his armies to be ready for
MICHAEL by 20 March and GEORG I and II by early April. After the
plans were wargamed in early February, Rupprecht proposed cancelling the
difficult MARS operation south of Arras. Instead, he proposed diversions to
draw British reserves to Flanders, but Ludendorff rejected this suggestion.
Further meetings took place across February and March but the overall
shape of the offensive from now on changed little.
Much of the discussion at these meetings centred on tactics. Ludendorff,
and indeed the whole German army, has been criticized for being obsessed
with the tactical level of war and paying too much attention to the trees at
the expense of the forest. Ludendorff had logical reasons for his approach,
however, as the words with which he had opened the Avesnes conference
on 21 January show:
We talk a lot about Operations and not much about Tactics. I have taken part
in many operations. But I have never known in advance how an operation
would unfold. Decisions must be made day by day, often hour by hour.
Whether one will be in a position to strike in the direction originally
intended, or whether one will be forced to veer off elsewhere, it’s not possible
to say in advance. Not even three or four days ahead. The picture can change
so much in the meantime, that the original design can’t be carried through.
So I warn you against committing to any particular course of action, even if
it’s the best one. Instead, one must think, using all possible means, ‘How do
I defeat the enemy? How can I break through his front line?’ What happens
thereafter in many cases depends on making a split-second choice. The
choice you make then must be correct. I advise you therefore to pay greater
attention to tactical matters.
During the same meeting, Ludendorff pointed out that ‘tactics must be
given precedence over abstract strategy. Without tactical success there could
be no strategy. Strategy which gives no thought to tactics is doomed to
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

208 Y e a r of De f e ats 1918

f­ ailure.’ He had a point, although uncertainty about the future hardly absolves
one from contingency planning. In any case, Rupprecht resented Ludendorff ’s
tactical interventions as interfering in the proper business of subordinate
commands and risking excessive standardization.10
Rupprecht was growing increasingly concerned that they were taking a
path which could lead only to disaster. ‘Do we have enough resources for a
major offensive?’ he asked his diary on 30 January. ‘I doubt it myself and am
deeply worried.’ When Ludendorff argued that ‘thorough preparation is
more important than surprise of execution’, Rupprecht disagreed: ‘I think
the opposite. If we can’t surprise the enemy with our attack, it will defin-
itely become a fruitless Materialschlacht.’ On 4 February he wrote to his
father: ‘Sadly I cannot see the military situation in a rosy light: reserves of
manpower are running short and we’re losing many horses to malnutrition.
Raw materials of all kinds are in short supply. . . . It is now impossible for us
to end the war victoriously.’ Extreme war aims in the east were a mistake
which would make a negotiated settlement harder. Kuhl shared Rupprecht’s
concerns: ‘we will do our best but I now have not much confidence in the
attack, unless we’re especially lucky’. The MICHAEL attack risked driving
into a devastated wasteland and creating nothing more than a bulge in the
line. On 7 February, Kuhl spoke out ‘unsolicited’ to Rupprecht after dinner,
arguing that they were not strong enough to succeed until peace with
Russia was final and the army there could be sent west.11
When Rupprecht visited Munich for his parents’ Golden Wedding cele-
brations on 20 February, he tried to persuade his father how poor the out-
look was. Ludwig III would not be convinced. More importantly, Rupprecht
also buttonholed the Kaiser, warning him of the difficulties facing the
offensive, including shortages of manpower and horses, the fact that they
were playing their last card, and the danger that they might gain nothing
more than a partial victory and a useless and vulnerable salient. Wilhelm II
was unconcerned. He replied that they were not seeking a major break-
through but would damage the enemy with successive operations in different
places. He was parroting Ludendorff ’s line. It remained doubtful, however,
whether Germany possessed the strength for multiple operations on the scale
required. Rupprecht made sure also to talk to the Kaiser’s military advisers,
Generals Plessen and Lyncker, along similar lines. ‘Who knows’, he asked his
diary, ‘whether we’ll ever again get so good an opportunity as we currently
have to begin peace talks with the Western powers and America?’ By early
March, however, as preparations began to come together, Rupprecht’s spirits
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Pla n n i ng t h e Spr i ng Of f e nsi v e s209

had risen once more: ‘when you spend three and a half years purely on the
defensive, you easily become a pessimist’, he noted ruefully.12
On 6 March Rupprecht issued orders for the offensive to his armies. The
first objective was to cut off the Cambrai salient ‘and thus win a major tac-
tical victory’. While it was not possible to be dogmatic about what should
follow, since circumstances might change, the general intention was for
Seventeenth Army to pivot north-west, rolling up the British line, while
Second Army headed west, driving deep into the enemy rear. Eighteenth
Army was to guard the southern flank. Hutier’s Eighteenth Army had been
pushing for a more ambitious role than flank guard for some days, so on
7  March Ludendorff called a meeting to ensure proper coordination
between Eighteenth and Second Armies. Ludendorff agreed that it made
sense for Eighteenth Army to seize bridgeheads on the far side of the
Crozat Canal if possible, but refused to unleash Hutier fully until they had
all seen how the battle was playing out. Another decision made at Mons that
day set Thursday 21 March as the start date.13 OHL issued its formal written
orders for Operation MICHAEL on 10 March, with final orders from Army
Group Crown Prince Rupprecht going out six days later. By now there
were indications that the British were strengthening their defences all along
the front, but there was no evidence that they knew where the attack would
fall. Nothing, however, could halt MICHAEL’s momentum now.14
The sky was clear the night before the attack, the weather better than
expected. Rupprecht was relieved that the British had done nothing to dis-
rupt his build-up. He was disturbed that evening, however, when Ludendorff
told Kuhl that, if the Allies were defeated, his ultimate intention was for
Army Group Rupprecht to establish itself along a line St Pol–Doullens–
Amiens while Eighteenth Army held Roye–Noyon. Eighteenth Army was
clearly no longer meant purely to guard Rupprecht’s flank, but was expected
itself to form a major part of the offensive. What had always been a highly
complex manoeuvre was now developing into a widely divergent attack.
This risked dissipating, rather than concentrating, German strength. When
Rupprecht protested, Ludendorff replied: ‘In Russia we always set ourselves
only a close objective. We just cut a hole in the enemy’s lines. The rest takes
care of itself!’15
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

20
Operation MICHAEL

R upprecht woke up on 21 March to find the countryside blanketed in a

dense fog.Visibility in places was just a few metres. At twenty to five in
the morning the German guns opened up. The preparatory bombardment
went on for five hours. Altogether, over 6,600 guns and 3,500 trench mortars
fired 3,200,000 rounds that day, including gas shells, along a 100-kilometre
front from Croisilles down to La Fère. The climax was five minutes all-out
shelling of the enemy front line, after which, at a quarter to ten, the barrage
began to creep forward. A first wave of twenty-one specially trained and
equipped assault divisions followed close behind. They were supported by
twelve regular divisions of the line. A further twenty-nine assault and six
line divisions formed two further waves behind. The defenders were
eighteen divisions from the British Third and Fifth Armies, with a further
eight infantry and three cavalry divisions in reserve. In the air, for once the
Germans outnumbered the British by almost two to one.1
At Rupprecht’s headquarters, it took until midday for any news of the
attack to come in. The first reports were highly encouraging. The enemy
front line had fallen almost without resistance and second-wave divisions
were moving up as scheduled. British units were transmitting by wireless en
clair: ‘We are in extreme distress!’ The prospects for cutting off the British
salient at Flesquières looked encouraging. Once that had fallen it would
be possible to shift heavy guns to get ready for MARS-SOUTH. The only
oddity Rupprecht spotted was that his men did not seem to be capturing as
many enemy guns as expected. The British batteries seemed to have pulled
back out of harm’s way. There was no need to worry too much yet, however:
they would soon fall into German hands.2
Rupprecht’s instinct proved correct. Returns the next day showed that
Second Army had captured just fifty British guns. The first day of the offensive,
however, undoubtedly constituted a German tactical victory. The artillery
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ope r at ion M ICH A E L211

preparation had been effective, surprise was achieved and infiltration tactics
worked well, especially in the fog. Particularly on the British Fifth Army
front in the south, the defenders were stretched too thinly. Even where
resistance was stout, it was disjointed. British communications collapsed,
so combined arms cooperation broke down, too. Guns were left blind while
all around them the battle swirled.Yet German progress fell short of expect-
ations. Eighteenth Army penetrated deepest, in places up to 13 kilometres,
closing up towards the Crozat Canal. Hutier had deployed only two of
his second-echelon divisions and still held twelve in reserve. Seventeenth
Army’s attack on the right flank, on the other hand, faced much tougher
resistance and at best made only some 5 kilometres, finishing the day still
nearly 8 kilometres short of its objectives at Bapaume and Ytres. It had been
forced to commit all the divisions of its second wave to get even that far.
Ludendorff chose to reinforce success rather than relative failure, sending
six  fresh reserve divisions to Eighteenth Army. On the other side, Haig
transferred five divisions down from the north to prop up his Third Army
south of Arras.3
The next day brought Rupprecht more encouraging reports. Seventeenth
Army forced Byng to withdraw and abandon the vital high ground at
Monchy-le-Preux. A counter-attack by British tanks was driven off with
heavy losses. Below’s flank was now more secure and the way seemed clear
for Ludendorff to extend his attack northwards. He ordered Kuhl to prepare
for Operation MARS both sides of the River Scarpe. At the same time Sixth
Army was instructed to get ready for an attack on the Lorette Spur under
the code name RIDE OF THE VALKYRIES. Second Army drove on
despite tough resistance, and captured, among other places, Gouzeaucourt,
forcing the British to evacuate the Flesquières Salient that night. Further
south still, disarray within the command of Gough’s Fifth Army led to a
premature retreat, leaving a gap in the British line north of Ham which the
Germans were quick to fill.4
German hopes were high. Reserves remained plentiful. Some twenty fresh
divisions were standing by and morale was good. Rupprecht found a train of
wounded soldiers, on their way home for treatment, waiting outside his
headquarters. He stopped to chat to them, finding them full of high spirits
and convinced of victory. Events had moved faster than he had anticipated:
I always hoped indeed that, if the attack restored mobility—which in any case
is not such an easy thing to do—the British higher command would fail.
They’re only used to set-piece attacks with very detailed painstaking planning
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

212 Y e a r of De f e ats 1918

while our leaders are better trained and more skilled at manoeuvre warfare.
But I am surprised that the endurance of the British soldier—which I know
well—did not last longer and that things have moved so quickly. The British
are now in a very tricky spot.

Now that the British would be forced to fall back over the Somme, ‘their
biggest defeat of the war definitely awaits’. When news arrived that the left
of Second Army had broken clean through the British defensive system and
was out into open country, Rupprecht exulted: ‘Victory is ours! Who could
have hoped for such a success 24 hours ago!’5
The next day brought further good news, south of the Somme at least.
Indeed, 23 March saw the deepest single advance of any day of MICHAEL:
16 kilometres. Along most of the front attacked, the Germans had now
bundled the British out of all three lines of their defensive system. Eighteenth
Army seized crossings over the Somme and broke out from its bridge-
heads along the Crozat Canal. The British Fifth Army was beginning to
disintegrate, with most of the divisions hit by the first assault reduced
to mere remnants. Worse, a gap was growing between British Fifth and
Third Armies which even the arrival of some French reinforcements was
not yet able to fill. British artillery fire was weakening noticeably. Reports
of chaos in the British rear were multiplying. Although north of the
Somme progress was slower, the Germans did manage to capture Mont
St Quentin, just above Péronne. They had taken 40,000 prisoners and seized
some 400 guns.
Ludendorff estimated that the British were down to about fifty divisions
and that the French would stand on the defensive to cover Paris. He issued
orders to exploit the success won so far. Seventeenth Army would head
north-west, resting its left flank on the Somme all the way to Abbeville.
Second Army would drive due west towards Amiens while Eighteenth
Army headed south-west. This was no new brainstorm on Ludendorff ’s
part. As we saw above, he had outlined something very similar to Kuhl on
the eve of the offensive. His estimate of the British and French was roughly
right for now: not until 26 March, when the Doullens Conference gave
Foch authority to coordinate the Allied response, did more aggressive
French support for the British become possible. Nevertheless, however
good the intelligence seemed, Ludendorff was increasing the ambition of
his offensive considerably. Where the original plan had envisaged merely
holding off the French while the British were crushed, now Ludendorff
expected his men to be able to take on and defeat both at the same time.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ope r at ion M ICH A E L213

Figure 20.1.  German troops in the streets of Bapaume, March 1918

He was over-estimating the resilience of his increasingly tired troops and the
ability of his logistic system to maintain momentum along ever-lengthening
supply lines across devastated terrain. Moreover, an eccentric attack of this
kind would dissipate strength, rather than concentrate it. An attack like that
might come off against an already defeated enemy, but the British army was
not yet weak enough for it to work.6 MARS-SOUTH was now pencilled
in for 27 March, MARS-NORTH for the next day, and RIDE OF THE
VALKYRIES for 29 March. Eight days or so later, Sixth Army would attack
on the River Lys. The deep objective for this last attack, code-named
KLEIN-GEORG, was ambitiously set as Boulogne.
Second Army occupied Bapaume again and started to fight its way across
the old Somme battlefield on 24 March. By the end of the day, fighting had
broken out in High Wood once more.The British could no longer organize
a continuous line. Instead the defence comprised a series of islands around
which the rising German tide lapped. Soon the Germans were back on the
River Ancre. Eighteenth Army, however, was starting to feel the effects of
fatigue and stretched supply lines, managing to advance only 4 kilometres
rather than the planned 12. Still, Hutier managed to drive a wedge between
the French and British lines around Roye.7
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

214 Y e a r of De f e ats 1918

We now know that this period marked the nadir of Anglo-French

cooperation. Neither commander-in-chief trusted the other. By 24 March
Pétain felt that the British were not trying hard enough and that Haig was
already falling back on the Channel ports. Paris remained his own priority.
Exactly what happened at a meeting between Pétain and Haig late that
evening is disputed. So are Haig’s actions that night in response. But
the  next day he warned Byng that he could no longer rely on either
Gough’s army or the French and must assume responsibility for the British
Expeditionary Force’s (BEF’s) southern flank. At GHQ, Haig’s staff officers
began working up logistics plans in case the line of communications
southwards was cut. Foch was concerned that the French and British were
fighting different battles and, by drifting ever further apart, were playing
into German hands. At Doullens on 26 March senior political and military
leaders from both Britain and France decided to forestall this danger by
agreeing to unify command under Foch. His first order was that Pétain
and Fayolle should defend Amiens and maintain contact with the British.
David Stevenson has described the Doullens Conference as ‘one of the
great symbolic moments of the war, although’, he warns, ‘its practical
import was greater for the future of the 1918 campaign than for the battle
in progress’.8
Foch possessed the attribute Napoleon most looked for in his generals: he
was lucky. He took over at just the right time. Although Hutier took Roye
and Noyon on 26 March and Marwitz’s men occupied Albert, the power of
the German offensive was fading. Seventeenth Army was still struggling to
make headway near Arras. Resistance had been determined and German
casualties were heavy. They had lost 90,000 men in five days. The troops
were getting tired. Of the thirty-seven divisions in action, eight were already
doing their second stint in the line. Supplies of ammunition, food, and even
water were problematic.The railheads had been left far behind and transport
was in short supply. The first thing the troops who captured Albert did was
loot the town for food and drink.
Ludendorff was frustrated at the lack of progress. He telephoned
Rupprecht in a rage. Units were being left too much to their own devices,
he argued. They must be pushed harder to attack. Worse, they were
using the wrong tactics. Rather than infiltrating in the proper stormtroop
style, they were attacking head on in close-order formations. Rupprecht
told Kuhl to find a way for Second Army to ease the pressure on
Seventeenth Army and help it forward. Oberste Heeresleitung’s (OHL’s)
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ope r at ion M ICH A E L215

solution was to order the launch of MARS either side of the Scarpe
for 28  March.9
Rupprecht described 27 March as ‘a disastrous day’. His staff officer, Major
Prager, considered it the turning point of the whole campaign. Neither
Seventeenth nor Second Armies managed to get much further forwards that
day. Although Eighteenth Army seized Montdidier, it was clear that even
here resistance was intensifying as French reinforcements arrived. That was
bad enough. Worse still, however, was a dispute between Rupprecht and
OHL over three divisions held in reserve. Rupprecht wanted to commit
them to support the bogged-down right wing of Seventeenth Army. OHL
refused. It diverted all three south of the Somme. When Rupprecht heard
this, in the early afternoon of 27 March, he exclaimed ‘now we have lost the
war’. Why he felt three divisions were likely suddenly to make a difference
on a flank where the Germans looked increasingly stuck, and why OHL
was wrong to refuse to reinforce failure, he did not make clear. In any case,
it is possible that Rupprecht was getting his dates confused. According to
Kuhl, the three divisions were sent to Second Army only in the afternoon
of 28 March, after it became clear that MARS was misfiring. Prager could
not recall which day the transfer took place and Rupprecht admitted that
his notes for both days were sketchy. The true significance, perhaps, of this
story is that it shows how tightly nerves within the German high command
were stretched and how swiftly the mood could and did swing. In only four
days, triumph had collapsed into despair.
This was not only the case at Rupprecht’s headquarters. On the night of
27/8 March Ludendorff sensed that the road from Montdidier to Amiens
lay open. He told Crown Prince Wilhelm to send forces up the valley of the
Avre and seize this vital city and rail junction, but Eighteenth Army refused
to advance. At the time, Wetzell thought this showed that MICHAEL had
run its course and should now be suspended. He argued that resources
should be shifted north for GEORG. Later, he felt the failure to push for
Amiens a decisive error, although it is not clear that Eighteenth Army could
realistically have done so. Amiens lay over 30 kilometres from Montdidier,
deep in the enemy rear. To ask tired troops, who had taken a week to
advance 50 kilometres and whose supply lines were already stretched, to
wheel right and strike for it was surely to ask a lot. French reinforcements
were arriving all the time and making the task ever harder. Eighteenth
Army eventually did send a corps up the road towards Amiens but found
the way blocked. On 28 March Rupprecht’s troops overran British supply
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

216 Y e a r of De f e ats 1918

columns at Le Hamel. For a few hours he thought that perhaps at last they
were about to break through, but such hopes soon evaporated. General
Debeney of French First Army was beginning to close up the gaps, not
only between British and French, but also within the French line. He no
longer had to throw all available troops straight into the line as soon as they
arrived. Instead he was at last able to build up reserves near Amiens.10
If 28 March was a day of missed opportunities south of the Somme, north
of the river it was one of dashed German hopes. Operation MARS was
designed to break the British defences around Arras and get the right wing
of MICHAEL moving once more. Nine German divisions launched the
attack, supported by 1,250 guns. The bombardment was less thorough than
it had been one week before, counter-battery fire less effective, and the
barrage crept forwards too quickly for the infantry, who took time to fight
their way through largely intact defences. Partly because not all the troops
had been properly trained in stormtroop tactics, and partly because the
enemy positions here were well established and complex, the tactics used
were relatively traditional. Close-order replaced small-group infiltration.
With little fog to mask the attackers, the British shot them down with ease.
By early afternoon MARS had clearly failed and OHL suspended the
operation. Kuhl later judged that Ludendorff was excessively concerned
about casualties on 28 March, that he lost his nerve that day, and that
one  more push might have broken the BEF’s backbone. But now that
MARS had failed, any attempt to capture the Vimy and Lorette Ridges was
obviously out of the question. Second Army was stuck in the Ancre valley.
Rupprecht felt the need to remind his men that, with resistance now stiffening,
frontal attacks without proper artillery preparation invited only casualties
and must be avoided. Instead, they should probe to identify weak points and
work around the flanks of enemy defences.11
Good Friday, 29 March, was the quietest day on the Western Front in
over a week, with the only change being a small German advance north
of Montdidier. More interesting were the deteriorating relations within
the high command. In the morning Ludendorff telephoned Kuhl and told
him to put Seventeenth Army on the defensive while Second and Eighteenth
Armies continued the attack in the Amiens–Noyon–Montdidier sector.
Kuhl and Rupprecht objected that this deviated from the original plan to
crush the BEF first and only then turn on the French. It would give the
British breathing space at precisely the time when one more good strike
might finish them off. Further, they were worried that the scaled-down
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ope r at ion M ICH A E L217

GEORGETTE, now intended for launch in a little over a week, was

being rushed.
Ludendorff was proving increasingly awkward to work with. He had
always been impatient and highly strung but was now becoming especially
aggressive and hostile. Further, he was increasingly bypassing the proper
chain of command, intervening directly in the operations of Rupprecht’s
armies. Often Kuhl and Rupprecht learnt of orders issued not from OHL
but from their subordinates. Rupprecht thought Ludendorff had lost his
nerve. A couple of days previously the second of his stepsons to die in the
war had been lost in action and the news had hit him hard. Captain von
Leeb of Rupprecht’s staff reckoned ‘Ludendorff ’s nerves have completely
gone. It is a crime to leave him in charge of commanding the armies. All our
divisions will be placed in an operationally impossible position, facing
stronger resistance in tactically disadvantageous conditions. So our last
divisions will be rendered useless. We could have forced a decision, if we’d
been handled properly! Now it will be too late, we’ve used up our strength
to no purpose.’12
Ludendorff continued to push his men hard, but even he was beginning
to realize there were limits. When the attacks of Eighteenth and Second
Armies ground to a halt on 30 March, he ordered them to regroup and try
again. Marwitz protested that his troops were worn out. They could do no
more. Ludendorff gave way, suspending the MICHAEL operation until
4 April. The Germans spent the first three days of April resting their men
and bringing up supplies. This was no easy task: reinforcements heading
south had to march across the lines of communication of two attacking
armies and kilometres of devastated ground. Rupprecht’s diary entry for Easter
Sunday, 31 March, gives a vivid impression of the chaos and devastation he
observed as he motored through endless columns of marching men to visit
Second Army. Leaving Mons at eight a.m., he first came across evidence of
fighting at Cambrai, where almost no window was left intact. He crossed
the St Quentin Canal at the ruined village of Masnières. From a distance,
Gouzeaucourt looked intact, but in reality it lay amidst ‘a labyrinth’ of
trenches, wire, and shell holes. Destroyed tanks and abandoned artillery guns
lay round about. From there he drove up Mont St Quentin to look down
at Péronne. The town appeared welcoming and picturesque in the rays of
the spring sun. Again, though, the reality was different. The houses were
mere shells, glowing only because their shattered walls let the sunlight
through. The nave of the gothic church lay in ruins, its towers missing.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

218 Y e a r of De f e ats 1918

Second Army headquarters had been set up in an old British field hospital
in the little village of Tincourt, which had been otherwise wiped from the
map. In a valley north of Cléry he found the remains of a British artillery
regiment which had been shelled as it tried to get away. Field guns stood
abandoned, together with a piece of heavy artillery and a tank. Most of the
crews had already been buried, but horses still lay dead all around. He drove
on to Bapaume over the old Somme battlefield. Grass had already grown
over the trenches and shell craters. No villages were to be seen: they existed
only as place names on sign posts. All that remained of the mighty forest of
St Pierre-Vaast was two clumps of low tree stumps. All over the steppe-like
rolling downs were planted countless wooden crosses with red, white, and
blue cockades, marking graves. ‘A bleak landscape!’ he called it. ‘A sad pic-
ture of unbounded devastation which I’ll never be able to forget.’ The
Bapaume–Cambrai road was in a very poor state, but Bourlon Wood, for all
the fierce fighting of the autumn, was relatively intact. He got home at nine
p.m. having travelled about 250 kilometres in thirteen hours.13
The German breathing space proved even more useful to the defenders.
The British and French were able to bring up fresh reinforcements and
reorganize themselves, consolidating their hold on Amiens and improving
positions all along the line. When Ludendorff renewed the offensive on
4 April, his objective was to take Amiens, or at least to get close enough for
his heavy guns to interdict the important railway lines running through the
city. Fourteen divisions attacked in the early morning of 4 April but only
four of these were fresh formations and progress was limited. Second Army
reached the outskirts of Villers-Bretonneux, 16 kilometres east of Amiens,
but stalled there. Attempts to get the attack moving again the next morning
collapsed. At Army and Army Group level, the feeling was unanimous:
MICHAEL must be suspended. During the evening of 5 April, OHL agreed.
All efforts now would be directed to the attack code-named GEORGETTE,
up in Flanders.14
In some ways, Operation MICHAEL had been a stunning success.
The concentration and delivery, in secrecy, of such an immense offensive
constituted, as even the British official historian noted, ‘one of the most
remarkable pieces of staff work that has ever been accomplished’. On 21 March,
David Zabecki has calculated, the Germans captured more ground than the
Entente did in 140 days on the Somme. Overall, they penetrated up to
60 kilometres into the Allied lines and captured 1,200 square kilometres of
France. As we have seen, Ludendorff ’s objectives kept changing but the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ope r at ion M ICH A E L219

important fact remains that he achieved none of them. Initially, he sought

to hold the French at bay while he destroyed the BEF. Apparent success
emboldened him to order his armies to fan out across northern France and
defeat both British and French simultaneously. When this more ambitious
plan soon became obviously impractical, he scaled his objectives right down,
now seeking merely to capture Amiens and the vital supply lines running
through the city. This failed, too.
If he did not break the British or French armies, however, he did hurt
them. Eighteen British divisions were worn out by the fighting and could
not be reconstituted, at least in the short term. A further sixteen were being
rebuilt. Haig’s GHQ reserve consisted of just one fresh division. Altogether,
the British lost nearly 178,000 men, 75,000 of them as prisoners of war.
More British troops went into captivity during the two weeks of MICHAEL
than in the whole of the war on the Western Front so far. French casualties
totalled 77,000, of which 15,000 were prisoners. The Entente, therefore, lost
a little over a quarter of a million men. The Germans also captured 1,300
guns, almost half the number the British had started with on 21 March.
German losses, however, were also very heavy: maybe up to 239,000, almost
equally split between the three attacking armies. Ludendorff had thrown in
about ninety divisions, including nearly all those trained and outfitted to
lead the assault. Each division had lost 2,000 men or more. Neither side
could easily afford losses on such a scale, but Germany could spare fewer
than Britain and France. Time, also, was working against the German army.
Both factors made it all the more vital that such a high price be made to
count for something. Perhaps another attack, this time in Flanders, would
knock over the tottering BEF. Once more, Rupprecht’s Army Group had a
lead role in executing it.15
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

21
Operation GEORGETTE
and Summer 1918

A t quarter past four in the morning of 9 April 1918, guns of the German
Sixth Army began shelling the British and Portuguese troops defend-
ing a 27-kilometre stretch of front between Givenchy-lès-la-Bassée and
Bois Grenier. After four and a half hours of bombardment, ten German divi-
sions advanced to launch Operation GEORGETTE. Further divisions of
Sixth Army were in support, and the next morning Fourth Army would
extend the front another 16 kilometres northward, driving into the British
line between Frelinghien and the Ypres–Comines canal. Altogether, twenty-
nine divisions were on hand. Overhead, nearly 500 aeroplanes set off on
missions in support.Thus the second of the German spring offensives began.1
Rupprecht and Kuhl were far from happy with the concept of
GEORGETTE. It fell a long way short of their idea for a decisive haymaker
aimed at the Channel ports to split the British Expeditionary Force (BEF)
in two, trap a large part of it in Ypres, and cut it off from its supplies. At one
stage, planning envisaged 30–40 divisions taking part in the main attack,
with another twelve to fifteen in support. As we saw in Chapter 19, how-
ever, Ludendorff preferred MICHAEL. This reduced the resources likely to
be available for operations in Flanders. GEORG became KLEIN-GEORG
and eventually the even smaller GEORGETTE. What was originally intended
as a knock-out blow became first a diversion and then a contingency plan
in case MICHAEL alone failed to win the war. At one stage in late March
it fleetingly almost seemed that no operation in Flanders might be necessary,
but on about 26 March the decision was taken to launch GEORGETTE.
It was not just wounded pride making Rupprecht and Kuhl unhappy.
They had other reasons for unease. First, they were unclear how much good
a half-strength attack could do, but it was obvious that Calais and Dunkirk
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918221

were now out of the question as objectives.The strategic effect, even if they
could crush the British Second Army and clear the Ypres salient, was likely
to be limited. The MICHAEL experience, indeed, set Rupprecht wonder-
ing whether Ludendorff had any plan at all: ‘it is striking that in all OHL’s
[Oberste Heeresleitung’s] instructions no single intention can ever be dis-
tinguished. Instead the talk is always only of territorial objectives to be
reached. It gives me the impression that OHL is living from hand to mouth
without professing any definite operational aims.’ In a footnote to his diary
he remarked: ‘before starting the second operation of the spring . . . I asked
OHL what the operational objective was. Ludendorff ’s reply on the tele-
phone was “I forbid the word ‘operation’. We bash a hole.The rest takes care
of itself. That’s how we did it in Russia!” ’ Kuhl complained that ‘OHL’s
orders for GEORGETTE are again not clear, no clear objectives and mis-
sions, always just points on the map and tactical measures.’ Second, Rupprecht
and Kuhl were concerned that GEORGETTE was being rushed. Most of the
artillery and aircraft needed for the attack, and even some of the infantry, had
to be redeployed northwards from MICHAEL. This could not even begin
until they were no longer needed there, and then the transfer would take time.
Lack of rail capacity meant some of the infantry divisions would have to
march the whole way. The decision to continue with MICHAEL, even after
30 March when it was clearly losing momentum, ate into that preparation
time. Leeb was furious, accusing Kuhl of refusing to stand up to OHL:

The Army Group plays a miserable role. If I were in Kuhl’s shoes I’d ask them
to back me or sack me. I’m just sorry for the Crown Prince. Kuhl is a very
clever man but he has no character. He’s not nicknamed ‘Mr Look-out-for-
number-one’ for nothing. He definitely knew, quite clearly, that OHL was not
running operations on the right lines.

With Kuhl’s position and seniority, Leeb thought, it was his responsibility to
say something. ‘He owed the Crown Prince that much, at least.’ In fact, after
the war Kuhl at first tried to defend Ludendorff ’s decision to persevere with
MICHAEL into April. In his evidence to the government commission
investigating the causes of Germany’s collapse in 1918, he gave two reasons.
First, he argued that any pause would have given the enemy time to regroup,
too. There is something in this: Sixth Army had already warned that they
would not be ready until about 8 April and there was an argument for keep-
ing up pressure on the British somewhere, especially since driving towards
Amiens might interdict rail movement there and ease GEORGETTE’s
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

222 Y e a r of De f e ats 1918

p­ rogress. Second, Kuhl suggested that it is always difficult to know when to

suspend an attack. One suspects that Kuhl was endeavouring to present a
united military front to the politicians looking to criticize the army in the
immediate aftermath of the war. A few years later, in his 1929 history of the
war, when the political storm had passed, Kuhl accepted that it would have
been better to suspend MICHAEL sooner. Either way, where MICHAEL
had taken seven weeks to organize, now GEORGETTE was being thrown
together in twelve days.2
Whatever his reservations beforehand, Rupprecht was pleased by Sixth
Army’s results on 9 April. The barrage crept forwards more slowly than on
21 March so the German assault troops were better able to keep up with it.
Infiltration tactics again proved effective in thick fog. The Germans easily
rolled over the defending divisions, many of which had been battered in the
recent fighting on the Somme and were still in the process of being rebuilt.
A Portuguese division evaporated almost at once. Rupprecht’s men had
overrun the front three trenches by noon and by nightfall had seized some
of the crucial crossings over the River Lys. Soft ground was making it diffi-
cult to move artillery forwards and caused some of the immense German
tanks to ditch, but only on the left, where the British were defending
Festubert and Givenchy stubbornly, was there cause for concern. Fourth
Army joined in the next morning and enjoyed similar success, storming
Messines village and forcing the British to evacuate Armentières.
On 11 April progress was slower, but the lead German units finished the
day well established across the Lys. They were ready to drive through the
Nieppe Forest for Hazebrouck, having advanced 14 kilometres in three
days. Haig was worried enough to issue his famous ‘Backs to the Wall’ order,
calling for every position to be held ‘to the last man’ and for each soldier to
‘fight on to the end’. On 12 April Plumer, worried for his southern flank,
pulled his troops back at Ypres, giving up, among other hard-won ground,
the ruins of Passchendaele. Elsewhere, though, the pace of the German
advance slowed further and Ludendorff, impatient at Sixth Army’s progress,
switched the main effort from Hazebrouck to the capture of Bailleul. He
was doing exactly what Kuhl had warned against: swinging north too soon,
in close behind the Ypres salient, rather than driving deep into the British
rear. Perhaps Ludendorff realized that he had not the strength to cut off and
destroy the BEF north of his break-in and would have to settle for the more
limited achievement of clearing the Ypres salient. There is evidence in
Rupprecht’s diary that Ludendorff was giving up on the idea of large
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918223

k­ nock-out blows and expecting instead to launch a series of smaller-scale

offensives. His attention was beginning to drift to planning a renewed attack
towards Amiens by Second Army. Rupprecht thought this a mistake which
demonstrated that ‘Ludendorff is a superb organizer, but no great strategist.’
Whatever the reasoning, the change of German direction gave the British
time to reinforce this sector: any chance to seize Hazebrouck and pull off a
deep envelopment of the Allied north flank had gone.3
Over the next few days the pressure from OHL to keep up the pace of
advance remained intense but, as Rupprecht noted, ‘what good are orders to
attack, if the troops are no longer capable of it?’The situation reminded him
of the First Battle of Ypres, when attack orders kept coming which the men
were in no position to carry out. One of the symptoms of exhaustion was a
slip in discipline, with looting and drunkenness a common feature in cap-
tured towns: ‘what a difference from our magnificent army of 1914!’
Rupprecht complained. ‘We are all stressed and over-tired’, he admitted in
his published diary. In the unpublished version, he was more direct, noting
that ‘Kuhl seemed entirely exhausted’ and saying of himself: ‘I am still very
depressed and my nerves let me down badly.’ Friction within the German
command was reaching new heights, and Ludendorff seems to have been at
the root of the problem. On 2 April Kuhl had noted that
Ludendorff is very agitated, intense and nervous, he wanted to some extent
to excuse himself for the multiple changes of decision and orders and blames
the army commands for not orienting him sufficiently and not giving him
important information. For example, a few days ago Second Army all of a sud-
den reported that they didn’t have enough ammunition and that consequently
they would be unable to execute the attack he wanted, and so on. But that
kind of friction cannot be eliminated in warfare: one just has to deal with it. If
you always want to have perfect information for your decisions, you’ll never
reach any decision at all. What he said about his intentions was not at all clear,
he gave no end-objectives or clear missions for the armies, just lots of tactical
odds and ends.

In the middle of April the Sixth Army chief of staff complained to Kuhl
about Ludendorff ’s interfering. Not only was he uptight and micro-managing
the smallest details, he also had a habit of demanding changes to orders
issued by the army after they had been sent out, which caused severe fric-
tion and disorder. Kuhl was in little position to help:
I myself am continually being called up by Ludendorff about tiny details, such
as that one division doesn’t have the correct maps, and so on. Unfortunately
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

224 Y e a r of De f e ats 1918

his liaison officers go around looking for negative things to tell him which will
make them look good. Recent days were frankly dreadful because of all this
flapping around, you can’t even think for the endless telephone calls.The army
group was cut out of the chain of command with Ludendorff dealing directly
himself with the army commanders and giving them their orders. We had to
ask the armies first of all what Ludendorff had ordered them to do, so as not
to contradict him. I had to keep these interventions secret from the Crown
Prince [Rupprecht]: he is already very worked up by them and it could easily
turn into a proper fight, since the Crown Prince is furious about it. He is
becoming in fact entirely excluded.
Later, during the fighting on the Aisne, OHL cut Crown Prince Wilhelm
out of the command loop in much the same way. This sometimes extended
even to important personnel decisions. When Kaiser Wilhelm II returned
from a visit to General von Stetten’s II Bavarian Army Corps, he com-
plained that Stetten seemed out of touch and had made a poor impression
on him. However little trust anyone had in the Kaiser’s judgement, his will
must be obeyed. Stetten, even though he was a personal friend of Rupprecht’s,
was unceremoniously sacked.4
On 15 and 16 April the Germans managed to establish toe-holds on the
lower slopes of the Flanders Ridge but within a couple of days the advance
had come to a complete standstill. Ludendorff ordered a pause to prepare a
set-piece resumption of operations on 25 April. This new assault seized the
vital high ground of Mount Kemmel from its French defenders but the
Germans proved unable to exploit their success. Operation GEORGETTE
had run out of steam. Indeed, when Second Army’s drive south of the
Somme on 24 April did not manage even to take and hold Villers Bretonneux,
it reinforced the feeling that Rupprecht’s whole Army Group was worn out
and needed a rest. Ludendorff would have to look elsewhere for his victor-
ies. GEORGETTE was officially suspended on 29 April. A few days later,
Ludendorff suggested using a couple of divisions to renew the attack on
Festubert and Givenchy. Rupprecht refused.5
Once again, as on 21 March, the Germans had managed an impressive
tactical break-in and caused considerable concern within the Allied high
command. At the operational and strategic levels, however, even though the
British lost more ground around Ypres in three weeks than they had gained
in four months of bloody fighting the previous year, GEORGETTE was a
failure. Again, the fundamental problem was the difficulty of maintaining
the momentum of an attack. The scaled-down operation launched on
9 April was, perhaps, never strong enough to have much chance of achieving
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918225

its most ambitious aims. With the forces available, to take Hazebrouck, drive
for the Channel ports, and cut off the British and Belgian coast was not
considered a very realistic objective even by the most optimistic voices at
OHL. Only if the BEF instantly crumbled might that be on the cards.
When it became clear that the British would not collapse, the Germans
began to work towards the more limited objective of levering them out of
the Ypres salient. This also failed. The cost to the German army was 86,000
men, a little over one-third as many as MICHAEL. British casualties were
roughly the same: 82,040. Overall, in a little over a month, more than
650,000 men had become casualties, half of them German.6
Today, we know that the spring offensives so far had proved a double
defeat for Germany. For Entente commanders at the time, blessed with none
of our perfect hindsight, however, the situation still appeared extremely dan-
gerous. At the end of April, German armies remained poised 16 kilometres
from Amiens and only 60 from both Abbeville and Calais: one good victory
away. The British army had taken a beating. Of its sixty divisions, all bar five
had been heavily engaged, with twenty-nine used twice and six in action
three times.The French situation was better: of their 103 divisions, forty-one
had fought in March and April, three of them twice. Nevertheless, the French
had been compelled to extend their front by some 100 kilometres, eating
into their reserves. On the other hand, French intelligence estimated that,
thanks largely to transfers from Italy and Russia, Germany’s available reserves,
which had been seventy-four divisions on 20 March, had been reduced to
only sixty-four by 1 May. Worse still, the whereabouts of forty-nine of them
were unknown. The threat remained very real and very large.7
Ludendorff felt that the Allies were now too strong north of the Somme.
While defeating the British remained the main priority, he would have to
draw French reserves away down south before attacking Haig again. His
plan was for Crown Prince Wilhelm’s Army Group to hit the French line
along the Chemin des Dames, north-west of Reims, with Operation
BLÜCHER. Eventually some forty-two divisions took part. Execution was
planned for 27 May. Rupprecht’s Army Group was to follow this up as soon
as possible with a resumption of its offensive against the BEF. Early thinking
favoured NEU-MICHAEL, a renewed push between the Somme and Arras,
targeting Doullens, but Kuhl eventually managed to convince OHL that a
NEU-GEORG in Flanders would be more practical and offered greater
rewards. NEU-GEORG would attack between Ypres and the Forest of
Nieppe, aiming for Poperinghe and Cassel. It was later re-designated HAGEN.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

226 Y e a r of De f e ats 1918

Since it would take time to re-deploy guns and equipment for this up from
operations further south, the launch was originally planned for late June,
although by 31 May that had been pushed back to mid-July.8
While planning continued, the German army tried to pick out the ­lessons
of the offensives so far. The way it did so reveals a further deterioration in
its capacity to think. All the attacks of March and April 1918 bogged down
in the face of logistic difficulties and the exhaustion of the attackers. As the
Germans began to analyse these offensives, however, they concentrated not
on operations, but tactics. The lessons-learnt system had started to creak at
Arras and in Flanders the previous year. It now seemed to be losing the
­ability to identify even the most important questions to ask. Worse, it was
even less able to put forward honest answers to the questions it did ask.
After-action reports parroted the dogma that recent fighting had demon-
strated the efficacy of tactical doctrine, even as commanders complained
that their troops were unable to shake the habits of trench warfare and keep
up with the increased tempo of mobile operations. Many units did not even
employ the new tactics. Corps and army headquarters, far from leaving sub-
ordinates free to use their initiative, had been trying to exercise too much
control, intervening even in the movements of individual regiments. The
General Staff seemed to Rupprecht to be part of the problem:
Higher commands are paid ever less attention to.Their reports go immediately
to OHL to mostly young General Staff officers, who have little of the import-
ant understanding of personalities.These judge them and the judgements they
make often have huge impact. The quest to present one’s own performance in
the best possible light tempts them to over-sharp and not always accurate
criticism.
Albrecht von Thaer, an experienced staff officer, shared a similarly low
­opinion of what he called the ‘demi-gods’ who surrounded Ludendorff at
OHL, none of whom was prepared to pass on unwelcome news.The learn-
ing process was beginning to prove itself incapable of keeping up with the
pace of change. Indeed, more generally, paralysis was gripping the German
army. Control was becoming increasingly centralized but those at the top
who insisted on making more and more of the decisions were receiving
ever less reliable information on which to base them. The consequence,
inevitably, was poor choices.9
Rupprecht was growing intensely concerned: ‘if our next big blow fails,
our situation will certainly be poor and there will be little hope that we can
bring the war to a tolerably satisfactory conclusion’, he noted on 29 April.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918227

About a week later he reported to his father that although ‘the overall situ-
ation is at the moment not unhealthy’, Germany needed to win by mid-
summer if she were to avoid the war dragging on into the unforeseeable
future, but Kuhl and Rupprecht agreed that she would not be able to do so.
Since U-boats were unlikely to force a submission from Britain, the only
way to end the war was to drive the British into the Channel but, with
recent victories ‘largely Pyrrhic ones’ which ate into German strength, and
hunger a growing issue in the Central Powers, this would only become
more difficult over time. The replacement situation was deteriorating.
Divisions arriving from Russia consisted exclusively of men who were
either over forty or very young. One whole company of this kind had raised
the white flag the first time they came under enemy shell-fire. The only
solution, Rupprecht felt, was to restore independence to Belgium and trade
Alsace-Lorraine for colonies to achieve a compromise peace. OHL seemed
at one level to see the situation clearly: they grasped the need to avoid heavy
casualties.They were unwilling, however, to follow their own logic through,
accept that military victory was beyond Germany’s strength, and start to
negotiate. ‘It’s the policy of the ostrich’, Rupprecht noted.10
One man who agreed with Rupprecht was the aspiring politician, Prince
Max von Baden. One of the candidates to replace Bethmann Hollweg the
previous summer, he offered a relatively moderate voice in German affairs.
Since February, he had been pushing for peace talks, quoting the view of
the well-respected Germanophile Swiss war correspondent, Hermann
Stegemann, that Germany’s position was ‘splendid, but hopeless’. Continuing
the war, in Max’s view, increased both Germany’s war guilt and the risks of
Bolshevik revolution across Europe. Military action would only drive her
enemies together, while diplomacy might prise them apart. Max was gath-
ering like-minded people to build a power base. On 19 May he began a
three-day visit to Rupprecht’s headquarters, in the course of which personal
empathy developed into political collaboration. Rupprecht agreed with
Stegemann’s analysis. Max and Rupprecht worked together to influence
other German royals, Ludendorff, the chancellor, and even the Kaiser. For
instance, Rupprecht suggested Max to Ludendorff as a foreign minister to
replace Kühlmann. Results, however, as we shall see, were slow to come.This
should not be mistaken for an exclusively progressive alliance to build a bet-
ter future. Their aims remained poorly defined but centred on making the
Kaiser see the true seriousness of the situation in which Germany found
herself. Their main motivation was a shared dread that defeat would bring
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

228 Y e a r of De f e ats 1918

down the monarchy. Only by contrast with a world as stubborn as the

Kaiser’s court might theirs seem truly liberal voices.11
Rupprecht’s pessimism was growing again. He told Kuhl that, if the war
could not be concluded by winter, there would be a revolution. Kuhl
thought he had picked up this idea from Max, but other factors also were
conspiring to lower Rupprecht’s morale. On 12 May, Rupprecht made a dis-
piriting visit to the Bavarian Lifeguard Regiment, with which he had begun
his military career. They had been one of the first units to the summit of
Mount Kemmel. It proved ‘a painful reunion! Many dear old comrades
were missing. The regiment’s casualties in the recent fighting were quite
severe, in officers alone they lost 40.’ He was having trouble with insomnia,
barely managing four hours of unbroken sleep per night. This left him feel-
ing exhausted.12
BLÜCHER began on 27 May with another stunning success. For the
first time, the German artillery all fired by the map as the British had at
Cambrai six months earlier. This helped achieve surprise. The bombard-
ment was heavy but very sharp, lasting less than three hours. The German
infantry attacked just before first light and quickly overran a garrison
which, rather than defending in depth, was packing the front line. They
seized the Chemin des Dames, overran much of the enemy’s artillery, and
swept on across the River Aisne. By nightfall Seventh Army had advanced
over 20 kilometres, the largest one-day advance of any Western Front
attack. Rupprecht was especially pleased that the Germans were finding
and defeating British divisions in the line: five of them had been sent there
to recover from the battering they had received in March and April.
BLÜCHER seemed to be doing its job of softening up the BEF. German
forces captured Soissons and were threatening both Reims and the River
Marne. By 1 June their forces were only 60 kilometres from Paris. French
reserves were being rapidly chewed up and Foch had ordered troops down
from Amiens to prop up defences in the Marne sector. Ludendorff was so
keen to exploit the success of Crown Prince Wilhelm’s Army Group, how-
ever, that he also began stripping divisions from Rupprecht. A diversion
was becoming a distraction. Even worse, BLÜCHER was now losing steam.
Progress slowed and stopped. On 4 June Seventh Army had to pause to
regroup. An attack between Montdidier and Compiègne on 9 June
(Operation GNEISENAU), designed to ease pressure on one shoulder of
the huge salient created by BLÜCHER and suck further Allied reserves
down from Flanders, accomplished little.13
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918229

Neither Rupprecht nor Kuhl agreed with Ludendorff that HAGEN

would decide the war. With the thirty-nine divisions likely to be available,
the most they could hope for would be to take Hazebrouck and Cassel and
clear the Ypres salient. In any case, Ludendorff decided on 6 June that the
next attack after GNEISENAU should not be HAGEN, but an operation to
push across the Marne and capture Reims.This was eventually scheduled for
15 July and given the codename MARNESCHUTZ-REIMS. HAGEN
would not be feasible before 1 August at the earliest.14
While Rupprecht planned his next move, the British demonstrated that
they would not sit on the defensive for ever. June saw the beginning of a
series of local, limited-objective attacks. Most famously, on 4 July ten bat-
talions of Australian infantry, accompanied by four companies of Americans
and supported by sixty tanks, attacked north of Villers Bretonneux and, in
a model combined-arms action lasting ninety-three minutes, captured the
village of Hamel. On 19 July, the Australians were in action again, this time
with South African and Scottish troops, to capture Meteren village.15
As usual, the first place OHL sought explanations for the success of these
British attacks was in tactics. The high number of German prisoners taken
suggested to the General Staff that the defenders were packing too much of
their strength forward. Maintaining a Forward Zone only 100 or 200 metres
deep was insufficient: 500 or 1,000 metres would be more appropriate. OHL
re-emphasized that the primary defensive effort should concentrate on the
fight, not for the front line, but for the Main Line of Resistance. What OHL
did not comment on was that much of the German army was by now in a
bad way. Rupprecht noted that some of his infantry companies, which should
have had 124 men in the field, in fact numbered only fifty.This undercut the
troops’ ability to mount a proper defence in depth. Food was in short supply.
There were no potatoes for front-line troops and they were only receiving
fresh meat nine days per month. With widespread exhaustion and hunger
lowering resistance, disease spread quickly. A first wave of the notorious
Spanish ’flu which ravaged Europe and America in 1918 and 1919 swept the
army from late June for a month. A second wave followed in September. In
Sixth Army alone, some 15,000 men were hospitalized.16
Meanwhile, Rupprecht was finding out for himself that there was plenty
of aggression left in his enemy. Soon after midnight on 2 July enemy aero-
planes raided his headquarters. The account Rupprecht left is vivid:
When the alarm sounded, I went down into the cellar, as I previously never
had. No sooner had I entered than I heard the buzz of the enemy aeroplanes
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

230 Y e a r of De f e ats 1918

and the whistle of dropping bombs. A dull thud, a violent crack! Suddenly the
doors from the shelter out into the garden blew open and down the stairs
stumbled the sentry who had been standing outside, a member of the
Landsturm, blown in by a bomb which exploded in the garden. I caught him:
he was bewildered, but unwounded.
The walls were shaking and the doors were banging backwards and for-
wards. We closed them.There was no doubt about it: the conservatory adjoin-
ing my bedroom was hit. You could clearly hear the sound of the glass roof
falling in. Captain Gerke, who was struggling to count the bombs, had just said
‘five’ when a bomb fell immediately on the shelter itself, which was installed
in the cellar under the stone staircase leading into the garden. The blast was
extremely strong. I saw the roof of the shelter buckle, then the light went out
and a thick cloud of dust and smoke blew through the doors which had once
again sprung open.
I grabbed my electric torch and checked that everything was all right. But
where was my adjutant, Captain von Hirschberg? We searched the cellar, then
heard his voice. He tottered into the shelter covered in blood. He had been
slow to get out of bed and was caught by the blast of an exploding bomb as
he neared the steps down into the cellar. He had bomb splinters in his legs and
was blown into the cellar, where he sustained other further injuries from splin-
ters of glass. We examined his wounds at once: luckily, they were light.
Already we wanted to leave the dark cellar, but then heard other bombs land
further away. In the morning it became clear that it was not only the house I
lived in and its outbuildings that were badly damaged. A bank was also hit and
reduced to rubble. The only person inside, a Belgian, was miraculously spared.
A building diagonally across from mine, which was occupied by Belgian nuns,
was likewise completely destroyed without its inhabitants coming to any
harm. An ill Franciscan sister was eventually dug out of the ruins of the
upstairs, after much hard work.
The story of the raid seems to be as follows. The enemy aircraft had followed
one of our bomber squadrons returning from a night flight to hide the sound
of their motors. Then they split up. Some bombed the nearest air defence
­batteries while the others attacked my house. It’s not clear how many enemy
squadrons were flying over Tournai that night, but in any case there were
­several. Bombs were also dropped near army headquarters. In the morning,
two squadrons appeared again, the first consisting of only three machines, to
inspect and photograph the damage.
I took my usual morning ride to the park at Froyennes for my exercise,
accompanied by Major von Behr. I always got back from there in time for the
mid-day briefing at the HQ building. This time, I luckily stayed a bit longer
than normal in the park. While I was still there, I heard firing from a nearby
anti-aircraft battery and immediately thereafter the whistle of a bomb, prob-
ably dropped on the flak battery. After the first impact came several others, bit
by bit moving away towards the city. As it turned out, the fliers flew along the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ope r at ion GEORGET T E a n d Sum m e r 1918231

road to the town and damaged some houses by the entrance to the city. It was
as if someone knew at what time I took this route every day. This suspicion
was close to the mark, as also in Douai, at the hour at which the general com-
manding Seventeenth Army usually journeyed from his residence to his HQ,
his route was bombed from the air.
When searches were carried out they turned up hidden wireless transmitters
in various places, used by spies to communicate with the enemy.

In all, four bombs fell on Rupprecht’s house or in his garden. Hirschberg

was hospitalized but there is no record of other casualties. Contrary to
Rupprecht’s suspicions, it is unlikely the RAF were targeting him directly
that night. Especially by night, they had no way of achieving that level of
accuracy at this time. It is more probable that he was caught up in a raid on
Tournai station, one of a series of experiments targeting the German ­railway
network.17
HAGEN never happened. When MARNESCHUTZ-REIMS began on
15 July, it did not go at all to plan. A total of nearly fifty German divisions
took part, supported by 6,400 guns, over 2,000 mortars, and 900 aircraft.
Although the artillery support thus came close to the level available on 21
March in absolute terms, the ratio of German to Allied guns was the lowest
of any of the 1918 offensives, at 2:1. Worse, the French defenders were well
dug-in and alert. Foch was actively preparing a counter-attack on the west
face of the Marne salient, designed to retake Soissons. This was planned for
18 July so the French were watching the sector very closely and picked up
indications of the impending German attack. For once, the Germans failed
to achieve surprise. In the face of intense enemy artillery fire, many of the
assault troops lost the cover of the creeping barrage. A few divisions man-
aged to gain toe-holds on the far side of the River Marne but found them-
selves very exposed there. Their artillery support was still struggling to get
forward. It soon became clear that Reims would not fall anytime soon.
Within less than thirty-six hours the German advance stuttered to a stop. By
the end of 17 July OHL was ordering the evacuation of the furthest bridge-
heads over the Marne and everywhere attackers became defenders. Within
hours, during the early morning of 18 July, the German troops manning the
western face of the Chateau-Thierry salient were attacked by General
Charles Mangin’s Tenth Army, supported by 346 tanks, swinging out of the
Forest of Villers-Cotterȇts and striking towards Soissons.
OHL suspended all transfers of men and resources for HAGEN. It des-
patched two of Rupprecht’s divisions, which he was keeping back for that
operation, down to prop up the defence in what is now known as the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

232 Y e a r of De f e ats 1918

Second Battle of the Marne. Although over the next fortnight of heavy
fighting the French failed to cut off and destroy the German salient, they
did push its defenders back across the Aisne and Vesle rivers. In all, seven
divisions earmarked for HAGEN had been diverted to the Marne. More
importantly, the French had made it clear that the initiative now lay, not
with the Germans, but with the Allies. Loßberg proposed an immediate
withdrawal into the Hindenburg Line. Ludendorff was not prepared to go
so far. Maybe he did not think it militarily necessary yet. Maybe he was
unwilling to acknowledge in public that his offensives had failed. He was
prepared, however, to cancel HAGEN, at least for now, despite objections
from Kuhl. This took place on 20 July. Kuhl noted tersely that ‘this is the
turning point of the war!’The next task for Rupprecht and his Army Group
would be not to attack, but to defend.18
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

22
The Hundred Days

I t is easy to forget now that, even after the Second Battle of the Marne
(15 July–6 August 1918), most people still had little clue how and when
the war would finish. This war had been unlike any other and had touched
so many so deeply. How could it possibly end? What might victory or defeat
even look like, when the cost in blood and treasure had been so high?
Could old-world diplomacy possibly silence the still-loud beat of the drums
of war? In the summer of 1918 to suggest that it might all be over by
Christmas would have seemed a joke in the worst of taste. Part of the story
of what we now know to have been the last few months of the First World
War is how and when the German military at last began to understand the
true nature and meaning of defeat.
In July and early August 1918 Rupprecht and Kuhl began doing what
they could to improve their defences against an Allied attack. In places,
they straightened the line, pulling out from some of the most vulnerable
positions. They were handicapped by a shortage of labour and transport,
and by poor morale. The 111th Division, for example, was exhausted. It had
been in the front line for 138 straight days, with just one twelve-day break.
It had lost nearly 6,000 men. Letters home were arguing for strikes and
revolution. One man in the sick bay wrote that he was being treated worse
than a convict or an animal. The replacement situation was so dire that
only Saxony still had men to spare: the King of Saxony decided to allow
his soldiers to be drafted into Prussian regiments, to Rupprecht’s predict-
able disapproval. Oberste Heeresleitung (OHL) reorganized command
arrangements. Rupprecht’s Army Group had become unwieldy: control-
ling five separate armies was too burdensome. A new Army Group was to
be created on 12 August under General Max von Boehn to take over the
three armies between the rivers Somme and Oise, including Rupprecht’s
Second Army.1
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

234 Y e a r of De f e ats 1918

Before that reorganization could take place, though, in the early morning
of 8 August, a ferocious Anglo-French assault began on Marwitz’s Second
Army east of Amiens. Spearheads from the Australian and Canadian corps,
supported by 430 tanks and over 2,000 guns, attacked along a 17-kilometre
front. Debeney’s French First Army soon joined in. Nearly 2,000 British
and French aeroplanes took to the air. German resistance crumbled fast. By
nightfall the British and French forces had driven up to 12 kilometres and
captured almost all their objectives. The Germans lost 27,000 men that day,
12,000 of them as prisoners. Kuhl’s diary describes how they were taken
completely by surprise and how the attack had simply overrun the burnt-
out defenders. It was, he concluded laconically, ‘a hard day’. Rupprecht put
part of the blame on a morning fog which blinded his anti-tank guns and
allowed enemy armour and infantry to work closely together. Surprise was
another important factor. As at Cambrai, British guns had fired no preliminary
bombardment and Rawlinson’s Fourth Army did a good job of concealing
its other preparations.2
The next day, the pace of the British advance slowed, gaining only
another 5 kilometres. Although some German reinforcements had come up,
the resistance Marwitz could offer was still very limited. Again, however, the
British had not planned adequately for the second day of operations. They
were unable to coordinate their actions effectively and launched a weak
series of piecemeal attacks. Nevertheless, a worried Marwitz proposed
falling back behind the Somme. Ludendorff responded by firing Marwitz’s
chief of staff and ordering him to hold firm. On 10 August the British
proved Ludendorff right.They struggled to make another 1,500 metres and,
after another day of disappointing results on 11 August, Haig and Rawlinson
suspended the Amiens operation. The British had suffered casualties of
about 20,000 men; the French a similar number; and the Germans 48,000.
Of those, no fewer than 30,000 had been taken prisoner.3
The Battle of Amiens has tended to occupy a more prominent place in
histories of 1918 than, say, the Second Battle of the Marne, especially in the
Anglo-Saxon world. This is partly because the British Expeditionary Force
(BEF), rather than the French, took the leading role. Partly also, however, it
is because the Battle of Amiens is often seen as the beginning of the German
army’s final decline. The German official monograph argued that ‘as the
evening of 8 August settled over the battlefield of the Second Army, the
heaviest defeat of the German army since the beginning of the war was an
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days235

established fact. . . . The first steps along the dark road through the forest
of Compiègne to the Hall of Mirrors at Versailles were taken with the
catastrophe of 8 August 1918.’ In his memoirs Ludendorff famously
called 8 August ‘the black day of the German Army in the history of this
war’, primarily because it showed how far the combat effectiveness of the
army had fallen. ‘The war’, he decided, ‘must be ended.’ His most recent
biographer, however, argues that although he offered to resign, he was far
less convinced that Germany had lost the war than he later pretended.
Defeats, for Ludendorff, remained the result of poor individual decisions
rather than systemic failings. He blamed Wetzell for the Marne and
Marwitz’s chief of staff for Amiens. He was constantly on the look-out
for scapegoats and redoubled his efforts to win the war single-handed.
Rupprecht was
amazed at the patience of my chief of staff, who doesn’t let himself lose his
cool at the endless telephone calls from Ludendorff, even when he’s trying to
lay down what even individual battalions . . . should do. To keep Ludendorff
calm he always agrees with him or at most says ‘that depends on how things
play out, we can’t predict that yet’, or so on. The telephone has some draw-
backs: it gets used too often, and in direct communications between OHL and
the chiefs of staff of armies and corps, the commanders get totally shut out.
Just recently one army commander said to me in this regard: ‘I no longer really
know what I’m there for, since everything has always been decided before
I am consulted.’ Ludendorff ’s impatience is to blame for much of this, however
understandable it may be in the current circumstances. If only Ludendorff
would not ring up every single corps direct, as well as the army group and
army chiefs of staff!

Ludendorff ’s nerve was clearly cracking, his screaming at his staff officers
audible throughout his headquarters. He was changing chiefs of staff ‘like
underwear’ and developing the shakes.4
It was not just Ludendorff feeling the strain. Rupprecht was finding it
hard to recover from a bout of influenza and his spirits, too, were low.
He  could not see how Second Army would be able to hold out. On
13 August he went back to Bavaria for a couple of weeks to recuperate, but
his mood did not improve much. He wrote to Prince Max that ‘I no longer
believe that we can hold out over the winter.’ Rupprecht spent his leave
fishing with his son and stalking stag in the Alps. He also acquired a wife
thirty years younger than himself: Princess Antonie of Luxembourg. Since
July 1917 he had been spending time with the young Grand Duchess of
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

236 Y e a r of De f e ats 1918

Luxembourg and her five sisters, aware that his son needed a mother and
the Wittelsbach dynasty needed a spare heir. The fourth sister, Antonie,
seemed keenest and got on well with her future stepson, Albrecht. The
engagement was announced on 25 August.5
By the time Rupprecht got back to his headquarters on 2 September, the
situation had deteriorated badly. While he was away, Foch had orchestrated
a sequence of attacks up and down the Western Front. The French struck
one day, the British somewhere else the next, and the Americans on a third
in a different sector again. As part of this, Rupprecht’s Army Group came
under British attack between Arras and Albert on 21 August. Employing
similar tactics to those used at Amiens, the BEF drove forwards 5 kilometres.
To the south Rawlinson’s Fourth Army soon resumed their push along the
Somme. The whole right flank of the British army began chewing through
the German defences across the old Somme battlefield towards Bapaume
and Péronne. In all, seven British and French armies were locked in battle
with five German ones over 200 kilometres of the Western Front, from Lens
almost all the way to Reims. Not since the autumn of 1914 had the fighting
been on such a scale.
The strain on the Germans was intense. In Kuhl’s view:
We stand now at the crisis. Questions of prestige have no role to play. Cold,
sober calculation is all. We can’t pull the wool over our people’s eyes. We can’t
force a decision. We are worn out. The divisions are in disorder. We can only
fight a delaying action, with skilful retreats, losing ground in some places and
holding it elsewhere. I wish I was wrong.
The war could no longer be won, but might it still be possible to force
a  draw? Kuhl envisaged fighting a long delaying action, slowly pulling
back from defensive line to defensive line. In this way the Allies could be
prevented from reaching the Rhine until sometime in 1919. This would
require coherent operational planning if it was to work, but Kuhl saw two
obstacles to this taking place. Loßberg, now Boehn’s chief of staff, was one:
he would have to be involved, especially given the influence he had with
Ludendorff, but he was a tactician, incapable of thinking operationally on the
scale required. Rupprecht’s absence in Germany would also make decision-
making harder.6
Through the last days of August, German resistance was fierce. Nonetheless,
the Allied armies ground inexorably forward. Progress was slow: it took the
New Zealand Division eight days to drive 12 kilometres and liberate the
ruins of Bapaume, for example. This was lightning fast compared with earlier
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days237

years, but at this rate it might indeed take nine months to reach the Rhine,
assuming they did not run out of manpower first. Boehn’s attempts to hold
the old German reserve position from 1916 along the Bapaume–Péronne
road failed. He was outflanked by Australian troops to the south at Mont
St  Quentin and Canadians driving down the Arras–Cambrai road to the
north. OHL accepted the inevitable and ordered Boehn’s and Rupprecht’s
Army Groups to fall back into the Hindenburg Line. Finally Ludendorff told
Hindenburg that ‘we no longer have any prospect of winning the war’.7
By 7 September the Germans were back where they had begun the year.
The Allies began closing up to the Hindenburg Line and positioning all
the men, machines, and munitions they would need to assault it. While
preparations went on, relative calm descended on the front. The Germans
had recoiled up to 50 kilometres from their summer high-water mark.
Losses on both sides had been heavy. British Third Army had lost 23 per
cent of its infantrymen, while 29 per cent of German Seventeenth Army’s
combat troops had become casualties. Over half of the German soldiers
went missing in action, most of them ending up fit and healthy but in
British hands.This was partly because open fighting made it easier to cut off
and capture individuals and whole units. Partly, however, it was also a symp-
tom of declining morale. Other signs of defeatism came across in the mail
censorship summaries, in increased desertion and ‘shirking’ behind the lines,
and even in reports of whole units running away at the first rumour of an
enemy. The story of the morale of the German army during late 1918 is
complex and historians are still debating it, but all can agree that it was
significantly worse than it had been earlier in the war.8
On 6 September Hindenburg, Ludendorff, and some of the chiefs of staff
met to discuss what had gone wrong and what to do next. Ludendorff
looked ‘completely ill and exhausted’. He argued that MARNESCHUTZ-
REIMS had been defeated because security had failed and the French had
been expecting it, while conversely the Germans had let themselves be
taken by surprise by the French counter-attack. More recent setbacks in
Picardy were the result of poor leadership. ‘So it is everyone’s fault—except
OHL’s’, Kuhl noted sourly: ‘in reality, it’s the other way around’. OHL’s
view was superficial but some units carried out sophisticated analysis which
gives a better explanation of the recent setbacks. For example, the 3rd Marine
Division drew four conclusions from the fighting for Bapaume (21–31 August).
First, German infantry was weaker than before, particularly in morale. Second,
defensive doctrine was too complicated and cumbrous. Using multiple
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

238 Y e a r of De f e ats 1918

different names for each defensive line was causing confusion. Third, divisions
were being committed piecemeal, mixing up units and reducing cohesion.
Lastly, ‘tank shock’ was a problem. Units were dissolving at the sight of
enemy armour. Poor leadership, of the kind Hindenburg and Ludendorff
were fixated on, was only one factor among many.9
The German supreme command was not entirely blind to the tactical
problems front-line troops were facing. Between July and September OHL
issued eleven different memoranda tweaking defensive tactics to counter
new Allied methods.There were multiple problems with how it went about
this, however. First, it once more assumed that its doctrine was perfect and
defeats must be the result of failure to follow instructions properly. There
was little evidence for this. Second, defence in depth required manpower
that was no longer to hand. The average infantry battalion had begun 1918
with 850 men. By the end of September this was down to 540. Even
achieving this strength had required breaking up no fewer than 22 whole
divisions. Third, as we have seen before, the standard of instant communica-
tions assumed in OHL’s instructions was impossible given the technology
available and conditions on the battlefield. Fourth, setting up field guns near
the front line to kill tanks proved highly effective when visibility was good.
In fog, however, they were useless and quickly overrun. More importantly,
detaching guns in this way distorted and unbalanced the overall defence.
Lastly, and most significantly, there was a hidden assumption underpinning
OHL’s analysis that German troops were better than the enemy. So, if they
were beaten, they must have done something wrong. The possibility that
the enemy might be doing something right was never admitted. Now, of
course it made sense to concentrate effort on improving things the Germans
could control, rather than those they could not. Moreover, to accept that the
British or French had got better and developed tactics to which the Germans
had no answer might damage morale. Nonetheless, failure to face the tactical
facts infected and prejudiced operational and strategic decision-making.
Wishful thinking was no basis for sensible planning. Tactically, the Germans
were being outfought every time and they had run out of useful responses.
On 21 September OHL revised its tactical doctrine again. Units were no
longer to defend elastically but were to fight in, and at all costs hold, the
main line. The wheel had turned full circle, back to the tactical rigidity
Falkenhayn had espoused in 1915 and 1916. Operationally, too, the Allies
were posing problems all up and down the front too quickly for the Germans
to respond, even had the resources to do so been available. The only concrete
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days239

decision reached on 6 September was to try more of the same of an

­operational recipe that patently was not working: they would hold in the
Hindenburg Line as long as possible to buy time to build the ‘Hermann
Line’ 15–20 kilometres further back.10
The time had come to address the increasingly obvious problem of Erich
Ludendorff himself. Doctors prescribed more sleep and relaxation to coun-
ter his nervous strain. Boehn went to see him on 5 September, on behalf of
all the commanders of armies and Army Groups. He asked Ludendorff to
stop short-circuiting the chain of command and interfering so aggressively
in day-to-day operational details. Ludendorff promised to behave better. He
also agreed to a reorganization of his staff.Wetzell, after two years as chief of
operations, was out. Colonel Wilhelm Heye became Ludendorff ’s deputy.
Speculation that Ludendorff himself might be replaced was widespread. For
a while, Ludendorff was calmer. Before September was out, however, he was
back to his old tricks, sacking army commanders he felt had let him down,
including Marwitz from Second Army, and fiddling with the deployment of
the anti-tank defences of one of Rupprecht’s corps.11
The middle of September was taken up with battles for the outposts of the
Hindenburg Line at Havrincourt (12 September) and Épehy (18 September)
as well as the American clearance of the St Mihiel salient (12–16 September).
The Germans remained uncertain about where the Allies would strike next.
Rupprecht was still most concerned about his left flank, south of Cambrai.
He could not understand why the British were not being more aggressive
here. He also wondered, however, whether Haig might assault Sixth Army
north of the River Scarpe. Later in the month he began looking for reinforce-
ments to send to Fourth Army, having received intelligence of an imminent
enemy attack on the Flanders coast. A simultaneous attack on all three armies
seemed possible and Kuhl was well aware that they did not have the strength
to withstand that. Ludendorff was also worried about threats at Verdun, in
Alsace, and in Lorraine. No part of the front seemed safe.12
One huge, coordinated offensive was precisely what Allied supreme com-
mander Ferdinand Foch had in mind. On 26 September the largest battle of
the Great War opened with a Franco-American drive between the Meuse
and Reims towards the rail junction at Mézières. The next day, the British
attacked west of Cambrai, crossed the Canal du Nord, and established
bridgeheads across the St Quentin Canal. Belgian, British, and French
troops launched an offensive around Ypres on 28 September. Finally, on
29 September a Franco-British force attacked the Hindenburg Line around
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

240 Y e a r of De f e ats 1918

St Quentin. Bitter fighting flared from Flanders to the Meuse. Not one of
these attacks created a clean break in the German lines. Indeed, early results
were disappointing. After three days of savage combat in the Argonne Forest,
American forces had only advanced 12 kilometres at best. General Pershing
suspended his attack on 29 September after losing 45,000 men.The drive on
Cambrai, similarly, ran out of steam by the end of the month.The Australian,
American, and British troops of Rawlinson’s British Fourth Army managed
to break in to the Hindenburg Line on 29 September, but it took them
several bitter days of combat to clear the Germans out. Only at Ypres did
the Allies win quick success. Within forty-eight hours they had cleared
the ridges up which the British had toiled so painfully in 1917 and were
heading down the far slope towards the rail lines at Roulers. The Flanders
operation only ground to a halt when the difficulty of transport over the
shattered battlefields of the salient choked off supplies and prevented further
progress. Foch’s plan, however, did not depend on a breakthrough at Ypres
or indeed anywhere. He was betting that the cumulative effect of these
attacks would exceed the sum of the parts.13
Foch was proved spectacularly correct. At 6 p.m. on 28 September
Ludendorff and Hindenburg agreed that it was time to ask for an armistice:
now, without delay. The next day they informed the Kaiser and decided to
form a new government to approach President Woodrow Wilson and
request a ceasefire. In his memoirs, Ludendorff tried to maintain that it was
not the military situation on the Western Front, but the imminent collapse
of Bulgaria, which provoked his decision. This was a transparent attempt to
deflect blame. Both Kuhl and Rupprecht thought little of this spin at the
time and the briefing Ludendorff gave to his staff makes clear that he was
desperate rather to forestall the collapse of an already unreliable army. He
wanted the army to remain, at a minimum, strong enough to crush any
Bolshevik revolution back home. If the army were defeated in battle, not
only might it not be strong enough to do so but it might even precipitate
the very revolution it was supposed to be guarding against.
Also on 28 September a telegram had arrived at Rupprecht’s headquarters
from Hindenburg warning all Army Groups that OHL was no longer in a
position to send them any more reserves. Manpower was not available and,
even if it had been, the transport situation prevented movement. Even
before he heard of the decision to put out peace feelers, Rupprecht had
been deeply worried: ‘our situation has further deteriorated and is now
critical as never before’, he wrote. The only way to save Fourth Army in
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days241

Flanders, he thought, would be to retreat to Ghent, even at the cost of

giving up the U-boat bases along the Belgian coast. Even Loßberg,‘the man
with nerves of iron’, had lost hope and was reporting that Army Group
Boehn was at the end of its strength. Rupprecht told his father that ‘the mili-
tary situation has got worse once again’. OHL was still playing the ostrich
game. Rupprecht re-emphasized the point in another letter: ‘What I have
been afraid of already for a long time, has now come to pass as I expected:
the decisive defeat. . . . We must definitely achieve peace at once and at any
price, since we are as good as defenceless. . . . It is the decisive moment in
Bavarian history since 1866. If only I could be at home!’ The same day, Leeb
wrote in his diary that ‘the campaign is lost’. Not everyone agreed yet, how-
ever. When Ludendorff rang Kuhl to admit defeat and tell him about the
request for an armistice, Kuhl’s first reaction was that this was premature and
that they should at least try to make one more stand before giving up.14
The new Chancellor, Rupprecht’s friend Prince Max von Baden, des-
patched a Peace Note to the United States overnight on 4/5 October. Even
now, Ludendorff harboured unrealistic expectations of the Allies’ likely
demands, under-estimating their determination to ensure that Germany
was in no position to resume hostilities. He thought he had more room for
negotiation over the territory Germany still occupied than was ever likely
to be the case. He told Kuhl that it would take nine months for German
troops to evacuate Belgium and France. As Kuhl sardonically noted in his
diary, ‘I was afraid that the enemy might not give us so much time.’15
By 4 October the Germans had been pushed back into the reserve posi-
tions of the Hindenburg Line. To overwhelm this last belt of defences the
Allies prepared another set-piece attack.While they did so, there was another
brief lull in operations, during which Rupprecht slipped away to Luxembourg
to visit his fiancée. On 8 October came news that the British were attacking
once more. After heavy fighting and the last German counter-attack of the
war, eventually the British smashed through into open country. Canadian
forces finally liberated Cambrai in the morning of 9 October. By then,
Rupprecht’s armies were in full retreat back to the so-called Hermann Line,
a sketchy set of defences on the east bank of the River Selle. Again, the Allies
rolled forward and began to prepare a full-scale assault. By 11 October,
Rupprecht’s Army Group reserves had run out. He could offer his armies no
further reinforcement.16
Pessimism spread through the German command. On 8 October Boehn
and Loßberg reported that their troops were no longer capable of putting
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

242 Y e a r of De f e ats 1918

up a defence. This gave Ludendorff an excuse to sack both men and break
up their Army Group. Three days later, the commander of Seventeenth
Army, Otto von Below, also buckled. He telephoned Kuhl, insisting that his
troops must retreat at once. To avoid leaving a gap in the line, and since
anyway it was too late at night for such a manoeuvre, Kuhl said no. In that
case, Below threatened, he would march anyway, orders or no. Insubordination
of this kind could not be tolerated. Both Below and his chief of staff were
rapidly sacked.17
The state of morale lower down the army is harder to untangle. There is
evidence that, already depressed by the failure of the spring offensives and
the defeats of July and August, it entered a final crisis, especially among rear
area troops, after news of the Peace Note to President Wilson. Desertion
and shirking were problems, albeit probably not on the scale some historians
have suggested. Equally, however, there are indications that, particularly in
some front-line units, spirit remained remarkably firm even in October.
Better food, now that supply lines were shorter, played a major role in
improving mood and the number of men taken prisoner declined in both
absolute terms and as a proportion of total casualties. The slower tempo
of  operations in October made the situation appear less critical. Wilson’s
second note, which arrived in Berlin on 16 October, provoked resentment
and stiffened resolve in some quarters. The idea that the German army was
a morally beaten force, even after the Peace Note, is an over-simplification
and it is the resilience, not the vulnerability, of the German army, even into
the dying months of the war, that is one of its most striking characteristics.18
On 17 October the Allies went on the offensive once more, clearing the
Belgian coast and liberating towns such as Bruges and Courtrai. Further
south the British marched into Douai and Lille, although an attempt to
bounce across the River Selle at first bogged down. Still, by 23 October the
British were advancing on Valenciennes and the Forest of Mormal, while
the French and Americans continued to grind forward in the south.
After weeks of increasingly erratic judgement and behaviour, Ludendorff
was finally removed from his post on 26 October. Within a fortnight of
demanding an immediate armistice he had begun to argue that the situation
had improved, that Germany could afford to take a hard line, and that it
might make sense to fight on into the new year. The final crisis came when
he issued an order of the day to the army which rejected peace negotiations
and so directly contradicted the civilian government. Max von Baden
refused any longer to serve alongside Ludendorff. The Kaiser could hardly
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days243

lose two Chancellors in less than a month, especially with his own position
and prestige with the population at large falling fast. He accepted Ludendorff ’s
resignation. Even Hindenburg made no more than a token effort to save the
man who had been his partner for more than four years. The two men never
spoke again. Ludendorff suggested Kuhl as his successor but the job went
instead to Wilhelm Groener, a railway expert, chosen because the government’s
priority was now demobilization, not fighting on.19
On 30 October Kuhl and some of his other staff officers managed to
escape their new headquarters in Brussels for a few hours’ sightseeing on
the battlefield of Waterloo, but from 1 November renewed Allied attacks
shattered any relative quiet. Ghent and Valenciennes fell in the north while
American and French forces resumed their drive towards the Meuse at
Sedan and Mézières. On 4 November three British armies attacked on a
65-kilometre front and crossed the River Sambre. Rupprecht’s report to
OHL was bleak:

The heavy fighting of 4 November on the fronts of Second and Seventeenth

armies caused heavy wastage in personnel. Units on the front line, whose
combat power had already been seriously reduced before the battle, have
mostly been weakened to the limit by the losses they suffered in this heavy
fighting. It is no longer possible to supply combat-capable reserves to the
front. . . . The commanders of Seventeenth and Second armies dutifully report
that, because of exhaustion, low unit strengths, and particularly the limited
steadiness of their men, they are not capable of withstanding a new major
offensive.There is a danger that a new large-scale attack would break through,
or at least cause us heavy losses, and that we would have to start the move to
the Antwerp-Meuse line in considerable disorder.

Soon after midnight, he ordered a general retreat. The rearguards detached

to slow the British advance did a good job of holding up the enemy. Poor
roads, worse weather, and an over-stretched supply network delayed the
British even more and by 8 November they were losing all touch with the
retreating Germans.20
Rupprecht’s units were not the only ones on the brink of collapse. The
navy mutinied in Kiel on 3 November and within days soviets of sailors,
soldiers, and workers formed all over Germany as revolution spread. The
German government hurried to send a delegation to discuss armistice terms
with Foch. On 7 November the negotiators arrived at the French lines near
Guise and were led to Foch’s train, parked near the village of Rethondes in
the forest of Compiègne. Several days of long and blunt talks ensued. While
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

244 Y e a r of De f e ats 1918

they went on, Germany descended into chaos. As soon as it became clear
that the army would not fight for his throne, the Kaiser abdicated and fled
into exile in Holland. On 9 November, Germany was proclaimed a repub-
lic, with the Socialist Friedrich Ebert taking over in Berlin as Chancellor.
By now, the Allies were standing on the banks of the Rivers Scheldt and
Meuse. On 11 November British troops reached the grimy Belgian town
of Mons, where their war had begun. Early that morning, the German
delegation finally signed the armistice and by noon the war, at long, long
last, was over.
By then, Rupprecht’s war had already finished, cut short by revolution in
Munich. On the afternoon of 7 November, as many as 60,000 people
packed onto the Theresienwiese fields in the south-west of the city, where
the Oktoberfest is now held. Overlooked by a huge bronze statue of Bavaria,
speaker after speaker demanded the immediate abdication of Kaiser Wilhelm
II and his son, plus a package of political and social reforms. No mention
was made of the Wittelsbachs.The demonstrators decided to process through
the government quarter to the Angel of Peace monument, high on the bank
of the River Isar and, as they marched, red flags appeared in the crowd. From
the steps of the Feldherrnhalle, the monument built outside the royal palace
to commemorate Bavaria’s military past, agitators harangued the marchers,
demanding an end to the monarchy. The leader of the Independent Social
Democrats, Kurt Eisner, began to seize control of armouries and barracks
throughout the city, persuading the soldiers to join the revolution. Resistance
was slight. The War Minister had no troops on whom he could rely and
soon even the royal guards at the palace began to slip away. By eight o’clock,
Eisner’s small but growing band controlled the barracks, the main railway
station, the post office, and government buildings. A Soldiers’ and Workers’
Soviet set itself up in the Mathäser Beer Hall. It was the very model of a
modern coup d’état: fast, effective, and mostly peaceful.
The masses on the streets of Munich made no attempt to storm the palace
or detain the royal family. The writer Thomas Mann found the atmosphere
peaceful and celebratory, rather than menacing. If there was any political
sentiment, he thought, it was pro-Bavarian and anti-Hohenzollern. The
government, however, was worried that revolutionaries might rush in
during the night and round up the royal family. They advised the King to
leave town until the situation became clearer. Later that night, three cars
slipped out of the royal mews carrying the King and Queen, four princesses,
Rupprecht’s son Albrecht, and various members of staff. After a dramatic
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

T h e H u n dr e d Days245

Figure 22.1.  The demonstration of 7 November 1918 on the Theresienwiese, Munich

night drive through thick fog, in the early morning of 8 November they
took temporary refuge at one of their countryside castles on the road to
the Austrian border. Within hours, Eisner proclaimed that the Wittelsbachs
had been deposed and a republic set up: ‘the Free State of Bavaria’. When
Ludwig heard that revolutionaries were on their way to force him to abdicate,
he fled, first to a hunting lodge in Berchtesgaden and then across the
Austrian border to Salzburg. On 12 November 1918, Ludwig III signed a
declaration releasing all officers of the Bavarian army from their oath of
loyalty. Although he never formally abdicated, thus ended over seven
centuries of Wittelsbach rule.21
About nine o’clock in the morning of 8 November, Rupprecht received
his first news of the revolution from Munich. The telephone call was inter-
rupted and information remained fragmentary for some time, but it was
clear that no troops could be relied on to restore order and there was little
Rupprecht could do from so far away. He issued a proclamation to all his
Bavarian troops, claiming that the King was acting under duress and the
government did not represent the majority. He called for a national convention
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

246 Y e a r of De f e ats 1918

to draw up a new constitution and a free vote to redefine the relationship

between people and the royal family. Rupprecht evidently still saw a future
for himself and his family in the government of a new Bavaria.
In the short run, however, it was possible that he was in personal danger.
Soldiers’ soviets were being set up throughout the army, including in
Brussels, the site of Rupprecht’s headquarters, just as they had been in Russia
in 1917. The Bolsheviks’ murder of the Tsar and his family was fresh in
Rupprecht’s memory. He may not yet have known that his father was
fleeing into Austria, but he did hear of the Kaiser’s flight on 10 November.
The city of Brussels was sliding into anarchy as armed mobs of soldiers
roamed the streets seeking loot, and it was becoming dangerous for
Rupprecht to stay. He was, in any case, according to one of his staff officers,
a ‘broken’ man. When Hindenburg ordered him to cooperate with the soviets,
he could not bring himself to comply. Instead, he burned his personal
papers, turned over command of the Army Group to Sixt von Armin, bade
an emotional farewell to his staff, and took refuge in the Spanish Embassy
late on 10 November.22
In some ways one almost feels that Rupprecht’s war had already been over
for a while. His behaviour during the Hundred Days was rather detached at
times. He was absent on crucial occasions and was less determined to organ-
ize the defence and influence the campaign than he had been. His influenza
did not help. Even after Rupprecht recovered from that illness, he continued
to be troubled with insomnia and migraines. It is not hard to diagnose
exhaustion and stress. Family was a further distraction. He was surely too
experienced to be carried away by his new fiancée, but relations with his
father remained an irritant. More importantly, one suspects he was demotivated
by Ludendorff ’s habit of marginalizing Army Group and commanders and
working through staff officers instead. The last occasion Rupprecht and
Ludendorff met in wartime was on 18 July. More than anything else, how-
ever, one suspects that Rupprecht had simply realized that the situation had
reached the point where there was no longer anything he could do.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

23
Rupprecht on the Run

T he Spanish ambassador provided Rupprecht with a car and a passport

into the Netherlands under a false name. Travelling as Mr Landsberg,
presumably to avoid getting interned or embarrassing the Dutch govern-
ment, Rupprecht crossed the border on 12 November. In Amsterdam he
stayed with an art connoisseur friend and studied pictures in the Rijksmuseum
while he waited for news of the political situation back home. Eventually
the government guaranteed his safety and, still incognito, he travelled back
to Bavaria. A rising wave of political violence soon forced the Wittelsbachs
abroad once more, however, and it was not until the autumn of 1919 that he
was able to settle permanently back home. He finally married Princess
Antonie of Luxembourg in April 1921.
Although she bore him a son and five daughters, the marriage seems
not to have been a happy one. The couple spent increasing amounts of
time apart as the years rolled by. Perhaps old habits died hard. Perhaps the
Crown Prince was unable to meet the expectations of a much younger
bride. Neither case would be the first or last of its kind in history. When
his father died in October 1921 Rupprecht was able finally to agree a
settlement with the government over the status and finances of the royal
family. The Wittelsbachs received a lump sum and possession or resi-
dence rights in a number of palaces, castles, and estates. The state took
over the rest.1
One threat hanging over Rupprecht was that of prosecution for war
crimes. He featured at number 33 on a list of 895 alleged war criminals
published in December 1919, facing accusations of ordering the murder of
British prisoners of war, of the destruction of villages and execution of
innocent civilians in Lorraine in 1914, and of the burning of Cambrai during
the German retreat.The whole war crimes process provoked testy disagree-
ments between France, Britain, and Germany, and collapsed soon after it
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

248 Y e a r of De f e ats 1918

began in Leipzig in May 1921. Ironically, the case which led the French to
withdraw was that of Major-General Stenger, accused of issuing a ‘take no
prisoners’ order in August 1914 while under Rupprecht’s command. With
the French gone, the prosecutions lost any teeth they might ever have had.
In January 1922 the Inter-Allied Commission declared the Leipzig process
‘highly unsatisfactory’ and, although war crimes trials of a kind continued
until 1933, it became increasingly clear that none of the accused would be
found guilty by a German court. The charges against Rupprecht were for-
mally dropped in June 1923 and he was free to travel all over Europe, includ-
ing to England and even France, as he had before the war.2
The death of his parents left Rupprecht the pretender to the thrones of
both Great Britain and Bavaria. The British claim could be dismissed: he
had already made clear that he had no intention or chance of becoming
King Robert I and IV. The Bavarian issue, however, had to be treated with
care. For the rest of his life, Rupprecht styled himself in public as ‘Crown
Prince’ rather than ‘King’.There were plenty of monarchists keen to restore
a Wittelsbach to the throne. Inevitably, they were concentrated on the pol-
itical right. The number of true conservatives looking only to wind the
clock back was relatively small. Most people, including Rupprecht, realized
that the obstacles to returning the sweeping powers that his predecessors
had enjoyed were immense. More often, the monarchy was seen as more of
a symbol and embodiment of Bavaria. Rupprecht seems to have shared that
view: while keen to take up the crown, it remained important to him that
he do so only by legal means and to popular acclaim. In theory seeing the
monarchy as rising above faction and politics, he was not prepared to let
himself be identified with any particular party.3
His attitudes, sympathies, and interests, however, left him much more in
tune with the right than the left. During the economic and political crisis
of 1923, amid rising political violence, dissatisfaction with the weakness of
the Weimar government set off rumours of putsches against the govern-
ment. Rupprecht backed the declaration of martial law and the installation
of the nationalist Gustav von Kahr as a kind of dictator in September 1923.
Kahr was to form a united front of relatively moderate Bavarian ‘patriotic’
groups to forestall the extremism of radicals such as Adolf Hitler’s Nazi
Party. One of Kahr’s policies envisaged expelling the so-called ‘Ostjuden’,
recent Jewish refugees, largely from Poland.This was both a convenient stick
with which to beat Berlin and reflected genuine anti-semitism. On both
grounds, Rupprecht supported the deportation policy.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t on t h e Ru n249

Figure 23.1. Crown Prince Rupprecht and Princess Antonie with their family
between the wars

It is hardly surprising that anti-semitism constituted part of the mental

furniture of a Bavarian aristocrat born in 1869. It was equally predictable
that he should find the vulgarity of the Nazi Party unappealing. Rupprecht
and Adolf Hitler certainly met, in the summer of 1922, but the ex-corporal
made little impression on the former field marshal. Rupprecht might have
seen the Nazis as possible anti-socialist allies but their pan-German nation-
alism did not attract him. He was too committed to preserving the rights of
individual states within the federation. On the other side, Hitler’s interest in
the monarchy was tactical and pragmatic. If he could exploit it to help
achieve his own objectives, he was prepared to do so, but he had no real
commitment to either the Hohenzollerns or the Wittelsbachs. Indeed, he
viewed both dynasties as degenerate. Men in his entourage, however, such
as the head of the Brownshirts, Ernst Röhm, remained keen to bring the
Führer and the Crown Prince closer together. As the strength of the far
right, led by Hitler and Ludendorff, grew during 1923, Rupprecht did hold
meetings with the latter. Their relationship was no easier than it had been
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

250 Y e a r of De f e ats 1918

during the last year of the war. Rupprecht attacked him in public, pointedly
remarking that ‘not every general can be a statesman like Frederick the
Great’. Although the Crown Prince declared himself willing to receive
Hitler once more, that meeting never took place.4
On the evening of 8 November 1923 Hitler stormed into a meeting
attended by Munich’s politicians in the Bürgerbräukeller. In what became
known as the Beer Hall Putsch, Hitler, claiming to be righting the wrong
done to the monarchy in 1918, held Kahr and other members of the gov-
ernment at gunpoint and forced them to endorse the Nazis’ plan to seize
power in Munich, march on Berlin, and overthrow the government there.
The whole scheme was poorly planned and incompetently executed and it
quickly began to lose momentum. As soon as Kahr regained his freedom, he
repudiated the putsch. Hitler and Ludendorff tried to fan it back into life
next morning with a march on the War Ministry but their followers found
their way blocked by loyalist policemen on the Odeonsplatz.There, in front of
the Feldherrnhalle, shots rang out. When the smoke cleared, four policemen
and sixteen Nazis lay dead. Hitler escaped but was arrested a couple of days later.
Rupprecht was out of town at his castle in Berchtesgaden, near the
Austrian border, probably seeking to put distance between himself and a
coup of which he had some foreknowledge, although we will probably
never know for certain the depth of his involvement. Even if Rupprecht
had a broad idea of the plotters’ general intentions, he probably had little
warning of the attempt itself, not least because it was put together only at
the last minute. The next day, when he was told what had happened, his
primary concern seems to have been to avoid bloodshed, rather than to
intervene on his own account. Later, Hitler bitterly blamed Rupprecht for
the failure of the putsch, but that hardly proves that the Crown Prince had
let the conspirators down. Hitler under stress was quite capable of lashing
out at even the most innocent, as his subsequent career amply demon-
strated. Ludendorff accused the Catholic church and royal family of cook-
ing up an unholy alliance to smash the putsch and claimed Rupprecht
himself had broken his word. Rupprecht’s attitude to the conspirators, dur-
ing the trial which followed, was ambiguous. On the one hand, he was in
favour of trying them thoroughly, which suggests he had little to hide. On
the other, though, he also argued that they should be treated clemently. He
felt Ludendorff was the prime mover and main danger: Hitler was no more
than Ludendorff ’s stooge.5
Rupprecht was far from the first person to under-estimate Adolf Hitler.
Tragically, he was also not the last. In 1923 Rupprecht had little reason to
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t on t h e Ru n251

fear the enmity of a failed rabble-rouser now locked up in prison. By the

time the gaol-bird had moved into the Chancellery in January 1933, how-
ever, there was plenty of reason to be scared. Politicians in Bavaria, across the
political spectrum, identified the threat the Nazis posed and scurried to find
some way to protect the rights of Bavaria and themselves against a Hitler
dictatorship. One option might involve Rupprecht in some capacity. The
Crown Prince’s prestige was high. He had spent more than a decade estab-
lishing himself as a king without a crown, working with Bavarian veterans’
organizations and touring the country. His popularity underpinned renewed
ideas of restoring the monarchy in some form. When he visited the Opera
soon after Hitler came to power, he was greeted with applause and the
­audience roared out the old royal anthem. In a measure of the growing
­desperation of those opposed to the Nazis, even some Socialists were com-
ing around to the idea of installing Rupprecht as State President or in the
emergency post of State Commissioner. Prime Minister Heinrich Held,
however, was unwilling to concede the all-party government of national
unity which Rupprecht demanded.
While the politicians talked and talked, the Nazis acted. Having secured
the loyalty of the army and emboldened by success in the plebiscite of
5 March, the Nazis seized control of the Bavarian government and simply
tore up the old constitution. Rupprecht’s protests to the German President,
Hindenburg, received no more than a polite acknowledgement. Within a
fortnight, the concentration camp at Dachau, a short tram ride from down-
town Munich, opened its gates. If Rupprecht’s restoration had ever been a
serious option, it never was again.6
That Hitler feared Rupprecht’s prestige, and had neither forgotten nor
forgiven him for the collapse of the Beer Hall Putsch, was soon made dev-
astatingly clear. On 13 March four men from Rupprecht’s inner circle were
arrested. One remained incarcerated for over a year. In autumn 1933, his son
Albrecht was denounced as an enemy of the Nazis and was only saved from
Dachau by the intervention of Ernst Röhm. The murderous ‘Night of the
Long Knives’ in 1934 was further proof, if any were needed, of the Nazis’
violent lack of scruple: among the many victims was Gustav von Kahr,
killed for his ‘betrayal’ of Hitler in 1923. Nonetheless, Rupprecht would not
allow the Nazi swastika to be flown outside his Munich palace and continued
to appear in public into 1935, mainly at church or veterans’ events. Hitler
and Rupprecht took care to avoid each other when their paths might other-
wise have crossed. When the Nazis established their summer playground in
the Alps above Berchtesgaden, Rupprecht stopped using his castle there, and
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

252 Y e a r of De f e ats 1918

during the second half of the 1930s he retreated into ‘internal exile’. He did
his best to protect his children from Nazi indoctrination, sending his son
Heinrich to be educated by independent-minded monks at Ettal Abbey
while his daughters attended the Convent of the Sacred Heart in Roehampton.7
Approaches from monarchist resistance groups were treated with extreme and
warranted caution: the Gestapo was watching and in August 1939 most of the
members of these were rounded up and thrown into gaol.
These arrests were a clear sign that conditions were becoming more
­dangerous in Germany, and on 13 August 1939 Rupprecht left Germany
to  visit his brother in Hungary. When he returned to Munich briefly in
October 1939, he faced police questioning about possible links to monarch-
ist sympathizers in the anti-Hitler resistance group known as the ‘Harnier
Circle’. Thereafter he spent the rest of the war in exile in Italy. Here, with
support from the Italian king and Pope Pius XII, he divided his time
between Florence and the beach at Forte dei Marmi, while his wife pre-
ferred the mountain air of the Tirol or Dolomites. In September 1943 Italy
surrendered to the allies and the Germans seized effective control of the
north of the country while the British and Americans advanced slowly from
the south.
With the Gestapo active in Florence and military pressure on Germany
growing, life was becoming dangerous for anyone with a record of anything
other than wholehearted support for the Nazis. In early June 1944 Rome
fell to the Americans and on 6 June the Allies landed in France. Rupprecht
thought it best to disappear. On 19 June he went to Florence railway station
and bought a ticket to travel and rejoin his wife in the Dolomites. Instead
of boarding the train, however, he disappeared into the crowd on the plat-
form, left the station by a side entrance, and doubled back to the apartment
of an Italian friend. Here he went into hiding.
The wisdom of doing so soon became apparent. On 20 July 1944 Hitler
narrowly escaped with his life when a bomb went off at his headquarters in
Prussia, the ‘Wolf ’s Lair’. The Nazi regime lashed out at anyone with even
the vaguest record of opposition to Hitler or the most tenuous link to the
conspirators responsible for the attempted assassination. Rupprecht was
­vulnerable. He knew the uncle of the would-be assassin, Claus Schenk von
Stauffenberg. Indeed, much of the conspiracy centred on the traditional
officers of the old general staff, and their families, some of them Bavarian,
whom Rupprecht had known for many years. The Gestapo searched
Rupprecht’s old apartment in Florence, but they had missed him and now
time was running out. As the boom of British guns crept ever closer, the
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t on t h e Ru n253

Gestapo scoured the city but eventually even they gave up and pulled out.
After a lawless few days of street fighting between Italian Fascists and par-
tisans, by 14 August it was finally safe for Rupprecht to emerge into daylight
once more. He had been indoors for nearly two months and was pleased to
be reunited with his son, Heinrich.
Rupprecht did not yet know what had happened to the rest of his family.
On Hitler’s personal order, the whole Wittelsbach family had been arrested
in the summer of 1944. They spent the winter that followed imprisoned in
a series of concentration camps, including Dachau. American soldiers finally
liberated them on 30 April 1945. By that stage Princess Antonie weighed less
than 40 kilograms. She refused ever to return to Germany. The family met
up but she never lived with Rupprecht again. Antonie died in Switzerland
on 31 July 1954, aged just fifty-four. At her request, she was buried not in
Bavaria, the country of her birth, but in Rome.
Rupprecht returned to Munich in November 1945. He had spent much
of the previous year writing memoranda for Eisenhower and the Allies with
suggestions for a federal constitution of the new Germany, but his time was
past. The Americans had hardly fought a world war to restore crowned
heads to the thrones of Europe. Supporters of the Wittelsbachs set up a
‘King’s Party’ to argue for a constitutional, parliamentary monarchy but the
party was banned from taking part in municipal elections in May 1946 and
soon collapsed. Any hopes Rupprecht might have harboured for a restor-
ation died away and his life became less and less political. He continued his
travels, especially to Italy, and resumed his ceremonial ‘duties’ around Bavaria.
His eightieth birthday in 1949 was celebrated with a week of festivities at
his country estate at Leutstetten and with march-pasts and receptions at
Schloss Nymphenburg.
Age was now catching up with Rupprecht. In October 1954 he con-
tracted pneumonia. July the next year saw him return from Italy with a bit
of a chill and soon his heart was weakening fast. On 2 August 1955, at five
minutes to three in the afternoon, he died in his bed at Schloss Leutstetten.
Rupprecht was eighty-six years old.The government ordered Bavarian flags
to be flown at half-mast. Over two days some 50,000 people filed past his
open coffin, lying in state in Schloss Nymphenburg, to see his body, dressed
as a Bavarian field marshal. On 7 August, some of his veterans, dressed in the
uniforms of a bygone age, led the procession which carried his body to its
final resting place in the Theatinerkirche St Kajetan, opposite the royal pal-
ace. He was buried, with ceremony and pomp fit for a king, next to his first
wife Marie-Gabriele and their son, Luitpold.
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

PART V
Conclusions
OUP CORRECTED PROOF – FINAL, 19/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

24
Rupprecht the Field Marshal

T his chapter continues the assessment of Rupprecht’s command which

we began in Chapter 10, picking out broader lessons about Germany,
her army, and the role of the Kaiser. Chapter 25 then builds on this by look-
ing at Rupprecht’s attempts to have his voice heard in wartime politics.
Finally, Chapter 26 sets Rupprecht in the context of broader history and
explores some of the lessons his career teaches us about the First World War
and modernity itself.
Rupprecht’s promotion to field marshal and command of an Army Group
in August 1916 changed his daily routine and relationship with his chief of
staff very little. His duties remained the same mix of administration, oper-
ational decision-making, and ceremonial as before. The only difference was
that it took longer to drive to visit units spread out over a broader front. His
partnership with Kuhl remained solid. Inevitably, relations cooled from
time to time. Between December 1916 and February 1917, for instance, an
exhausted Kuhl sulked after he was shut out from personnel decisions about
his staff. On the whole, though, they made a formidable team, complement-
ing each other well. On occasion, Rupprecht was the dominant partner. He
seems to have been first to see the counter-attack opportunity presented by
the Battle of Cambrai in November 1917, for instance. At other times, Kuhl
was the prime mover. It was Kuhl who went to Ludendorff on 6 April 1918
and demanded the suspension of Operation MICHAEL.
The cooperation between Rupprecht and Kuhl was eventually destabil-
ized, not by any disagreement between the two, or even by enemy action,
but by Ludendorff.We saw in Chapter 11 how Ludendorff and Hindenburg
took over at Oberste Heeresleitung (OHL) and began by delegating author-
ity to subordinates such as Rupprecht to fight their own battles without
interference. Quickly, however, OHL started to demand information on an
ever-widening list of questions, some of them minute details. Already by
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

258 Conclusions

8 October Kuhl was noting that the burden of reporting was obstructing
the  conduct of operations. Worse, Ludendorff ’s demands for immediate
responses to his many enquiries grew increasingly shrill: a hint perhaps that
his nerves were already becoming strained. When some people realize they
have taken on too much, they cut back to their core roles and delegate
­secondary tasks to others. Ludendorff ’s response was the opposite. As the
pressure mounted he tried to exert greater control, centralizing power
and authority in OHL and intervening even in low-level tactical decisions
and doctrine.1
From the Battle of Arras on, as noted in Chapter  15, instances of
Ludendorff increasingly being told what he wanted to hear multiplied, so
he was making decisions on the basis of incorrect reports and warped ana-
lysis. With information not always flowing freely around the institution, the
army’s intellectual honesty and ability to innovate were threatened. General
von Gebsattel summed it up well:
Very often there have indeed been false reports made by many units, on all
sorts of topics, even very important ones. I am firmly convinced that several
severe defeats we had to suffer during this war were caused by such false
reports. . . . From the very beginning of the war it was clear that army com-
mands and higher reacted very negatively to reports that relayed inconvenient
facts and did not suit the plans of the commanders in question.Whoever made
such reports was accused of being a merchant of doom, a pessimist, to be lack-
ing in energy or even worse. He would be heavily criticized and in some cases
even bluntly told that he could not be employed in such a position when
­acting like that.2

Ludendorff also began to intervene directly well down the chain of com-
mand. On 12 May 1917 he issued orders that henceforth OHL must approve
all divisional transfers, bypassing three steps in the hierarchy: corps, army,
and Army Group headquarters would all be sidelined. ‘Such a reduction of
my powers is definitely too much’, complained Rupprecht. ‘What’s the
point of having army groups, if every decision must be approved by OHL?
It is impossible to attend to every detail on the whole front from one place.’
Such an approach risked both poor decisions and the destruction of any
sense of initiative further down the chain of command.3
A good example of this came during the Flanders campaign, with the
to-and-fro over defensive tactics around the Battle of Broodseinde
(4 October 1917) discussed in Chapter 17. Ludendorff ’s intervention created
three problems. First, his suggestions were not always practical. For instance,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e F i e ld M a rsh a l259

the new revised defence in depth introduced in October relied on a stand-

ard of real-time infantry–artillery liaison which was impossible with 1917
communications. Second, it dented the self-confidence of his subordinates,
who became scared of being second-guessed by the chief. Third, he set a
poor example which some of his subordinates began to imitate. Loßberg,
for example, tried to dictate every detail of his army’s operations. The
­arrogance of Ludendorff ’s clique of ‘demi-god’ staff officers increasingly
­alienated them from the line commanders, further reducing the flow of
information.The result, Rupprecht thought, was over-schematic approaches
which tied the hands of subordinate leaders and made their moves predict-
able to the enemy. ‘There is no cure-all. A pattern is harmful’, Rupprecht
wrote on 9 October. ‘The situation must be dealt with sometimes one way,
sometimes another. It would be better if he [Ludendorff] did not bother
himself with intervening too much in the details of tactical methods.’4
As the stress mounted in 1918 and Ludendorff tried to control ever tinier
details on the basis of distorted reports, he grew increasingly out of touch
with everyday reality. He lost sight of the broader picture and his capacity
to make decisions deteriorated. We saw a clear example of this process at
work during spring 1918 in Chapter 20. On 9 June, Kuhl complained that:
Since the Army Group was set up we have hardly had a single quiet day. The
period of the big offensive, which should have been fine, was made dreadful
by Ludendorff ’s nervous excitement. The continual rush, blame and threats of
dismissal were appalling. One got stuck on the telephone so much that one
could hardly issue any orders or even think properly.

By August, Ludendorff was telephoning Kuhl at least five or six times a day:
There’s no dealing with Ludendorff . . . Ludendorff has endlessly something to
say about every detail, speaks directly to every army and their chiefs of staff,
and gives them detailed instructions often completely opposite to what he
ordered me. When we talk to the chiefs of staff, we find out that they’re doing
something completely different from what we told them to.That makes every-
thing much harder. He is extremely nervy and won’t take no for an answer.5

Ludendorff increasingly worked through junior members of the General

Staff, bypassing his Army Group and army commanders because he could
bully them if necessary. Accordingly, generals such as Rupprecht became
increasingly marginal and demoralized. The clearest sign of this lies in
Rupprecht’s attitude to home leave. As we saw in Chapter 10, he took none at
all in the first two years of the war. From November 1916, he began to do so,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

260 Conclusions

both because he realized his health required it but also because he found out
that counterparts, such as the Prussian crown prince, were enjoying lots. At
first, he was extremely diligent about returning to his post if bad news
arrived. Reports from Flanders in September 1917, for instance, made him
cut short a trip home. By August 1918, however, when the situation was
arguably more serious, with his Army Group under British pressure and
Bapaume just lost, Rupprecht made sure to finish his leave before he
returned to duty. He only got back a fortnight after fighting had resumed.
Here was a man who no longer felt indispensable.6
Rupprecht’s career demonstrates that the German army’s approach to
command was not, as the myth would have it, universally pragmatic and
decentralized, with authority consistently delegated to the man in the best
position to give the orders required. Instead, decision-making power shifted
up and down the command hierarchy in a highly contingent manner, driven
primarily by the pressure of events and force of personalities. Other factors,
such as internal politics within the clique-ridden German officer corps,
were also at work.7 We saw how Rupprecht had trouble getting rid of
Lochow in May–June 1915, largely because he was a client of Falkenhayn’s
whom Lambsdorff was not prepared to alienate. Men such as Dohna-
Schlobitten were selected for important posts on the strength of imperial
connections rather than experience or qualifications, with sometimes disas-
trous results. Not all the Kaiser’s appointments were useless, though: indeed,
some proved extremely successful. The selection of Rupprecht’s chiefs of
staff, Krafft and Kuhl, for example, depended on politics as much as merit
but in the event succeeded well. Even Lambsdorff ’s selection was justified
by his record, although in fact he did not work out.
The case of Major Max Stapff offers another example of this. Stapff began
1917 as chief of staff to Fourth Army. As we saw in Chapter 16, after the
Battle of Messines Stapff was transferred to the less threatened Sixth Army
around Arras. Here, he soon began to worry that his sector might be the
next British target and wanted reinforcements. To get them, he bypassed
Rupprecht and the chain of command, going straight to his friend Wetzell,
the chief of operations at OHL. Stapff did not get the men he wanted but
he did convince Wetzell that Artois posed a major threat. This proved a dis-
traction from the real danger in Flanders. Even when Stapff had lost the
confidence of his corps commanders, Wetzell’s influence kept him in post.
Wetzell resented the role Kuhl had played in sidelining Stapff back in June,
which contributed to friction between the OHL operations department
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e F i e ld M a rsh a l261

and Rupprecht’s Army Group. Wetzell was ‘a dangerous enemy, since he

can’t be controlled and can do much damage behind my back’, Kuhl
thought. In late October 1917, Wetzell sent reinforcements Rupprecht badly
needed to the Aisne instead, where they achieved nothing. The Stapff–
Wetzell friendship undermined Kuhl’s ability to extract the resources he
needed from OHL.8
In the German army, therefore, rational modern efficiency coexisted
with patronage structures which Frederick the Great would have r­ecognized.
It was simultaneously capable of disinterested meritocracy and p­ ersonal
intrigue. Human tensions within the officer corps introduced an element of
uncertainty and instability to its processes. There were at least two other
factors getting in the way of the German army fighting its war, however:
distrust and antagonism between the different regional contingents, and the
Kaiser himself.
The German army was the midwife of national unity, its guardian and
guarantor. The Reich, founded on Prussian battlefield victories between
1864 and 1871, brought together communities long divided by religion,
money, culture, diet, dialect, and geography. The extent to which these and
other fault-lines divided—and still divide—German society never fails to
surprise non-Germans. The army was far from immune. Much of the time,
the fault-lines ran beneath the surface, but under wartime pressure several
broke into the open.The relationship between Bavaria and Prussia is a good
example. Between 1914 and 1918 the two largest German states were
­frequently wrangling. There were two underlying sources of friction. First,
Bavaria was sensitive to any infringement of her rights under the 1871 consti-
tution. Second, a deeper thread of low-level prejudice also infected relations.
There was, for instance, a long tradition in Prussia that Bavarians made
poor soldiers. They had been ineffectual opponents in 1866 and performed
scarcely better when they were allies in 1870. This was in turn a symptom
of a deeper cultural divide between the educated, Protestant north of large-
scale farms and industry and the bucolic peasantry of the Catholic south.
As the second largest block in the German federation, Bavaria sometimes
played a special role as representative of the non-Prussian states. She held the
permanent chair of the foreign affairs committee of the federal upper house,
the Bundesrat, for instance. She also had other, less formal, paths to influence.
Her ambassador in Berlin, Lerchenfeld, had been in post since 1880 and was
both highly experienced and well respected at the Kaiser’s court, for example.
Nevertheless, as Stefan März has argued, ‘the tension between unitary and
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

262 Conclusions

federal characteristics constituted an essential element of the Kaiserreich’

and the Prusso-Bavarian relationship was rarely straightforward.9
Relations between the two royal families were cordial without being
close. Religious differences prevented intermarriage and Ludwig III had
been wounded fighting against the Prussians in 1866. He limped for the rest
of his life as a result. His relations with Wilhelm II were not always smooth.
Both men were volatile and prickly. At the coronation of Tsar Nicholas II
in 1896, for instance, Ludwig had complained loudly at being treated like
the Kaiser’s vassal, rather than his ally. Wilhelm II summoned Ludwig to his
yacht and dressed him down ‘like a schoolboy’. Nonetheless, Ludwig
retained the right of access to the Kaiser and the senior members of both
families crossed paths frequently, not least at official events such as the
funeral of the Habsburg Emperor Franz Josef in 1916.Wilhelm attended the
golden wedding celebrations for the Bavarian king and queen in February
1918. Rupprecht had known Wilhelm II, who was ten years older than him,
since at least his student days in Berlin in 1890.10
Rupprecht supported a united but federal Germany led by Prussia, seeing
that as the best defence against the clericalism and Vatican interference a
closer union with Austria might entail. For him, though, Germany had to
be more than just a Greater Prussia. The constitutional rights and preroga-
tives of the other states must be safeguarded. Even before the war began, he
had been nervous of Prussia’s centralizing ambitions. In January 1914 his
suspicions were reinforced when Moltke told him that Rupprecht was to
have a Prussian chief of staff in the event of war. The clear implication was
that only thus could Moltke be sure to avoid mistakes which might other-
wise endanger Prussia. The outcome of war, Moltke tactlessly reminded
Rupprecht, would be crucial for Prussia. ‘No less so for Bavaria’, Rupprecht
drily replied. Krafft, the Bavarian who eventually got the job, possessed a
level of anti-Prussian paranoia which reinforced Rupprecht’s suspicions.11
Once war broke out the pious hope was that ‘as in 1870 will Bavarians,
Prussians and Badeners fight together and win as German brothers’.
National sensitivities soon surfaced, however. Perceived threats to Bavarian
reserve rights under the constitution generated resentment. Rupprecht was
especially sensitive to encroachments on the military convention, such as
any suggestion that the Kaiser might appoint or promote Bavarian officers
or install Prussian officers in command of Bavarian units. Likewise, he
opposed the extension of censorship from Berlin to include the tradition-
ally lightly regulated Munich newspapers.12
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e F i e ld M a rsh a l263

At times, Prusso-Bavarian tensions had a direct and evident impact on

operations. An early example occurred in late August 1914 when OHL
decided to send two army corps over to East Prussia to oppose a Russian
invasion. When a railway expert suggested that I Bavarian Corps was one
of the easiest formations to send, Moltke’s deputy exclaimed: ‘that’s
impossible. We Prussians can’t let it be said that East Prussia was liberated
by Bavarians.’ Prussian units were transferred instead. Second, as the war
continued, it proved neither militarily efficient nor politically desirable
to maintain Sixth Army as a predominantly Bavarian formation. After the
Lorraine campaign, contrary to Rupprecht’s wishes, Bavarians were dis-
persed up and down the front and formed only a minority of Rupprecht’s
command.Third, in 1916 Ludendorff originally wanted to give Rupprecht
overall command of the whole Western Front. To promote the Bavarian
crown prince over his Prussian counterpart, however, might prove diplo-
matically tricky and Ludendorff had to drop the idea. A fourth example
took place in October 1916: at a parade in front of the Kaiser, General
von Boehn made a speech boasting that no enemy would break through
his corps while a single Prussian heart was still beating. Unfortunately,
two of his three divisions, and many of the men on parade, were in fact
Bavarian. After complaints reached King Ludwig III, Boehn was trans-
ferred away at the first opportunity. Lastly, in March 1917 Rupprecht
wanted to resign in protest when ordered to scorch the earth as he fell
back to the Hindenburg Line. He did not do so for fear that this would
be interpreted at home and abroad as betraying a split between Bavaria
and Germany.13
Where Prussian arrogance met Bavarian inferiority complex, further
friction arose. The Bavarian general Fasbender stormed out of a meeting to
plan an attack at Arras in October 1914 after being provoked with ‘typical
Prussian condescension’ by Claer. Casual contempt was widespread: during
the Somme fighting, when told that the unit which had lost a given pos-
ition was Bavarian, Loßberg had muttered ‘of course’. During a tricky
period for his relations with Kuhl, Rupprecht wrote:

I regret ever more that I haven’t had a Bavarian chief of staff since the begin-
ning of the campaign. It would have made things much smoother. The
Prussian chiefs of staff tend to be overbearing and behave as if they are only
subordinate and responsible to OHL. In the future we must try to ensure that
Bavarian troops fight together under their own staff and are not treated as
mere contingents of the Imperial army.
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

264 Conclusions

Low-level discourtesy did little to promote smooth working and at times

could cause discomfort. General Wenninger at imperial headquarters noted
on 6 September 1914 that ‘my first week in our new headquarters as the only
Bavarian was as miserable as that in Koblenz was enjoyable.The t­ hermometer
rises and falls with the performance of our troops on the battlefield.’14
Wartime not only exacerbated existing tensions: it also created new ones.
Any suspicion that Bavarians were being exploited and making a dispropor-
tionate sacrifice provoked fury both within the army and at home. Nowhere
were emotions rawer than when it came to food. In October 1916, for
example, the Munich meat ration was cut to 187 grams per head, while it
remained 250 grams in Berlin. Outrage was the predictable result, made
worse by the fact that Bavaria supplied much of the food to Berlin.15
Not only should Bavaria pay a fair share of the costs of the war and no
more; she should also receive her deserts when it came to sharing out the
spoils of winning. From Bethmann Hollweg’s ‘September Programme’ of
1914 onwards, the German leadership debated the annexations which would
follow victory and how to split them up between the different federal states.
Rupprecht was as keen as the government in Munich that Bavaria should
gain her rightful share of anything on offer and his contacts kept him well
informed about the various schemes which came up.Tracking the evolution
of all the competing proposals would fill much space to little purpose, since
none of them became a reality. One point is crucial, however: annexations
represented an impossible problem for Germany because there was no way
to reconcile everyone’s interests. The tone was set as early as November
1914 when the question of the post-war future of Alsace and Lorraine came
up. The 1871 solution of making them directly administered imperial prov-
inces had created an awkward administrative anomaly and failed to make
good Germans of the inhabitants as hoped. Instead, it seemed that Alsace
and Lorraine should be transferred to one or more of the German states.
How to do this was not straightforward, however. Bavaria was not prepared
to let Prussia take both: this would make her too powerful and undermine
federalism. Some conservatives in Prussia did not even want Alsace-Lorraine,
since they worried that the Catholic population there would disturb the
confessional equilibrium in their largely Protestant kingdom. The possibil-
ity of Bavaria taking either or both, however, provoked opposition from
Württemberg and Saxony, who worried about the spread of Catholic influ-
ence. Baden, meanwhile, wanted Upper Alsace and was not prepared to let
herself be encircled by Bavarian territory. No consensus was possible.16
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e F i e ld M a rsh a l265

After Russia’s collapse, Hindenburg and Ludendorff designed a new

Prussian order for Europe which ignored, or was ignorant of, complexities
of this kind. In December 1917 Ludendorff wanted Prussia to annex
Lithuania, Courland and Riga in the East, and both Alsace and Lorraine in
the west. The furthest he was prepared to go to buy off the lesser German
states was to offer the rulers of Bavaria and Saxony the thrones of Belgium
and Lithuania respectively, via a personal union. In other words, a Wittelsbach
would rule Belgium, although it would not become part of Bavaria.
Rupprecht was scornful:
The Prussian motto ‘Suum cuique’ (‘each to his own’) should be translated
into German as ‘everything for me’. Since the start of the war, from my know-
ledge of the Prussian character and the weakness of our leading men in
Bavaria, I foresaw that, whatever her commitment to the war, Bavaria would
suffer only harm and lose respect.
The idea of a personal union was an echo of a bygone age which could not
work in modern conditions.17
What this brief discussion of the annexations problem has demonstrated
is that Germany did not know what world she wanted to create even if she
somehow managed to win. She had gone into the war with no clear polit-
ical vision and proved unable to develop one. Even in the best case, there
was no territorial settlement which could possibly satisfy all those who were
bleeding to achieve it. Just as annexing Serbia would destabilize the fragile
balance between Austria and Hungary, any move to incorporate Alsace-
Lorraine threatened the 1871 Reich settlement. Adding Belgium, Poland,
and the Baltic states into the mix only made affairs more intractable.
Probably only Ludendorff ’s brutal approach of ignoring the sensitivities of
the lesser states and giving Prussia whatever she wanted was viable in the
long run. He certainly did not care about the difficulties involved, even if he
was aware of them. Whether he, or anyone else, could have successfully
managed the necessary constitutional change from a federal to a unitary
Germany must remain an open question.
The fact that Germany went to war with no clear vision of the settle-
ment she might fashion if she won, and knowing that any division of the
spoils of victory would prove difficult at best, reinforces the view that
Germany lacked a coherent strategy. If anyone was supposed to articulate
one and bring statesmen and generals together to execute it, that person was
the Kaiser.Wilhelm II is often written off as a ‘shadow Kaiser’ who did little
to determine the course of the First World War, whatever his role in starting
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

266 Conclusions

it. His biographer, John Röhl, argues that his power evaporated ‘almost over-
night’ as soon as the first shots were fired. He delegated the conduct of the
war to his Chief of the General Staff and, according to Holger Afflerbach,
only ever intervened directly in operational decisions rarely. The Kaiser
himself complained that ‘the General Staff tells me nothing and never asks
my advice. If people in Germany think that I am the Supreme Commander
they are grossly mistaken. I drink tea, saw wood and go for walks.’ Most
modern historians detect a further erosion of his power in August 1916
when Hindenburg and Ludendorff took over OHL. By offering the stra-
tegic leadership which Wilhelm had not, these two pushed the Kaiser even
further to the margins.They forced out the Chancellor, Bethmann Hollweg,
as well as other members of Wilhelm’s entourage, and led Germany back
into unrestricted submarine warfare. At the end, they almost sent the Kaiser
off to meet a ‘heroic’ death in battle.18
The Kaiser was more important than this picture suggests, in three main
ways. First, he retained significant power of his own, especially over appoint-
ments, and he was prepared to use it. Two examples show this clearly. In
April 1918, after a routine visit to II Bavarian Corps, he insisted that its com-
mander, General Otto von Stetten, be sacked. Whether he was correct that
Stetten was exhausted or not, this constituted an insult to Bavaria, an
infringement of her constitutional rights, and a disturbing intervention in
day-to-day operations by a man whose qualifications to make such deci-
sions were slight. Another case came up in Chapter 16, when Wilhelm II
petulantly forbade an exchange of staffs between Fourth and Sixth Armies,
designed to strengthen the command team likely to be hit by the next
British Expeditionary Force (BEF) offensive in Flanders, because he wanted
revenge on Rupprecht for sacking two corps commanders six months ­earlier
who happened to be imperial favourites.19
Unpredictable meddling of this kind was a clear source of uncertainty.
It was compounded by the second form of impact Wilhelm II had on his
army. A leader sets the emotional tone for those under his command. Voltaire
praised the Duke of Marlborough, for instance, for possessing ‘that calm
courage in the midst of tumult, that serenity of soul in danger, which is the
greatest gift of nature for command’. A calm and resolute commander instils
his staff and subordinates with confidence.The notorious emotional instabil-
ity of Wilhelm II had exactly the opposite effect. Rupprecht had little respect
for him, seeing him as a military dilettante who could not be bothered to
listen to any briefing longer than twenty minutes. ‘The Kaiser is an unstable
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Ru ppr ech t t h e F i e ld M a rsh a l267

character’, he noted; ‘on the one hand weak and soft, on the other at
times . . . brutal’. On 25 May 1917 the Kaiser told a parade of his troops at
Tournai to take no British prisoners. Rupprecht tried to hush it up, but
within a couple of weeks it was in the English newspapers all the same.
Rupprecht was exasperated:

If only he would learn to keep control of his words. There’s no hope in this
regard. He often does not know himself what he is saying. He means no harm
by what he says, however it sounds, he just loves striking a pose and this has
got him in trouble several times before.

Childish behaviour set a poor example and left his subordinates wondering
both where they stood with him and why they were fighting for such a man.20
The third way the Kaiser influenced command was less obviously under
his own control, although it was important all the same. In much the same
way as Thomas Ertman has argued that early modern states developed by
imitating the models they had available to them, the power networks of the
army had evolved by aping those of the imperial court. The Chief of the
General Staff served entirely at His Majesty’s whim (an especially appropri-
ate word in the case of the Kaiser). Falkenhayn, throughout his tenure, was
aware that he had not been a popular choice for the post and that influential
voices were calling for his head. He was entirely beholden to Wilhelm II. At
first, Ludendorff had rather more security by virtue of Hindenburg’s popu-
larity in the country at large. Nevertheless, he knew that even Hindenburg
might not be able to save him if the capricious Kaiser wanted him gone, as
indeed proved the case in October 1918. The command system was thus
built on shaky foundations.Worse yet, at lower levels this monarch–courtier
relationship was replicated, with subordinates so dependent on their super-
iors’ goodwill that their independence was undermined. Men such as
Tappen and Wetzell were no less dependent on the goodwill of Falkenhayn
or Ludendorff than the Chiefs of the General Staff were on that of the
Kaiser.We have already noted how influential cliques and personal relation-
ships could be. An exception will prove this rule: when Ludendorff was
temporarily too weak to object in September 1918, his Army Group com-
manders acted quickly to demand the break-up of his immediate cabal at
OHL and infuse fresh blood.21
This instability within the army command meant that every time
Falkenhayn or Ludendorff felt their position with the Kaiser threatened,
their knee-jerk response was to suck power and decision-making back
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

268 Conclusions

into OHL. Auftragstaktik had its place as a tactical measure at division level

and below, but among the senior command ranks delegation was the excep-
tion rather than the rule. It flourished only briefly during Ludendorff ’s
honeymoon period in autumn 1916. The default setting to which the army
reverted when stress mounted involved centralization and increasing rigid-
ity.This had two consequences. It made the job of operational commanders
such as Rupprecht very tricky. They had to negotiate a shifting structure
where responsibility and power flowed up, down, and around from day to
day or minute to minute in unpredictable ways. In these circumstances, it
was harder to make decisions and control compliance with orders than it
needed to be. Further, enemy pressure and a politicized command hierarchy
combined to bring poor reporting, an OHL ever more out of touch with
reality, and worse decisions. This circle rapidly became vicious.
The German army between 1914 and 1918 was far from the well-oiled
machine of myth. Instead, it was a deeply flawed institution which reached
poor battlefield decisions and by doing so contributed significantly to
Germany’s defeat. The First World War was not just lost at the tactical and
operational levels of war, however. Poor strategic and political decision-
making also played a part, and it is to these that we now turn.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

25
Rupprecht and Politics

I n addition to his role as a military commander, Rupprecht saw himself,

and was perceived, as the senior Bavarian on the Western Front. He thus
represented Bavaria and, to some extent, German unity. This gave him
political value above and beyond his military responsibilities. It also afforded
him a measure of independence and protection from the worst excesses
of army politics and offered him the opportunity to challenge prevailing
orthodoxies on the strategic and political issues Germany faced as she
fought. Rupprecht enjoyed another significant advantage, too. By dint of his
position and status, he had access to a network of multiple sources of useful
intelligence. Some of these were formal. The Bavarian government kept a
representative at his headquarters: Leopold Krafft von Dellmensingen,
a cousin of his first chief of staff. Together with the Bavarian military rep-
resentatives at the Kaiser’s General Headquarters, Leopold Krafft kept him
in touch with gossip from Munich and the imperial court. Less formally,
Rupprecht nurtured a wide range of contacts in the higher reaches of pol-
itics, business, and the military. Friends and like minds shared information
and worked together from time to time. At no point did this form a definite
group with a coherent agenda, however. Instead, informal coalitions came
together opportunistically to try to influence policy on particular issues.
The attempted coup against Falkenhayn is one obvious example, which we
will come back to below. This chapter examines how far Rupprecht was
able to exploit his independence to affect policy.1
Between 1914 and 1918 Germany faced a series of life-or-death decisions.
The original choice to go to war was the first of these. As we saw in
Chapter  1, neither Rupprecht nor any other Bavarian had any influence
over that decision, which was taken in Berlin by the Kaiser and a small
group of his closest advisers. There is no evidence that Rupprecht or his
father opposed it, however, and one suspects they backed German expansion
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

270 Conclusions

so long as Bavaria got her fair share. Once war had broken out, the questions
became how violently the war should be fought and how it should be ended.
The two were linked: the more unlimited the objectives, the more it made
sense to escalate the violence. The resorts to poison gas and unrestricted sub-
marine warfare were two controversial examples of the radicalization of war.
As we saw in Chapters 7 and 14, at the time Rupprecht agreed with the use of
both, whatever he sometimes said later. Rupprecht was broadly in line with the
German consensus on how violently the war should be fought.Where he was
out of step at times was when it came to how to end the war.
With the benefit of hindsight, it is clear that the French victory on the
River Marne in early September 1914 marked the turning point of the war.
Once the Schlieffen Plan had collapsed, Germany’s hopes of a rapid victory
evaporated. She became locked into exactly the kind of drawn-out war
on multiple fronts she had desperately wanted to avoid. Should Germany
accept a compromise peace (‘Verständigungsfrieden’)? Or fight on, win on
the battlefield, and then impose the settlement she wanted (‘Siegfrieden’)?
The idea that this was a choice open to Germany, however, was a mirage.
She no longer had options. Once she had failed to snatch victory in a short
war, she had no chance of winning a long one. Neither the German army
nor the civil government was ready to see things this way, however. When
Falkenhayn advised Bethmann Hollweg that military methods had failed
and Germany must pursue a diplomatic solution in November 1914, he was
denounced as defeatist. Rupprecht was one of those who did so, playing a
leading role in the cabal which sought to force Falkenhayn from office. He
then welcomed Falkenhayn’s replacement by Hindenburg and Ludendorff.
The two new men were firm believers in final military victory and intent
on harnessing all Germany’s strengths to achieve it. In retrospect, the switch
of leadership clearly prolonged the war.
Rupprecht’s faith in the Hindenburg/Ludendorff programme began to
waver as his belief in the possibility of a war-winning offensive fell away.
Where he had still been arguing for a single knock-out blow to decide the
war in early 1916, by the time he had lived through Verdun and the Somme,
his view had shifted. He came to believe that a major breakthrough was
highly unlikely and that, even if there was one, no single battle was likely to
yield victory. Rupprecht did not yet believe defeat was probable, much less
inevitable, but from the end of 1916 onwards he recognized it as a possibility,
even if he veered between optimism and pessimism at times. He was not
alone in this view: indeed, of the chiefs of staff of the armies under his
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t a n d Poli t ics271

command, only one disagreed in January 1917. The next month, even
Ludendorff and his staff seemed to accept that shortages of manpower,
horses, transport, and ammunition left the army incapable of winning the
war. They were relying on the U-boats instead. Neither Rupprecht nor any
of the others followed this insight through to its logical conclusion, however,
even when it became clear that submarine warfare was failing to deliver on the
extravagant promises made for it. When, in July 1917, the Reichstag passed a
resolution in favour of peace without annexations. Rupprecht’s first instinct
was to welcome it. This was not, however, because he was desperate for
peace at any price.The ‘peace resolution’, which has often been represented
as a pacifist rejection of war inspired by Bolshevik declarations in Russia,
was nothing of the sort. Rather, it envisaged Germany retaining the terri-
torial gains she had already made and marked an attempt to stiffen popular
resolve. Although Rupprecht was aware of underlying problems with man-
power and food supply and hoped for an end to the war in the autumn, the
immediate situation seemed manageable.
Even as late as January 1918, emboldened by Russia’s collapse, Rupprecht
was still backing Ludendorff and thinking one major effort might yet win
the war. His faith only began to drain away as his conviction grew that
Ludendorff was butchering the execution of the spring offensives. On
19 February he warned the Kaiser that they represented Germany’s last card,
but Wilhelm II parroted Ludendorff ’s blithe confidence. He also expressed
his concerns to members of the Kaiser’s entourage such as Plessen and
Lyncker. The next day he tried to explain the seriousness of the situation to
his father, but was unable to deflate Ludwig’s ‘amazing optimism’. By this
time Rupprecht had finally come round to the idea that the only way out
of the war was to negotiate.2
As we saw in Chapter 21, it was only in May 1918, after the disappoint-
ments of MICHAEL and GEORGETTE, that Rupprecht finally accepted
that Germany’s situation was becoming hopeless. This set him at odds with
Ludendorff who, despite wobbles in August and late September, remained
capable of convincing himself that the western powers would collapse
internally, as Russia had done, almost until the end. Rupprecht wrote to the
Chancellor, Hertling, suggesting peace talks while Max von Baden talked to
the Kaiser along the same lines. Hertling was old, tired, and little more than
Oberste Heeresleitung’s (OHL’s) stooge by this time. He felt none of the
urgency Rupprecht did and his reply was non-committal. An attempt to
persuade Ludendorff face-to-face was no more successful.3
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

272 Conclusions

Rupprecht’s efforts to convince the leadership that a negotiated peace

was necessary took three forms. First, he continued to feed information to
Max von Baden on the state of the army to use against those who even now
wanted to prolong the war. Second, he tried to awaken his father to the
dangers Bavaria and Germany faced. On 25 July he warned that the enemy
had the advantage and every effort must be made for peace, even if it cost
Germany. On 4 September, ‘peace—as soon as possible and even at a great
price—has become urgent’, he wrote. Later that month he criticized OHL
for refusing to accept that Germany was defenceless and must make peace
at once, whatever the price. On 7 October he pointed out that military
defeat might have broader political ramifications and suggested democra-
tization: ‘the failure of previous arrangements means the idea that the people
should have a decisive voice in their fate has some legitimacy’. Rupprecht
was beginning to despair: ‘poor Bavaria, that you must pay so for Prussia’s
mistakes! Catastrophe can strike any day!’ On 26 October, a new and even
more ominous note entered Rupprecht’s letters to Ludwig. Some Prussian
officers had begun muttering that the removal of Ludendorff and
Hindenburg would not go far enough. The Kaiser himself, and his dynasty,
might also be ditched. ‘That’s how far we have come, already!’ Worse,
Rupprecht warned, there was a danger that other states might then follow
the Prussian example. Finally, on 1 November, in the wake of press reports
that the Hohenzollerns had become obstacles to peace, he begged his
father to press the Kaiser to abdicate to enable an armistice. ‘The situation’,
Rupprecht wrote, ‘is extremely serious. . . . The [revolutionary and possibly
separatist] movements which are getting ready in northern Germany and
are already a fact in Austria could also transfer to Bavaria.’4
The King paid little attention to his son’s warnings. Ludwig consistently
under-estimated how far support for his regime was being weakened by
food shortages, war-weariness, and his government’s apparent failure to
defend Bavarian interests against Prussia. As late as July 1918 he remained
convinced that any peace treaty would result in German territorial expan-
sion. Ludwig seems to have resented and feared his son’s relative popularity.
Remarkably, he felt that late October and early November 1918 was the
perfect time to re-open an old argument about Rupprecht’s home in
Munich. In November 1916, after four years of bitter family dispute, it had
been decided that Rupprecht should use the suite of rooms known as the
Steinzimmer (‘Stone Rooms’) in the old palace known as the Residenz.
Ludwig now overturned that agreement, deciding instead that Rupprecht
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t a n d Poli t ics273

should move into another building on the other side of the street. This
triggered a new, unnecessary, and trivial row. Ludwig told Prime Minister
Dandl to form a parliamentary government on 2 November, demonstrating
how weak his grasp of the gravity of the situation was. It was too little and
too late to satisfy the growing radicalism of his opponents. His last message
to his son, sent on 7 November, while blithely asserting that he understood
the seriousness of the situation, showed clearly that he did not: he hoped
that ‘after the Armistice it will be possible for you to come here soon’.5
The failure of indirect approaches drove Rupprecht to take more direct
measures. He took a step he would never have dreamed of earlier in the war.
On 21 September he wrote directly to Wilhelm II with a summary of the
military situation and suggesting that Hertling had lost any value he had
once had; he should be replaced as Chancellor. There is no evidence that
Rupprecht ever received a reply. An appeal to the Kaiser in November, to be
allowed to take over the defence of the Tirol against the Italians, proved
equally unsuccessful.6
For all the advantages Rupprecht enjoyed, his ability to use them faced
significant obstacles. First, there were considerable limits to his ability to assert
his independence. His ultimate weapon was resignation, but the political
consequences of threatening this were so great that he hesitated to use it.
Any sign of disunity might encourage the enemy and weaken Bavarian
influence further. Thus, as we saw above, he held back from resigning in
protest at the ‘scorched earth’ order of March 1917. In private, he could
be  petulant: he seems to have mentioned resignation to members of his
entourage at least six times during the war. The only occasion on which he
formally threatened to resign, however, was carefully not on a general point
of policy or principle. That might have laid him open to charges of disloyalty
or irresponsibility. Instead the issue Rupprecht picked was Falkenhayn’s
undue interference in his command of Sixth Army. He never tried to
use resignation to unseat his opponents, as Hindenburg and Ludendorff
repeatedly did.7
Second, Rupprecht had no independent power. He was only heir to the
throne, not sovereign. Therefore much depended on his ability to sell his
ideas to others and work through them. Since one of his main routes to
influence lay via his father, it was unfortunate that personal relations with
Ludwig III were poor and widely known to be so. For instance, in January
1916, Rupprecht tried to persuade the King that Bavaria needed to stand
firm against Prussian infringements of her rights, only to find that all his
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

274 Conclusions

father wanted to discuss was uniforms. Rupprecht’s ability to act as a

channel to, or spokesman for, the Bavarian government was consequently
limited. Ludwig’s conservatism and poor political judgement only made
matters worse and fed into a strange passivity on the part of the Bavarian
government. Munich often lacked the confidence to demand a voice in
imperial policy. In July 1917, for instance, Ludwig ignored warnings from
his son that Prussian democratization might threaten the powers of the
princely states and that ‘the whole of Germany is watching and waiting
for Bavaria to intervene! Much depends on this game, maybe everything!’
Indeed, precisely because Bethmann Hollweg supported reforms which
threatened the prerogatives of the princes, Ludwig made no effort to
support him in his struggle with Ludendorff, despite Rupprecht’s advice
that Bethmann Hollweg offered a quicker end to the war and a better
deal for Bavaria than any alternative Chancellor was likely to. Ludwig
could not see, as Rupprecht could, that this inertia was undermining sup-
port for the monarchy. Bavaria’s lack of a role in policy was not all the
fault of the government in Munich, though. The Kaiser and those around
him often gave scant consideration to Bavarian interests and concerns.
When, for example, Hindenburg decided to set up a Supreme War Office
(‘Kriegsamt’), under the Prussian War Ministry, to coordinate industrial
mobilization across the whole of Germany, he gave Munich only twenty-
four hours’ notice. Rupprecht complained of this ‘typical Prussian lack
of consideration. But it is a scandal, how badly we are bossed about by
our allies in Berlin.’8
Rupprecht has sometimes been seen as sidelined politically because he
was unfashionably downbeat about Germany’s chances. As we have seen,
however, Rupprecht was not a consistently negative voice within her coun-
cils of war. Indeed, pessimism finally triumphed over optimism and he gave
up hope in victory only when the results of the spring 1918 offensives
proved disappointing. Rupprecht was, however, a frequent and trenchant
critic of Germany’s conduct of the war. This was largely a function of frus-
tration: one suspects he thought he could be making a better job of it. We
have also seen how prickly he could be if he felt his rights or dignity, either
personally or as a Bavarian, were threatened. This frustrated his superiors
and diminished his influence. So did his tendency, shared with many born
critics, to be better at identifying weak points in OHL’s arguments than at
finding constructive answers. When, for instance, he correctly identified
that Falkenhayn’s attritional approach could not deliver military victory,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Ru ppr ech t a n d Poli t ics275

he could offer no realistic alternative. Later, he pressed Ludendorff for a

compromise peace but could suggest no plan to deliver one. He was always
coming up with problems, never solutions.9
Rupprecht’s fallible judgement was another problem. To support Ludendorff
over Falkenhayn was to choose a longer war. Similarly, as Chancellor,
Bethmann Hollweg probably had a better chance of negotiating an end to
the conflict than Rupprecht’s candidate, Georg Hertling, who was seventy-
three years old and never likely to be able to stand up to OHL. When
Rupprecht observed the calls in Berlin for democratic reforms in July 1917,
he spotted the threat this posed to Bavarian independence, and indeed the
Hohenzollern dynasty, but failed to anticipate that pressure for constitu-
tional change might spread to Munich, too.10
One of Rupprecht’s most important judgements concerned ending the
war. As we have seen, he realized that Germany could not hope for military
victory, and needed to negotiate for peace, well before conservatives such as
Ludendorff, Ludwig III, and the Kaiser. Nonetheless, he did not give up
hope until April or May 1918. The interesting question is less why he lost
faith in victory than why he carried on believing that Germany could win
for as long as he did. Why did not Rupprecht, and the other warriors of the
German Reich, see that any chance of victory had evaporated by the end of
the Battle of the Marne?
Part of the explanation, no doubt, lies in a simple misjudgement about
the balance of forces and willpower. The most common error in warfare
is  to under-estimate the enemy’s capacity to resist. Faulty or inadequate
intelligence can cause this. So also can the plain arrogance of assuming one’s
own troops are better than the enemy’s—something we have seen the
Germans were prone to—and wishful thinking. Even when one has man-
aged to collate the right information, it remains easy to analyse it incorrectly,
especially within an organization in which intellectual honesty sometimes
took second place to politics.
A second explanation of the hard-liners’ resistance to peace negotiations
probably lay in their assumption that the Allies would be likely to drive
a  hard bargain. President Woodrow Wilson’s Fourteen Points included
demands that Germany evacuate Belgium and France, including the prov-
inces of Alsace and Lorraine. These were hard for the Germans to swallow
but were more moderate proposals than Germany was likely to be offered
by some French statesmen, for example.To the German conservatives, there
seemed no basis for compromise.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

276 Conclusions

Linked to this was a third consideration. If Germany accepted the

Fourteen Points as a basis for peace talks, she would effectively be admitting
defeat. A war which had supposedly been fought to secure Germany from
war for all time would have achieved less than nothing, at huge financial and
human cost.The idea that she might end up with less territory than she had
before 1871 was unthinkable. To accept defeat would be to value all the
sacrifices of the last four years at nil and destroy the prestige of the army.
Militaries are ‘can-do’ institutions. They pride themselves on their ability to
find answers to complex problems, even when resources are limited, and in
1914 the Schlieffen Plan seemed to offer a military solution to Germany’s
political problems. To admit failure would shatter the army’s self-image
and  so, since the military was one of the keystone imperial institutions,
undermine also the legitimacy of the Reich itself. Already by 1918 demands
for constitutional reform and a quick end to the war were growing. Once
military and home-front morale began to collapse in the summer, the threat
to the regime became increasingly obvious. The dissatisfaction of a war-
weary public with a government which seemed unable to end the fighting
fed back into frustration at those who had led Germany into a disastrous
war and then lost it. Ludendorff saw this danger clearly and indeed shifted
his priority to preserving the army intact as a bulwark against the socialist
revolution he increasingly feared.
The German military mind thus had multiple reasons for the obstinacy
with which it held out for final victory. Rupprecht possessed the tempera-
ment, and some of the independence of mind and position, to be intensely
critical about aspects of how the German government and army were con-
ducting the war. On the crucial question of ending it, however, it took him
a long while to escape the orthodoxy of his heredity and upbringing. The
fact that Rupprecht was able to do so proves that the militarist spell could
be broken. That it took him so long, however, shows how deeply ingrained
it had become as well as how seductive its ideas and glittering prizes were.
In the self-destructive mentality of the imperial military lies Rupprecht’s
tragedy—and Germany’s.11
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

26
Last Words

W hen I began writing this book, my first objective was to tell the
sad  story of how Rupprecht, proud Crown Prince of Bavaria and
German army general, became plain Mr Landsberg, a fugitive surviving on
the protection of strangers. In the course of doing so, however, Rupprecht
has also offered us a handy lens through which to study broader themes
about the First World War and to offer a new, German, perspective on a
history which has too often been written exclusively from a British or
Commonwealth point of view. This final chapter summarizes what we
learn when we examine the First World War from this perspective.
First, the German army was far from the perfect instrument for waging
war that it is often assumed to have been. It was neither rational meritocracy
nor blessed with the most flexible of command systems, effortlessly delegat-
ing authority down the chain of command to the man best placed to make
decisions. Command was highly personalized and the officer corps was
riven with cliques and patronage. Auftragstaktik was very much the excep-
tion rather than the rule. The German army never managed to decide who
should have control of the tactical defensive battle and especially of the
counter-attack. Falkenhayn could not resist reaching far down the chain of
command to micro-manage. At the operational level, the only time delega-
tion was ever properly tried was in the autumn of 1916, when Ludendorff
and Hindenburg took over Oberste Heeresleitung (OHL). As soon as things
began to go wrong and Ludendorff started to feel the pressure, however, he
began to exert central control once more. To some extent, this was dictated
by the personalities of the men in charge. It was also, however, largely the
default setting of German command during the First World War, reflecting
the instability of a non-system built on the shaky foundations of the
monarchy in general and Kaiser Wilhelm II in particular. To make imperial
Germany function required the determination and skill of a Bismarck.1
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

278 Conclusions

Whether even he could have overcome the challenge of the First World
War, we shall never know. What is certain, though, is that when Germany
needed talents like his, her leaders fell short.
Centralizing operational decision-making in the hands of one man carried
risks. First, there was the obvious danger of overload. Ludendorff broke
down under the strain of trying to control the deployment of almost every
single artillery battery. As one of his generals complained: ‘he did not know
how to delegate. Such centralization as he operated, only a Napoleon would
have been up to, and Ludendorff was no Napoleon. He stubbornly carved
his own course, with no view of reality, until a kind of military total insanity
gripped him.’2 A further, more insidious, danger existed, too. Information is
never perfect on the battlefield, but the worse it is, the greater the chance of
making bad decisions. Especially from 1917 onwards, the honesty of the
reporting system was increasingly compromised as subordinates competed
to tell their superiors what they wanted to hear. As Ludendorff grew ever
more detached from reality, his grasp of the capabilities of not only the enemy
but even his own forces faltered. During spring 1918 he under-estimated
Allied resilience and over-extended his armies. In the autumn, he failed to
understand the depth of the enemy’s determination and exaggerated the
ability of his men to hold on until it was all too late.
Poor information flow also affected the German army’s ability to learn
the lessons of, and adapt to, the challenges of modern warfare. When, from
1916 onwards, the Entente began to develop the material strength and battle-
field know-how finally to pose a serious threat, the German army proved
unable to keep up with the pace of innovation. There were two obstacles.
The first, as just discussed, was that the mechanism for transmitting know-
ledge around the army broke down. Failure to analyse experience honestly
led to the wrong problems being identified, the wrong lessons being learnt,
and the wrong solutions being put forward.
The second factor inhibiting innovation was more conceptual. Throughout
the First World War the German army sought to address operational, and
sometimes even strategic, problems with tactical solutions. Thus, for example,
it sought to answer Allied attritional attacks with a better scheme of defence.
Similarly, in March 1918 infiltration was elevated from battlefield tactic to
operational method, as Ludendorff looked to exploit weak spots wherever
they were found, rather than fix objectives in advance. This was not because
the Germans misdiagnosed the problem: they certainly understood the concept
of operations. Rather, it was the result of a well-founded appreciation that
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Last Wor ds279

when it came to strategic resources or operational scale, the balance lay

strongly with the Entente, so Germany’s comparative advantage lay at the
tactical level of war. On the battlefield, German soldiers might be able to
out-think and out-manoeuvre their enemies and offset enemy mass with
agility. Since the alternative to believing this was to admit defeat, it is hardly
surprising that they chose to believe. No amount of tactical innovation,
however, could hold back the tide of strategic reality forever.
The German obsession with tactics led to two further tragedies. First,
rather like the British in Blair’s wars, the German army’s confidence in its
ability to solve problems outran its capability and led it to allow a probably
unwinnable war to be fought. When Schlieffen and his successors pretended
that they possessed a military solution to the problem of Germany’s position
in the centre of Europe, sandwiched between France and Russia, the first
steps were taken on the road to disaster. Second, once war had begun, the
Germans still kept looking for tactical answers when it would have been
better for all to accept the strategic logic of the situation.This was especially
the case on the battlefields of the First World War, where manoeuvre was so
difficult and tactical brilliance traded at a discount. If there is a lesson for
today, it is surely that the military needs to be ruthlessly honest, both with
itself and with its politicians, about what it can and cannot achieve.
Third, Rupprecht forces us to revise long-cherished views about the armies
of the Entente, too. Most obviously, his diary reminds us how important the
French were. Not only were they the main enemy from August 1914 to
the end of 1916 but even thereafter he still viewed the French army as a
more dangerous opponent than the British Expeditionary Force (BEF).
Rupprecht tended to look down upon the British, considering them brave
enough most of the time but clumsily handled. Even into 1918 the BEF
remained sometimes poor at coordinating attacks and exploiting any
temporary success it managed to achieve. The main factor behind British
success, Rupprecht thought, was weight of numbers, in men and especially
in artillery. There is little evidence that Rupprecht was aware of, much less
worried by, any British tactical improvement. He was less likely to note
threatening British tactical innovations than instances of them repeating
the  same mistakes. The important exception was Plumer’s artillery-heavy
‘bite-and-hold’ approach in late September 1917. This did cause concern
but was too resource-intensive to be repeated often. Even in 1918, Rupprecht
reckoned the average German soldier superior to a British one. This was a
common view among German troops and officers during the First World
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

280 Conclusions

War and it is hard to know how much weight to give to such opinions.
Were  they an objective description of what German soldiers saw before
them, or did they merely fit existing national stereotypes? It made obvious
sense for the Germans to think this way about their enemy: there were clear
morale benefits while the war went on and it offered a face-saving explan-
ation of defeat once it was over. Better to have been overwhelmed than
outfought, after all.
Fourth, Rupprecht enables us to view learning during the First World
War in a fresh light. Over the last twenty-five years revisionist historians
have painted a picture in which the BEF on the Western Front overcame
the challenges of modern warfare one by one until it had built a ‘weapons
system’ capable of beating the Germans in 1918.This useful corrective to the
‘butchers and bunglers’ stereotype had the benefit of capturing the central
truth that the British army overcame daunting challenges to be much better
at its job by 1918 than it had been earlier in the war. However, it also
risked creating the impression that learning was a one-sided, British-only
phenomenon and that the challenge of modern war was a largely static
one. Finding the right solution, it suggested, was difficult, but possible: a
single answer did exist, and once the BEF had worked it out, they could
unlock the Western Front. The problem is often presented as rather like
working out how to climb a fearsome mountain: far from easy; always
dangerous; subject to numerous unpredictable variables, such as weather;
but essentially a puzzle which, with time, thought, a lot of pluck, and a little
bit of luck, could be unpicked.
Rupprecht shows us that there was more to learning than that. Learning
was a front in the war, and contested just as fiercely as the others. It was a
site of intense and dynamic competition between the combatants. Both
sides took time to accept that they were now part of a new world but, once
they had, measure clashed with counter-measure, was refined, tried again,
parried once more, and so on, all in a never-ending and lethal race to a
destination no-one knew. That the Germans lost this struggle, largely due
to inherent institutional weaknesses within their army, does not diminish
the achievement of the British and French in winning it. The fact that the
learning environment was so much more complex and unforgiving than
some historians have previously estimated only magnifies the respect the
Entente armies are due. On both sides of the wire, ordinary men spent four
years grappling with some of the most extraordinary and complex problems
mankind has ever encountered. They did so under conditions of stress and
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Last Wor ds281

danger which few of us today have ever encountered. That they adapted as
well and as fast as they did testifies to human creativity even in the midst of
the cruellest destruction.
Crown Prince Rupprecht of Bavaria was one of those men.Too proud to
consider himself ordinary, he is not always an immediately sympathetic
character to modern eyes. Like many royals, he could turn on easy-going
charm when required, but just as smoothly behave with chilling reserve and
ruthlessness. His social and political attitudes belonged to the nineteenth
century. He could be short-tempered and prickly, especially when he
thought he was not receiving the respect he was due. His moods swung
rapidly from exhilaration to gloom. He was clearly a nuisance to command.
He was, however, capable of inspiring loyalty and affection in those who
worked for and alongside him. Some of them carried on fighting to defend
his wartime record well into the 1930s, even after Rupprecht himself had
sunk back into dignified silence. He was widely respected for his integrity,
patriotism, and dedication. Rupprecht, to me at least, is not necessarily a
very likeable man but he was a straight, honourable, and decent one, better
perhaps as a business partner than a friend.
In 1914 some, as in Rupert Brooke’s famous sonnet ‘Peace’, gladly turned
from a world turned old and cold and weary and leapt into the cleanness of
war. For them, the greys of peacetime resolved into blacks and whites. Here
they found beauty among the horror, comradeship among the killing, and
some even took delight in destruction.3 Rupprecht was not one of these
men. By temperament he was a professional, rather than a natural, warrior.
He did what he must to execute his duty but derived little pleasure from it.
He took his job seriously. The fact that he kept so full a diary is in itself
evidence of that. He found release and relaxation in physical exercise, art,
and hunting, not battle. He was highly sensitive to personal criticism. His
paranoia was sometimes justified: some people, such as Falkenhayn, genuinely
were trying to undermine him. Perhaps also, however, he was defensive out
of a sense that he was not in his natural element and needed to work extra
hard to prove, both to others and to himself, that he could handle the challenge
of war. In this he was largely successful. He proved far from a figurehead,
much less a dilettante. He set a high personal standard for his officers to follow:
his army and, later, his Army Group, seem to have been well run and to have
operated smoothly, at least until the army as a whole began to break down.
His military judgement was generally good; he made sensible decisions; his
troops were well prepared to fight; he oversaw an efficient logistics system
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

282 Conclusions

which kept them equipped during combat; and he managed and directed
the deployment of his reserves with care and skill. His ability was rewarded
with extended responsibilities in August 1916 and he might even have
been given command of the entire Western Front had it not been for
the  need to tip-toe around the Prussian Crown Prince. When he was
­bypassed by Ludendorff in 1918, this had little to do with his own per-
formance. The same process was occurring all across the German army as
Ludendorff struggled to keep control of a war he could feel slipping away
from him.
Nonetheless, it would be silly to claim that Rupprecht was one of the
great captains of history.The First World War did not produce many of those.
It did shatter the reputations and careers of many, though, and Rupprecht
was good enough to avoid such a fate. He made an uncertain start in 1914:
Sixth Army failed first to tie down French forces in Lorraine and then to
trap the BEF in Flanders. Nonetheless, for the next three years, despite days
of disaster such as 25 September 1916 and 9 April 1917, his men were able to
prevent Entente forces from achieving their objectives in seven major bat-
tles.4 Only in 1918 was he clearly on the losing side in all the battles he
fought. It could be argued that, being largely on the defensive, he had an
easier job than his enemies. On the Western Front the outcome of single
battles was less important than the cumulative effect of many, so the idea
of  victory and defeat is more complicated than it might have been for
Napoleon or Julius Caesar. If we could look at individual battles in isolation,
however, and define winning as achieving one’s own objectives or prevent-
ing the enemy from gaining his, then Rupprecht won more than he lost.
Not many generals in history can say that.
As time went on, he found it harder to deal with the pressures of war,
both mentally and physically. He had trouble sleeping and his health began
to weaken him. Of course, Rupprecht’s sufferings were slight relative to
those of his men. His daily life was comfortable. He slept under a roof and
in a bed every night. His clothes were clean, his food always hot. He was
rarely in personal danger, although he dealt with it well enough when he
was. Most importantly, he survived the war sound in body and mind. Friends
died but the only close family member he lost directly to the war was his
cousin Heinrich, killed in Romania in 1916.The war had, however, torn him
away from his two motherless sons and prevented Rupprecht from looking
after, or even burying, one of them. The defeat and revolution of 1918
robbed him of his throne and his future. The war cost him much. Was
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Last Wor ds283

Rupprecht, however, to return to one of the questions raised in the

Introduction, merely one of its victims? Was he, as I first thought, a man
born too late, overwhelmed by the onrush of the modern world?
Without doubt, the circumstances Rupprecht had to handle were tough
and eventually proved beyond him, as they proved beyond Germany herself.
They brought catastrophe for Rupprecht, for Germany, and eventually
for  the whole world. The old certainties Rupprecht knew and fought to
preserve were lost forever, sucked into the murderous whirlpool of the
twentieth century. Inevitably, he did not have complete freedom of action.
He lived and worked within structures, both institutional and conceptual,
which constrained the decisions he took. As a general, he had to maintain
the trust and respect of his subordinates while satisfying his superiors.
In politics, he had always to work through others. His personality proved
sometimes a help, sometimes a hindrance. Moreover, breeding and upbring-
ing equipped him with the set of cultural and philosophical assumptions
within which he operated. Eventually Rupprecht was able to see through
the strategic group-think which gripped the senior ranks of the German
military government. He saw that the war was hopeless when many others
remain fixated on fighting on. He was more realistic than either his father
or the Kaiser about the political consequences of defeat. Sadly, however, his
conversion came too late: by 1918 the cumulative consequences of four years
of poor decisions, made both individually by Rupprecht and collectively by
Germany, were making themselves felt. Even then, he under-estimated the
cost of failure and how radically his world was about to change. The paths
Rupprecht took were not completely freely chosen, but neither were they
entirely pre-determined.The more I studied Rupprecht, the more I realized
that the choices he made, and indeed the decisions he did not make in time,
affected what happened to him.We tend to describe any terrible accident as
‘tragic’, but Rupprecht’s story is a tragedy in its original, ancient Greek,
sense: a tale of human suffering brought about by the hero’s weakness
and mistakes.
There is another, possibly more important, way of looking at this question.
Reading Rupprecht’s diaries has shown me that he definitely saw himself as
the subject of his own life, not just the object. A proud man with a strong
sense of mission and responsibility, when Rupprecht looked in the shaving
mirror every morning he surely saw himself as a man of action. The whole
purpose of publishing his diary, after all, was to put forward his own account
of his war in the face of criticism from the official historians. Rupprecht’s
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

284 Conclusions

diary thus contradicts the argument of the distinguished cultural historian

Samuel Hynes, himself a combat veteran, that soldiers’ tales are ‘not con-
cerned with why. War narratives are experience books; they are about what
happened, and how it felt. Why is not a soldier’s question.’5 Many of the
most powerful soldiers’ tales from the First World War are stories of suffering.
W. B. Yeats famously excluded poets such as Wilfred Owen, Siegfried Sassoon,
and Isaac Rosenberg from his 1936 Oxford Book of Modern Verse on the
grounds that ‘passive suffering is not a theme for poetry’.6 Between 1914
and 1918, it seemed, ‘men no longer made war; war was made on men’.7
Rupprecht would never have shared such a view. He would have been
repelled by the idea that he was no more than destiny’s plaything. Rupprecht
spoke in the active, not the passive, sense. Tempting as it may be to bucket
together all the soldiers of the First World War as victims, when we do so we
not only distort the historical record but also steal something of their sense
of self from men who can no longer speak for themselves.This betrays their
memory. It also conflicts with one of the primary purposes of history which
surely is to let those whose voices have been stilled, by exclusion or death,
speak again.
There is another wider point about the use of history in general, and the
history of the twentieth century in particular, here. Before the First World
War had even finished, those who saw themselves as its ‘vanquished’ began
to harness victimhood to drive their agendas of vengeance and violence.8
Mussolini and Hitler are only the two most obvious and chilling examples.
Rupprecht, by never trying to exploit victimhood in that way, shows that
there was nothing inevitable about doing so. Both the poets and the men
of violence, for very different reasons and in very different ways, made a
conscious choice to present themselves as pawns. We can see and feel the
literary legacy of the Western Front around us still: due to the power of its
appeal to empathy and emotion, it dominates the historical memory of the
Great War and can be felt in much of how we remember and commemorate
it today.
The consequences of the revanchists’ myth of martyrdom were altogether
more disastrous. Diabolic men led the world to the brink of a new Dark
Age by the lights of their perverted history. Only after the Second World
War had killed millions did they and their lies die in the flames of ruined
Berlin.The final lesson we can learn from Rupprecht, perhaps, is that history
is a powerful weapon which must be kept safe from those who would steal
and twist it to their own evil ends.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Appendix: Note on Military

Terminology

This note is designed to introduce some of the jargon used in the armies of the
First World War and to explain their structure and hierarchy. In 1914 the British,
French, and German armies were all set up along broadly similar lines, but organ-
izations did change as the war went on. Here is a simplified version of the situation
when war broke out. It reflects establishments on paper: practice on active service
might deviate considerably, especially when casualties affected unit strength. As the
war went on all armies tweaked the way they were organized. This note makes no
attempt to keep up with all of these changes.Where they were important, they have
already been identified in the text.
In 1914 the British army was manned entirely by long-serving regular soldiers,
volunteers who had signed up for seven years’ active service followed by five in the
reserves. If war broke out, they could be called back to the colours. As the war
went on, the original, regular British Expeditionary Force (BEF) was reinforced
both by territorial units made up of part-time soldiers originally intended for
home defence and by the wartime volunteers who famously signed up for
Kitchener’s New Armies. By 1916 Britain was running out of volunteers and turned
to conscription, everywhere except Ireland. In Germany and France the system was
different. There, conscripts formed the backbone of the army. After two or three
years of active service, men became reservists until they reached the age of forty-
five. After five years in the first-line reserves German men were transferred to the
second-line Landwehr for ten years and thereafter to the Landsturm, where they
might expect to serve mainly in rear areas. During the war the German army
deployed active, Reserve, Ersatz, and Landwehr divisions in the front line. Apart
from some minor differences in their orders of battle, there was increasingly little
to choose between them. The same applied to the active, Reserve, and Territorial
divisions of the French army, which employed a similar system of conscription and
reservists.
British, German, or French, a new recruit arriving in the trenches to serve as an
infantryman would be assigned to a section of 10–12 men. This was the team with
which he would eat, sleep, and fight. Four sections made up a platoon; four platoons
constituted a company; and four companies made up a battalion. In the German
and French armies three battalions formed a regiment and fought together. Each
division had four regiments. In the British army the regiment served a rather dif-
ferent function. It was an administrative, rather than a fighting, unit and the battalions
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

286 A ppe n di x

of a regiment might never fight together. Different regiments had their own tra-
ditions and proud histories which served as a focus for esprit de corps. In the
British army four battalions would be grouped into a brigade, and three brigades
formed a division. In all three armies, therefore, the division comprised twelve
infantry battalions. The division also included, as part of its permanent establish-
ment, machine-guns, some cavalry, artillery, engineers, and medical units. In other
words, the division was a combined arms formation capable of acting independently.
It was also the building block used to construct higher commands on an ‘as required’
basis. Thus a corps in 1914 generally commanded two divisions, but by 1917–18
might have three or four attached. Four or five corps might be put together into an
army, with a number of armies attached together in an Army Group, such as
Rupprecht commanded.
The table below gives a (simplified) hierarchy of combat ranks within the British,
French, and German armies and gives an idea of the job men at each rank might
perform.The reality was considerably more complex. For instance, the British army
distinguished between ‘ranks’ and ‘appointments’ and between substantive, brevet,
temporary, and acting ranks. The practical effect of these distinctions was limited.
Likewise, it makes little difference that what one line infantry regiment might call
a ‘private’, the Guards would designate as a ‘guardsman’ and the Royal Engineers a
‘sapper’. Or that a Hauptmann in the German infantry would be known as a
Rittmeister in the cavalry. Unit strengths, of course, could vary dramatically, and
casualties might mean that junior officers and non-commissioned officers (NCOs)
found themselves temporarily in command one or even two levels above their
normal station.
Rank Command Approximate number German equivalent French equivalent
of men under command

Field Marshal Army Group 2,000,000 Generalfeldmarschall Maréchal de France

OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

General Army 300,000 Generaloberst Général d’Armée
Lieutenant-General Corps 60,000 General der Infanterie/ Général de Corps
Kavallerie/Artillerie d’Armée
Major-General Division 12,000 Generalleutnant Général de Division
Brigadier-General Brigade 3,500 Generalmajor Général de Brigade
Colonel Oberst Colonel
Lieutenant-Colonel Battalion commanding 1,000 Oberstleutnant Lieutenant-Colonel
officer (CO)
Major Battalion second Major Commandant
in command
Captain Company 200 Hauptmann or Rittmeister Capitaine
Lieutenant or Second Platoon 50 Oberleutnant or Leutnant Lieutenant or
Lieutenant Sous-Lieutenant
Sergeant Platoon second Feldwebel or Sergeant Sergent
in command
Corporal or Lance Section 12 Unteroffizier or Gefreiter Caporal
Corporal
Private None 0 Soldat Soldat
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Notes

Introduction
1. See the Appendix for a note explaining ranks, units, and some other military
details.
2. The passport is in the Geheimes Hausarchiv, Abteilung III, Bayerisches
Hauptstaatsarchiv, Munich (GHA) Nachlaß (NL) Kronprinz Rupprecht (KPR)
255.
3. Early hagiographies in German include: Josef Breg, Kronprinz Rupprecht von
Bayern: Ein Lebensbild (Munich: Max Kellerer, 1918); Otto Kolshorn, Kronprinz
Rupprecht von Bayern: Ein Lebens- und Charakterbild (Munich: R. Piper & Co.,
1918); Kurt Sendtner, Rupprecht von Wittelsbach, Kronprinz von Bayern (Munich:
Richard Pflaum Verlag, 1954); Walter Wilhelm Goetz, Rupprecht: Kronprinz von Bayern
1869–1955 (Munich: Bayerische Akademie für Wissenschaften, 1956). More
modern, analytic treatments of Rupprecht’s political career include: Dieter
J. Weiß, Kronprinz Rupprecht von Bayern: Eine politische Biografie (Regensburg:
Verlag Friedrich Pustet, 2007); Stefan März, Das Haus Wittelsbach im Ersten
Weltkrieg: Chance und Zusammenbruch monarchischer Herrschaft (Regensburg: Friedrich
Pustet, 2013). For the military side, see: Dieter Storz, ‘Kronprinz Rupprecht von
Bayern—dynastische Heeresführung im Massenkrieg’, in Winfried Heinemann and
Markus Pöhlmann (eds), Monarchen und ihr Militär (Potsdam: Militärgeschichtliches
Forschungsamt, 2010), pp. 45–57; Holger Afflerbach, ‘Kronprinz Rupprecht
von Bayern im Ersten Weltkrieg’, Militärgeschichtliche Zeitschrift 75:1 (May 2016),
pp. 21–54.
4. For important and influential books on the First World War and modernity, see
Paul Fussell, The Great War and Modern Memory (London: Oxford University
Press, 1977) and Modris Eksteins, The Rites of Spring: The Great War and the Birth
of the Modern Age (London: Bantam, 1989). They mean very specific things by
‘modernity’, involving the use of irony and the scope allowed to the individual.
Here I use ‘modern’ in a more popular sense as a concept including senses of: a
break from the traditional past; progress; speed; scale; the machine; urbanization.
5. Notable exceptions to this rule, by historians with a good feel for military realities,
include: Hew Strachan, The First World War Volume I: To Arms (Oxford: Oxford
University Press, 2001); David Stevenson, 1914–1918: The History of the First
World War (London: Allen Lane, 2004); Stevenson, With Our Backs to the Wall
(London: Allen Lane, 2011), although all three works cover much more than
just the Western Front.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

290 Not e s to Page s 3–4

6. Books on the former include: Martin Kitchen: The Silent Dictatorship: The
Politics of the German High Command under Hindenburg and Ludendorff, 1916–1918
(London: Croom Helm, 1976); Robert B. Asprey, The German High Command at
War: Hindenburg and Ludendorff Conduct World War I (New York:William Morrow,
1991); better are: Holger Afflerbach, Falkenhayn, Politisches Denken und Handeln
im Kaiserreich (Munich: R. Oldenbourg, 1994); Holger H. Herwig, The First
World War: Germany and Austria-Hungary 1914–1918 (London: Arnold, 1997);
Annika Mombauer, Helmuth von Moltke and the Origins of the First World War
(Cambridge: Cambridge University Press, 2001); Robert T. Foley, German Strategy
and the Path to Verdun: Erich von Falkenhayn and the Development of Attrition 1870–
1916 (Cambridge: Cambridge University Press, 2005); Alexander Watson, Ring of
Steel: Germany and Austria-Hungary at War, 1914–1918 (London: Allen Lane, 2014).
For the latter, see: Benjamin Ziemann, War Experiences in Rural Germany 1914–
1923 (Alex Skinner, trans.) (Oxford: Berg, 2007); Bernd Ulrich and Benjamin
Ziemann (eds), German Soldiers in the Great War: Letters and Eyewitness Accounts
(Barnsley: Pen and Sword, 2010); Jack Sheldon, The Germany Army on the Somme
1916 (Barnsley: Leo Cooper, 2005); Sheldon, The German Army at Passchendaele
(Barnsley: Pen & Sword, 2007); Sheldon, The German Army on Vimy Ridge 1914–
1917 (Barnsley: Pen & Sword, 2008); Sheldon, The German Army at Cambrai
(Barnsley: Pen & Sword, 2009); Sheldon, The German Army At Ypres, 1914
(Barnsley: Pen & Sword, 2010) and others. There have also been excellent
­studies of individual battles, as well as useful overviews of the longer course of
German military history: Dennis Showalter, Tannenberg: Clash of Empires, 1914
(Hamden: Archon, 1991); Christopher Duffy, Through German Eyes: The British
and The Somme 1916 (London: Weidenfeld & Nicolson, 2006); David T. Zabecki,
The German 1918 Offensives: A Case Study in the Operational Level of War (New
York: Routledge, 2006); Robert M. Citino, The German Way of War: From the
Thirty Years’ War to the Third Reich (Lawrence, KS: University Press of Kansas,
2005). Robert Foley has published a useful series of specialist articles discussing
innovation and the German army: Robert T. Foley, ‘Learning War’s Lessons:The
Germany Army and the Battle of the Somme 1916’, Journal of Military History
75 (April 2011), pp. 471–504; Foley, ‘A Case Study in Horizontal Military
Innovation: The German Army 1916–1918’, Journal of Strategic Studies 35:6
(2012), pp. 799–827; Foley, ‘Dumb Donkeys or Cunning Foxes? Learning in the
British and German Armies during the Great War’, International Affairs 90:2
(2014), pp. 279–98.
7. Although the Zentrum für Militärgeschichte und Sozialwissenschaften
der Bundeswehr has produced an impressive array of work, such as: Christian
Stachelbeck, Militärische Effektivität im Ersten Weltkrieg: Die 11. Bayerische
Infanteriedivision 1915 bis 1918 (Paderborn: Ferdinand Schöningh, 2010); Gerhard
P. Groß, Mythos und Wirklichkeit: Geschichte des operativen Denkens im deutschen
Heer von Moltke d.Ä bis Heusinger (Paderborn: Ferdinand Schöningh, 2012).
8. Examples of this school of thought include: Trevor N. Dupuy, A Genius for War:
The German Army and General Staff, 1807–1945 (London: Macdonald and Jane’s,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 4– 6291

1977); Timothy T. Lupfer, The Dynamics of Doctrine: The Change in German

Tactical Doctrine during the First World War (Leavenworth Paper No. 4: Combat
Studies Institute, U.S. Army Command and General Staff College, Fort
Leavenworth, KS, 1981); Martin van Creveld, Command in War (Cambridge,
MA: Harvard University Press, 1985); Allan R. Millett and Williamson Murray
(eds), Military Effectiveness, 3 volumes (Boston, MA: Allen & Unwin, 1988);
Martin Samuels, Command or Control? Command,Training and Tactics in the British
and German Armies, 1888–1918 (London: Frank Cass, 1995).
9. Winston S. Churchill, TheWorld Crisis, 6 volumes (London: Thornton Butterworth,
1923–31); David Lloyd George, War Memoirs, 6 volumes (London: Nicholson &
Watson, 1933–6); Basil Liddell Hart, The Real War (London, Faber & Faber,
1930; expanded and republished as A History of the World War, 1914–1918 (1934)
and as History of the First World War (1970)); Alan Clark, The Donkeys (London:
Hutchinson, 1961); A. J. P. Taylor, The First World War: An Illustrated History
(London: Hamish Hamilton, 1963); John Laffin, British Butchers and Bunglers of
World War One (Stroud: Sutton, 1988).
10. Paddy Griffith, Battle Tactics of the Western Front: The British Army’s Art of Attack
1916–18 (New Haven, CT:Yale University Press, 1994); Gary Sheffield, Forgotten
Victory: The First World War: Myths and Realities (London: Headline, 2001). See
also, among many others, Bill Rawling, Surviving Trench Warfare: Technology and
the Canadian Corps 1914–1918 (Toronto: University of Toronto Press, 1992);
Albert Palazzo, Seeking Victory on the Western Front:The British Army and Chemical
Warfare in World War I (Lincoln, NE: University of Nebraska Press, 2000); Peter
Simkins, From the Somme to Victory: The British Army’s Experience on the Western
Front 1916–1918 (Barnsley: Pen & Sword, 2014).
11.  Patrick Leigh Fermor, A Time of Gifts: On Foot to Constantinople: From the Hook
of Holland to the Middle Danube (London: Penguin, 1988 [1978]); Leigh Fermor,
Between the Woods and the Water: On Foot to Constantinople from the Hook of
Holland, the Middle Danube to the Iron Gates (London: John Murray, 1986).
12. Author’s note, Erich Maria Remarque, All Quiet on the Western Front (Brian
Murdoch, trans.) (London: Vintage, 2005 [1994]), n.p. and p. 208; Wilfred Owen,
‘Anthem for Doomed Youth’, http://www.oucs.ox.ac.uk/ww1lit/collections/
item/3290, accessed 7 January 2017; Stephane Audoin-Rouzeau and Annette
Becker, 1914–1918: Understanding the Great War (Catherine Temerson, trans.)
(London: Profile, 2002), p. 2.
13. Paul Fussell’s seminal The Great War and Modern Memory (Oxford: Oxford
University Press, 2000 [1977]) is the most complete statement of the ‘soldier as
victim’ hypothesis, but see the critique in Robin Prior and Trevor Wilson, ‘Paul
Fussell at War’, War in History 1:1 (1994), pp. 63–80; also: Joanna Bourke,
Dismembering the Male: Men’s Bodies, Britain and the Great War (London: Reaktion,
1996); Bourke, An Intimate History of Killing: Face-to-Face Killing in Twentieth
Century Warfare (London: Granta, 2000 [1999]), p. 375; Niall Ferguson, The Pity
of War (London: Allen Lane, 1998), p. 357. For a more convincing exploration of
some of the ways in which soldiers tried to assert agency over their situation
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

292 Not e s to Page s 6 –13

and environment, see Tony Ashworth, Trench Warfare 1914–1918: The Live and
Let Live System (London: Macmillan, 1980) and, better still, Alexander Watson,
‘Self-Deception and Survival: Mental Coping Strategies on the Western Front,
1914–18’, Journal of Contemporary History 41:2 (2006), pp. 247–68.
14. Hans Binneveld, From Shell Shock to Combat Stress:A Comparative History of Military
Psychiatry (Amsterdam: Amsterdam University Press, 1997), p. 57.
15. I am grateful to one of the anonymous readers of the first draft of this manu-
script for suggesting this line of inquiry to me. As Holger Afflerbach points
out, if it is hard to establish the precise significance in history of supreme
commanders such as Wilhelm II or Hitler, it is even harder for those such
as Rupprecht a level or two further down the ladder: Afflerbach, ‘Kronprinz
Rupprecht’, p. 22.
16. The published version is Rupprecht, Kronprinz von Bayern (Eugen von
Frauenholz, ed.), Mein Kriegstagebuch (Berlin: E. S. Mittler & Sohn, 3 volumes,
1929), abbreviated hereafter in these notes to Rupprecht, Kriegstagebuch. The
manuscript version, which will be referred to as Rupprecht MS diary, is to be
found in GHA NL KPR, 699–708. For ease of reference, wherever possible
I have quoted from the former. Entry, 1 October 1917, Rupprecht MS diary,
GHA NL KPR 707, pp. 2988–96. For an excellent discussion of the dispute
between Rupprecht’s detractors and supporters, see Markus Pöhlmann,
Kriegsgeschichte und Geschichtspolitik: Der Erste Weltkrieg: Die amtliche deutsche
Militärgeschichtsschreibung 1914–1956 (Paderborn: Ferdinand Schöningh, 2002),
pp. 284–321.
17. Bayerisches Kriegsarchiv, Abteilung IV, Bayerisches Hauptstaatsarchiv, Munich
(BKA).
18. The German official history, Der Weltkrieg 1914 bis 1918, was published in four-
teen volumes between 1925 and 1944, although the last volumes were not made
public until 1956. The name of the institution producing it changed several
times. For simplicity, it will here be referred to by GOH with the volume num-
ber. Similarly for the French (Les Armées Françaises dans la Grande Guerre) and
British (Military Operations France and Belgium) official histories, with FOH and
BOH respectively.

Chapter 1
1. The details on Rupprecht’s early life which follow are primarily drawn from Weiß,
Kronprinz Rupprecht and Sendtner, Rupprecht von Wittelsbach. The Leuchtenberg
Palace, in which he was born, stands on Odeonsplatz and now houses the
Bavarian State’s Ministry of Finance.
2. März, Das Haus Wittelsbach, pp. 62–9.
3. Statistics largely from tables in Volker R. Berghahn, Imperial Germany, 1871–1914:
Economy, Society, Culture, and Politics (Providence, RI: Berghahn Books, 1994),
pp. 294–339; Roger Chickering, Imperial Germany and the Great War, 1914–1918
(2nd edn, Cambridge: Cambridge University Press, 2004): James Retallack (ed.),
Imperial Germany 1871–1918 (Oxford: Oxford University Press, 2008).
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 14–20293

4. Breg, Kronprinz Rupprecht, p. 13; letter to author from Rupprecht’s grandson,

Herzog Franz von Bayern, 3 July 2015; Kolshorn, Kronprinz Rupprecht, pp. 47–8.
5. Quoted in Weiß, Kronprinz Rupprecht, p. 68.
6. März, Das Haus Wittelsbach, pp. 148–56.
7. Foreword, Rupprecht MS Diary, GHA NL KPR 699.
8. John Röhl, Wilhelm II: Into the Abyss of War and Exile 1900–1941 (Sheila de Bellaigue
and Roy Bridge, trans.) (Cambridge: Cambridge University Press, 2014),
pp. 1092–3.

Chapter 2
1. Some of the numbers in this book are so large that they can be hard to visualize.
Some comparisons might help: 250,000 people use Waterloo Station every day;
Wembley Stadium holds 90,000 and Stamford Bridge a little over 40,000; the
University of Birmingham has about 35,000 students; 10,000 soldiers marching
four abreast would take about two hours to go past; a Northern Line tube train
holds about 1,000 people.
2. Citino, The German Way of War, pp. 131, 150; Crown Prince William of Germany,
My War Experiences (London: Hurst and Blackett, n.d.), p. 4.
3. Dieter Storz, ‘ “Dieser Stellungs- und Festungskrieg ist scheußlich!” Zu den
Kämpfen in Lothringen und in des Vogesen im Sommer 1914’, in Hans Ehlert,
Michael Epkenhans, and Gerhard P. Groß (eds), Der Schlieffenplan: Analyse und
Dokumente (Paderborn: Ferdinand Schöningh, 2006), pp. 162–204, p. 166; entry,
31 July 1914, Krafft Diary, BKA NL Krafft 145; entry, 1–5 August 1914, Rupprecht
MS Diary, GHA NL KPR 699, pp. 4–5.
4. Christian Stachelbeck, Deutschlands Heer und Marine im Ersten Weltkrieg (Munich:
Oldenbourg, 2013), pp. 116–37; ‘Aufmarschanweisung für Oberkommando der
6. Armee’, 2 August 1914, GHA NL KPR 476, paras 38, 44.
5. Entries, 1–5 August and 9 September 1914, Rupprecht MS Diary, GHA NL
KPR 699, pp. 4–5, 194; Ehlert, Epkenhans, and Groß (eds), Der Schlieffenplan,
pp. 399–484; Krafft,‘Kurzer Überblick über die Vorgänge beim Oberkommando
der 6. Armee im August 1914 bis zur Schlacht in Lothringen am 20.8.1914’,
GHA NL KPR 699, pp. 7–9; Robert Doughty, Pyrrhic Victory: French Strategy and
Operations in the Great War (Cambridge, MA: Belknap Press, 2005), pp. 12–13;
Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 10.
6. Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege (Munich: Verlag des
Bayerischen Kriegsarchivs, 1923), pp. 5, 16; Holger H. Herwig, The Marne, 1914:
The Opening of World War I and the Battle that Changed the World (New York:
Random House, 2009), p. 46; GOH Volume I, p. 144. One corps consisted of
1,500 officers, 40,000 men, and 14,000 horses, most of them pulling 2,400
wagons for supplies and ammunition.
7. Krafft, ‘Kurzer Überblick über die Vorgänge beim Oberkommando der 6. Armee
im August 1914 bis zur Schlacht in Lothringen am 20.8.1914’, GHA NL KPR
699, p. 2; ‘Erste Beurteilung der Lage’, Krafft Diary, BKA NL Krafft 145: Annex;
Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 13.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

294 Not e s to Page s 20 –25

8. Entry, 6–9 August 1914, Rupprecht MS Diary, GHA NL KPR 699, p. 8.

9. Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 15.
10. Sixth Army headquarters had no wireless and only one telephone in August
1914. Even once OHL moved to the Western Front, the one-way trip from
there to Sixth Army could take nearly three hours by automobile: Rudolf von
Xylander, Deutsche Führung in Lothringen 1914: Wahrheit und Kriegsgeschichte
(Berlin: Junker und Dünnhaupt Verlag, 1935), p. 22; Rupprecht, Kriegstagebuch
Volume I, footnote pp. 85–6; Krafft’s note on entry, 2 September, Rupprecht MS
Diary, GHA NL KPR 699, pp. 149–51.
11. Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716,
pp.  16–17; entry, 14 August 1914, Rupprecht MS Diary, GHA NL KPR 699,
p. 16; GOH Volume I, pp. 644, 647–9.
12. Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege, pp. 18–23; GOH Volume
I, p. 174; Rupprecht, Kriegstagebuch Volume I, p. 10.
13. FOH I:1, pp. 161, 179–80, 230–5; Michel Goya, La Chair et l’Acier: L’Invention de
la Guerre Moderne (1914–1918), p. 181; Bayerisches Kriegsarchiv, Die Bayern im
Großen Kriege, p. 24.
14. Rupprecht, Kriegstagebuch Volume I, p. 9; Krafft, ‘Kurzer Überblick über die
Vorgänge beim Oberkommando der 6. Armee im August 1914 bis zur Schlacht
in Lothringen am 20.8.1914’, GHA NL KPR 699, p. 14.
15. Krafft, ‘Kurzer Überblick über die Vorgänge beim Oberkommando der 6.
Armee im August 1914 bis zur Schlacht in Lothringen am 20.8.1914’, GHA
NL KPR 699, pp. 2, 11; entry, 17 August 1914, Rupprecht MS Diary, GHA NL
KPR 699, p. 36; Rupprecht, Kriegstagebuch Volume I, pp. 6–21; Konrad Krafft
von Dellmensingen, Die Führung des Kronprinzen Rupprecht von Bayern auf dem
linken deutschen Heeresflügel bis zur Schlacht in Lothringen im August 1914 (Berlin:
E. S. Mittler & Sohn, 1925), pp. 17–21; GOH Volume I, p. 256. See also pp. 208,
210–11.
16. Instruction Générale No. 3, FOH Volume I:1 Annex 474, pp. 441–2; FOH
Volume I:1, pp. 250–4; Elizabeth Greenhalgh, Foch in Command:The Forging of
a First World War General (Cambridge: Cambridge University Press, 2011),
pp. 14–19.
17. Entry, 20 August 1914, Rupprecht MS Diary, GHA NL KPR 699, p. 48; GOH
Volume I, p. 451; Rupprecht, Kriegstagebuch Volume I, p. 30.
18. FOH Volume I:1, pp. 252–60; Rupprecht, Kriegstagebuch Volume I, pp. 36, 41.
19. Entry, 21 August 1914, ‘Auszug aus einem am 6.6.1916 vom AOK 7 an die OHL
eingereichten Bericht, AOK 7 Operations 3 August–16 September 1914’, BKA
Armeeoberkommando (AOK) 6 Vorl. 9; Ludwig von Gebsattel, Von Nancy bis
zum Camp des Romains 1914 (Oldenburg: Gerhard Stalling 1924), p. 14; Herwig,
The Marne, 1914, p. 103; entry, 24 August 1914, Rupprecht MS Diary, GHA NL
KPR 699, pp. 69, 72.
20. Entry, 27 August 1914, Rupprecht MS Diary, GHA NL KPR 699, pp. 110–28;
entries, 24, 25 August 1914, BKA NL R. Xylander 12 Kriegstagebuch 1914/18
Volume I, pp. iv–v; Gebsattel, Von Nancy bis zum Camp des Romains, pp. 23,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 26 –32295

22–36; John Horne and Alan Kramer, German Atrocities 1914: A History of Denial
(New Haven, CT:Yale University Press, 2001), pp. 66–7; Fayolle cited in Michel
Goya, La Chair et l’Acier, p. 183; Rupprecht, Kriegstagebuch Volume I, p. 53.
21. Rupprecht, Kriegstagebuch Volume I, p. 58.

Chapter 3
1. Rupprecht, Kriegstagebuch Volume I, p. 62; telegram, 30 August 1914, correspond-
ence with King Ludwig III, GHA NL KPR 434; letter from Wenninger, 30 August
1914, in Bernd F. Schulte, ‘Neue Dokumente zum Kriegsausbruch und
Kriegsverlauf 1914’, Militärgeschichtliche Mitteilungen 25:1 (1979), pp. 123–85: p. 159.
2. Storz, ‘ “Dieser Stellungs- und Festungskrieg ist scheußlich!” ’, pp. 162–204:
pp. 187–8; Herwig, The Marne, 1914, p. 231; Wenninger diary entry, 28–9 August
and report, 1 September 1914, in Schulte, ‘Neue Dokumente’, pp. 158, 160.
3. Entry, 31 August 1914, ‘Auszug aus einem am 6.6.1916 vom AOK 7 an die OHL
eingereichten Bericht, AOK 7 Operations 3 August–16 September 1914’, BKA
AOK 6 Vorl. 9; Rupprecht, Kriegstagebuch Volume I, pp. 76–86.
4. Gebsattel, Von Nancy bis zum Camp des Romains, pp. 51–4.
5. Rupprecht, Kriegstagebuch Volume I, pp. 93–5, 99.
6. Rupprecht, Kriegstagebuch Volume I, pp. 103–7.
7. The post-war official German army medical report gave total losses (sick, wounded,
dead, and missing) in Sixth and Seventh Armies in August and September as
127,496, of whom 38,875 were dead. The two armies had begun with 345,000
men between them: Heeres-Sanitätsinspektion des Reichskriegsministeriums,
Sanitätsbericht über das Deutsche Heer (Deutsches Feld- und Besatzungheer) im
Weltkriege 1914/1918 Volume III: Die Krankenbewegung bei dem Deutschen Feld-
und Besatzungsheer im Weltkriege 1914/1918 (Berlin: E. S. Mittler & Sohn, 1934),
pp. 36–9; Storz, ‘ “Dieser Stellungs- und Festungskrieg ist scheußlich!” ’,
pp. 162–204: p. 204; Gebsattel, Von Nancy bis zum Camp des Romains, p. 78, foot-
note to p. 87; Rupprecht, Kriegstagebuch Volume I, p. 118; Elizabeth Greenhalgh,
The French Army and the First World War (Cambridge: Cambridge University
Press, 2014), pp. 58–9.
8. Horne and Kramer, German Atrocities 1914, pp. 20, 22, 63–7.
9. Horne and Kramer, German Atrocities 1914, pp. 74–6, 95, 194–5, 349–50; Isabel V.
Hull, Absolute Destruction: Military Culture and the Practices of War in Imperial
Germany (Ithaca, NY: Cornell University Press, 2005), pp. 208–12; Hull, A Scrap
of Paper: Breaking and Making International Law during the Great War (Ithaca, NY:
Cornell University Press, 2014), pp. 51–60.
10. Letter to Krafft from the Supreme State Prosecutor, b.J. 353/20/12, 3 March
1923, Miscellaneous Documents regarding Crown Prince Rupprecht, BKA NL
Krafft 196; Horne and Kramer, German Atrocities 1914, pp. 349–51.
11. Rupprecht, Kriegstagebuch Volume I, p. 34; entry 21 August 1914, Rupprecht MS
Diary, GHA NL KPR 699, pp. 55–6.
12. Hull, A Scrap of Paper, p. 57 summarizes views she developed in Absolute Destruction.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

296 Not e s to Page s 33–4 2

13. Entry, 21 August 1914, Rupprecht MS Diary, GHA NL KPR 699, pp. 55–6;
Army Order of the Day, 24 August 1914, Campaign 1914: Documents from
Section Ic/d, BKA AOK 6 Vorl. 220; Special Order for 30 August 1914, Ic No.
1129, 29 August 1914 and Special Proclamation, 9 September 1914, both in
Special Orders, BKA AOK 6 Vorl. 206;Army Order of the Day for 19 September
1914, Army Orders of the Day, BKA AOK 6 Vorl. 611 and Campaign 1914:
Documents from Section Ic/d Vorl. 221.
14. Rupprecht, Kriegstagebuch Volume I, pp. 65–6; Hull, Absolute Destruction, pp. 208–12;
Hull, A Scrap of Paper, chapter 3; Röhl, Wilhelm II: Into the Abyss, pp. 77, 1129.

Chapter 4
1. Rupprecht, Kriegstagebuch Volume I, pp. 126–7; entry, 18 September 1914, Krafft
diary, BKA NL Krafft 48, pp. 1a–2a; Foley, German Strategy and the Path to Verdun,
pp. 99–102.
2. Rupprecht, Kriegstagebuch Volume I, p. 132; entries, 18, 21 September 1914,
Krafft diary, BKA NL Krafft 48, pp. 1a–2a, 11a.
3. There had already been some sharp skirmishes in the area in August but the
autumn fighting was on a different scale and more important:William Philpott,
Bloody Victory: The Sacrifice on the Somme and the Making of the Twentieth Century
(London: Little, Brown, 2009), pp. 18–19.
4. Entry, 27 September 1914, Krafft diary, BKA NL Krafft 48, pp. 33a, 35b; Rupprecht,
Kriegstagebuch Volume I, p. 161.
5. Philpott, Bloody Victory, pp. 27–8; Rupprecht, Kriegstagebuch Volume I, p. 165. In
fact, no British units were nearby. Sheldon, The German Army on the Somme, p. 34.
6. Entry, 2 October 1914, Rupprecht MS Diary, GHA NL KPR 700. This sentence
is omitted from Rupprecht, Kriegstagebuch Volume I, p. 180.
7. Entries, 26, 28 September, 4 October 1914, Krafft diary, BKA NL Krafft 48,
pp.  28a–b, 36b, 52a; entry, 2 October 1914, Rupprecht MS Diary, GHA NL
KPR 700.
8. GOH Volume V, pp. 216–17.

Chapter 5
1. Rupprecht, Kriegstagebuch Volume I, p. 206. The sentences in italic feature only
in the MS diary: entry, 14 October 1914, Rupprecht MS Diary, GHA NL KPR,
700. Entries, 13, 14 October 1914, Krafft diary, BKA NL Krafft 48, pp. 81b–82b,
84a–87a.
2. GOH Volume V, p. 286; entries 14, 15 October 1914, Krafft Diary, BKA NL
Krafft 48, pp. 84a–87a, 91a; Rupprecht, Kriegstagebuch Volume I, p. 206.
3. Richard Holmes, The Little Field Marshal: Sir John French (London: Jonathan
Cape, 1981), p. 244; Rupprecht, Kriegstagebuch Volume I, pp. 215, 221–2.
4. Rupprecht, Kriegstagebuch Volume I, p. 216; AOK 6 War Diary, BKA AOK 6 Bd
1 Vorl 1, p. 62; report to King, 30 October 1914, Correspondence with King
Ludwig III, GHA NL KPR 428.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 43–46297

5. Armeetagesbefehl, 19 October 1914, Operations, GHA NL KPR 478; also

reproduced in Rupprecht, Kriegstagebuch Volume III, pp. 57–8.
6. ‘ “Annihilate the English” ’, Daily Mail, 2 November 1914, p. 4; ‘Hatred of
England’, The Times, 29 October 1914, p. 8; ‘National Hatred’, The Times,
1 November 1914, p. 3; ‘Germany’s “most hated foe” ’, The Times, 9 November
1914, p. 9; ‘Rupprecht’s Call for Help’, Daily Mail, 29 September 1915, p. 5;
‘Enemy Generals in the West’, Daily Mail, 25 March 1918, p. 2; ‘Shooting of
British Prisoners’, Manchester Guardian, 11 May 1915, p. 8; ‘Murder of British
Prisoners’, Daily Mail, 11 May 1915, p. 6; ‘House of Commons’, Manchester
Guardian, 12 May 1915, p. 4; Hansard HC Deb 19 May 1915 Volume 71 cc
2304–5; Rupprecht, Kriegstagebuch Volume I, p. 361. The allegations were a
mixture of hearsay and first-person testimony made by German deserters to
Dutch journalists. One of the stories was of forty British soldiers being
burned alive in an aircraft hanger. Rupprecht and Krafft always denied any
such order, and it is probably impossible to check the veracity of the reports
of murders at this distance. See the discussion of this issue in Weiß, Kronprinz
Rupprecht, pp. 174–8.
7. Rupprecht, Kriegstagebuch Volume I, pp. 227, 225.
8. One of the earliest official versions appears in German General Staff, Ypres,
1914: An official account published by order of the German General Staff (translation
of Die Schlacht an der Yser und bei Ypern im Herbst 1914 (Oldenburg: Gerhard
Stalling, 1918) by G. C. W[ynne]), (Nashville, TN: Battery Press, 1994 [1919]),
p. 43. See Ian F. W. Beckett, Ypres:The First Battle, 1914 (Harlow: Pearson, 2004),
pp. 46–8, 98–109, 237–40 for a good summary of the myth, which is more fully
explored in Karl Unruh, Langemarck: Legende und Wirklichkeit (Koblenz:
Bernard & Graefe, 1986); Bayerisches Kriegsarchiv, Die Bayern im Großen
Kriege, p. 137.
9. Rupprecht, Kriegstagebuch Volume I, pp. 232–4; Diary entry 27 October 1914,
MS Diary, GHA NL Kronprinz Rupprecht, 700, pp. 356–7. Sentence in italics
only in MS diary.
10. [Wilhelm] Solger, ‘Zur Ausarbeitung des Reichsarchivs “Die Erwägungen über
einen Durchbruchs-Angriff gegen die westliche Stellungsfront von ende
Februar bis 13.April 1915” ’, dated 24 January–2 February 1930, pp. 7–8, Schemes
for Breakthrough in the West, BKA NL Krafft 189; Bayerisches Kriegsarchiv,
Die Bayern im Großen Kriege, pp. 139–40; ‘Gedanken über die Führung der
Operationen und Erfahrungen im Weltkriege’, BKA, NL Krafft 324, p. 217a;
GOH Volume V, p. 336; German General Staff, Ypres, 1914, p. 63.
11. Rupprecht, Kriegstagebuch Volume I, p. 236.
12. German General Staff, Ypres, 1914, p. 65.
13. First estimates were that Fabeck had lost 10,000, or one-third, of his men
between 30 October and 2 November. The reality was even worse: 17,250. The
6th Bavarian Reserve Division suffered 40 per cent casualties in just five days.
Fourth Army lost 39,000 dead and wounded with 13,000 missing, while Sixth
Army, including Fabeck, suffered 27,000 dead and wounded with another 1,000
missing. Rupprecht, Kriegstagebuch Volume I, pp. 240, 244, 246; GOH Volume V,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

298 Not e s to Page s 46 –51

pp. 346, 401; entries 3, 5 November 1914, Rupprecht MS Diary, GHA NL KPR
701, pp. 394, 402.
14. Holmes, The Little Field Marshal, pp. 248–9, 252–3; BOH 1914 Volume II, p. 383;
entries, 4, 6 November 1914,War Diary, BKA AOK 6 Bd 1 Vorl No. 1, pp. 67, 73;
Rupprecht, Kriegstagebuch Volume I, p. 245; Hermann Cron, Imperial German
Army 1914–18: Organisation, Structure, Orders-of-Battle (C. F. Colton, trans.)
(Solihull: Helion, 2002), footnote 17, p. 136.
15. GOH Volume V, p. 401; GOH Volume VI, pp. 1, 10, 15; Foley, German Strategy
and the Path to Verdun, p. 103.
16. GOH Volume VI, p. 17.
17. Rupprecht, Kriegstagebuch Volume I, pp. 250–3; entry, 14 November 1914, War
Diary, BKA AOK 6 Bund 1 Vorl No. 1, p. 75; GOH Volume VI, pp. 19–23, 371–2.
18. Of these, 7,960 were dead and 17,873 missing. The ‘missing’ fell into three
groups: a few would have got lost somehow and rejoined their units later; many
or most would be dead, their bodies in some way unrecoverable in No Man’s
Land; lastly, some others might have fallen into enemy hands as prisoners. In
this case, slightly over 7,000 turned out to have been captured by the enemy:
BOH 1914 Volume II, p. 383; Statistical Abstract of Information Regarding the
Armies at Home and Abroad, No. 25, 1 October 1918,TNA WO 394/10, p. 137;
War Office, Statistics of the Military Effort of the British Empire during the Great War
1914–1920 (London: HMSO, 1922), p. 253.
19. Sixth Army lost 27,000 dead and wounded in the second half of October, plus
a thousand missing. The figures for the first ten days of November seem to have
been lost but 9,500 were killed or wounded, and another thousand went miss-
ing, between 10 and 18 November: Beckett, Ypres, pp. 225–6; GOH Volume V,
p. 401; GOH Volume VI, p. 25.
20. ‘Die Erwägungen über einen Durchbruchs-Angriff gegen die westliche
Stellungsfront von ende Februar bis 13. April 1915’, BKA NL Krafft 189,
pp. 8–13 and Beilage 1, p. 2; entry, 9 November 1914, War Diary, BKA AOK 6
Bd 1 Vorl 1, p. 72.
21. BOH 1914 Volume II, pp. 464–5; J. P. Harris, Sir Douglas Haig and the First World
War (Cambridge: Cambridge University Press, 2008), p. 377; Gary Sheffield,
The Chief: Douglas Haig and the British Army (London: Aurum, 2011), pp. 98, 240;
German General Staff, Ypres, 1914; GOH Volume V, p. 578; Foley, German
Strategy and the Path to Verdun, pp. 100–1; entry 12 November 1914, Rupprecht
MS Diary, GHA NL KPR, 701, p. 432.
22. Michael Howard, ‘Men against Fire: Expectations of War in 1914’, International
Security 9:1 (Summer 1984), pp. 41–57.
23. Rupprecht, Kriegstagebuch Volume I, p. 10; Terence Zuber, The Mons Myth: A
Reassessment of the Battle (Stroud: History Press, 2010), pp. 48–54; FOH Volume I:1,
pp. 179–80; Goya, La Chair et l’Acier, p. 181; Joffre to army and corps commanders,
13 August 1914, Annexe 67, FOH I:2 Annex Volume I, pp. 62–3; Report,
1 October 1914, Operations File GHA NL KPR 477.
24. Report to Sixth Army, 1 October 1914, Operations, GHA NL KPR 477; First
Army No. 1704, 11 September 1914,‘Instructions du général Dubail, commandant
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 51– 60299

la Ire armée, au sujet des procédés de combat à employer vis-à-vis des Allemands’,
Annexe 3258, FOH I:3 Annexe Volume 3, pp. 450–1; Krafft Diary, BKA NL
Krafft 145, p. 36b; Chef des Generalstabes des Feldheeres M.J. No. 6505,
11 October 1914, BKA AOK 6 Vorl. 587.
25. GQG No. 3565 ‘Note pour les commandants d’armée’, 15 October 1914,Annexe
2864, FOH I:4 Annexe Volume III, pp. 573–5; entry, 22 October 1914, Krafft
Diary, BKA NL Krafft 145, p. 111a.
26. ‘Aufmarschanweisung für Oberkommando der 6. Armee’, 2 August 1914, para-
graph 21, GHA NL KPR 476; Rupprecht, Kriegstagebuch Volume I, pp. 71–2; III
Bayerische Armeekorps (BAK) Lessons Learnt Report, 31 August 1914, BKA
AOK 6, Vorl. 164.
27. AOK 6 Order, 11 December 1914, BKA AOK 6, Vorl. 164; AOK 6 J. No. 5118
26 December 1914, Operations, GHA NL KPR 481; ‘Rundschreiben No. 6 der
7. Englischen Division vom 27.11.1914: “Lessons learnt in recent defensive
­battles” ’, BKA AOK 6 Vorl. 169; AOK 1 Ia No. 834 ‘Erfahrungen aus den fran-
zösischen Angriffen gegen die Stellung des Gren. Rgts. 89 in der Zeit vom 21.
bis 25. Dezember, 31 December 1914’, 4 January 1915, BKA AOK 6 Vorl. 164;
AOK 6 No. 5702, 7 January 1915, Operations, GHA NL KPR 481; Hermann
von Kuhl, Der deutsche Generalstab in Vorbereitung und Durchführung des Weltkrieges
(Berlin: E. S. Mittler & Sohn, 1920), pp. 187–8.
28. Michel Goya, La Chair et L’Acier, pp. 232–4; GQG Nos 1981 & 1982,‘Instruction
relative à l’emploi de l’artillerie’, 9 November 1914, Annexes 4077–8, FOH I:4
Annexe Volume 4, pp. 742–5; Nikolas Gardner has argued that the learning
processes of the BEF worked effectively in the autumn of 1914, although the
conceptual limitations of senior commanders prevented them from learning
the correct lessons: ‘The Beginning of the Learning Curve: British Officers and
the Advent of Trench Warfare, September–October 1914’ (ESRI Working Papers:
Salford, 2003) and Gardner, Trial by Fire: Command and the British Expeditionary
Force in 1914 (Westport, CT: Praeger, 2003).
29. Foley, ‘A Case Study in Horizontal Military Innovation’, pp. 799–827.

Chapter 6
1. Afflerbach, Falkenhayn, pp. 211–32; Foley, German Strategy and the Path to Verdun,
pp. 113–24; Rupprecht, Kriegstagebuch Volume I, pp. 306, 409; entries 2 March,
25 December 1915, Rupprecht MS Diary, GHA NL KPR, 702, pp. 634–4, 703,
1244–5.
2. Rupprecht, Kriegstagebuch Volume I, p. 265.
3. Rupprecht, Kriegstagebuch Volume I, pp. 253–4; entries, 29 November,
11 December 1914,War Diary, BKA AOK 6 Bund 1 Vorl No. 1, pp. 86, 91; entry,
25 November 1914, Rupprecht MS Diary, GHA NL KPR, 701, p. 463.
4. Goya, La Chair et L’Acier, pp. 255–6; BOH 1915 Volume I, p. 4; GOH Volume
VI, pp. 375–6; Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege, p. 146.
5. BOH 1915 Volume I, pp. 17–19; entries, 14, 16 December 1914, War Diary, BKA
AOK 6 Bd 1 Vorl No. 1, pp. 92–4; Rupprecht, Kriegstagebuch Volume I, pp. 276–8.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

300 Not e s to Page s 60 – 67

6. Casualties in the four French divisions involved were 542 officers and 7,229
men. Doughty, Pyrrhic Victory, p. 127; FOH Volume II, pp. 171–9, 188–9.
7. Rupprecht, Kriegstagebuch Volume I, pp. 278–9; BOH 1915 Volume I, pp. 20–2.
8. FOH Volume II, pp. 190–242, 410–81; entry 25 December 1914, Krafft diary,
BKA NL Krafft 289, p. 900; entries 25, 26 December 1914, Ruprecht MS diary,
GHA NL KPR 701, pp. 529–31; entry, 28 December 1914, War Diary, BKA
AOK 6, Bd 1 Vorl No. 1 Kriegstagebuch, p. 98; Sheldon, The German Army on
Vimy Ridge 1914–1917, pp. 37–8.
9. In August–November 1914, Sixth Army lost, on average, 37,630 men per month
in combat. In December 1914 this fell to 15,799. GOH Volume VII, p. 56;
Heeres-Sanitätsinspektion des Reichskriegsministeriums, Sanitätsbericht Volume
II, pp. 342, 365, 396.
10. BOH 1915 Volume I, pp. 69–154; G. C. Wynne, If Germany Attacks: The Battle in
Depth in the West (Westport, CT: Greenwood Press, 1976 [1940]), pp. 19–41;
GOH Volume VII, pp. 58–9; Robin Prior and Trevor Wilson, Command on the
Western Front:The Military Career of Sir Henry Rawlinson (Barnsley: Pen & Sword
2004 [1992]), pp. 19–73.
11. Prior and Wilson, Command on the Western Front, p. 23; BOH 1915 Volume I, p. 151.
12. Prior and Wilson, Command on the Western Front, pp. 23, 30–4; BOH 1915
Volume I, pp. 81–4, 91; Wynne, If Germany Attacks, pp. 22–4.
13. Quoted in Prior and Wilson, Command on the Western Front, p. 56; Sheffield, The
Chief, p. 110.
14. Prior and Wilson, Command on the Western Front, pp. 59, 61; BOH 1915 Volume
I, p. 128; Rupprecht, Kriegstagebuch Volume I, p. 310.
15. Entry 11 March 1915, Rupprecht MS diary, GHA NL KPR 702, p. 651.
16. Entry 11 March 1915, Rupprecht, Kriegstagebuch Volume I, pp. 310–12: see
Krafft’s comment, footnote, p. 312.
17. Entry 13 March 1915, Rupprecht, Kriegstagebuch Volume I, p. 314.
18. BOH 1915 Volume I, p. 149; Prior and Wilson, Command on the Western Front,
pp. 77–80, 84–5; Paul Harris and Sanders Marble,‘The “Step-by-Step”Approach:
British Military Thought and Operational Method on the Western Front, 1915–
1917’, War in History 15:1 (January 2008), pp. 17–42.
19. Mark Osborne Humphries and John Maker (eds), Germany’s Western Front:
Translations from the German Official History of the Great War Volume II: 1915
(Waterloo, Ontario: Wilfred Laurier University Press, 2010), pp. 384–7.
20. AOK 6 ‘Weisung für die nachsten Maßnahmen der Armeekorps’, 29 November
1914; Rupprecht, Kriegstagebuch Volume III, pp. 58–63; GOH Volume VII,
pp. 17–19; Humphries and Maker (eds), Germany’s Western Front Volume II: 1915,
pp. 22–4; Rupprecht, Kriegstagebuch Volume I, p. 286.
21. Entry 13 March 1915, Rupprecht MS diary, GHA NL KPR 702, p. 654;
Rupprecht, Kriegstagebuch Volume I, pp. 312–13; entry, 14 March 1915 War Diary,
BKA AOK 6 Vorl. 2 Bd 1 Akt. 2, n.p.
22. Entry 16 March 1915,War Diary, BKA AOK 6 Vorl. 2 Bd 1 Akt. 2, n.p.; Rupprecht,
Kriegstagebuch Volume I, p. 321 and Volume III, pp. 68–9; entry, 25 March 1915,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 68– 73301

Rupprecht MS diary, GHA NL KPR 702, p. 678; BOH 1915 Volume 1, p. 151;
GOH Volume VII, p. 59; entry 26 March, Krafft Diary, BKA NL Krafft 290, p. 1362.

Chapter 7
1. Humphries and Maker (eds), Germany’s Western Front Volume II: 1915, pp. 85–8.
2. Humphries and Maker (eds), Germany’s Western Front Volume II: 1915, pp. 89–94;
see the 1930 report drawn up for the Reichsarchiv by Dr Wilhelm Solger, ‘Zur
Ausarbeitung des Reichsarchivs “Die Erwägungen über einen Durchbruchs-
Angriff gegen die westliche Stellungsfront von ende Februar bis 13. April
1915” ’ and especially Beilage 2, ‘Auszug der Stellen aus meinem Tagebuch über
meinem Durchbruchs-Entwurf (u.a.) vom 2.3.1915’, ‘Erwägungen für einen
Durchbruch im Westen’, BKA NL Krafft 189.
3. Beilage 2,‘Auszug der Stellen aus meinem Tagebuch über meinem Durchbruchs-
Entwurf (u.a.) vom 2.3.1915’, pp. 4–5, ‘Erwägungen für einen Durchbruch im
Westen’, BKA NL Krafft 189; entries 20, 26 November 1914, BKA NL Krafft 289,
pp. 761, 784, and BKA NL Krafft 151, pp. 408–9; Rupprecht, Kriegstagebuch Volume
I, p. 261; entry 18 March 1915, Rupprecht MS diary, GHA NL KPR 702, p. 667.
4. GOH Volume VIII, pp. 35–49; BOH 1915 Volume I, pp. 171–359. The German
army lost about 35,000 men, the British nearly 60,000. German estimates that
the French lost 18,000 on the first day alone seem exaggerated: only two divi-
sions were directly affected.
5. Afflerbach, ‘Kronprinz Rupprecht von Bayern’, footnote 11, p. 24; Rupprecht,
Kriegstagebuch I, pp. 305, 327; Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–
18’, GHA NL KPR 716, pp. 33–4; entry 1 March 1915, Rupprecht MS diary,
GHA NL KPR 701, pp. 636–8.
6. Rupprecht, Kriegstagebuch Volume I, pp. 333–6.
7. Rupprecht, Kriegstagebuch Volume I, pp. 326–7, 329; GOH Volume VIII, p. 57.
8. Rupprecht, Kriegstagebuch Volume I, pp. 336–7; entry, 9 May 1915, Xylander
diary, BKA NL R. Xylander 12 Volume I, p. 301.
9. Adrian Bristow, A Serious Disappointment: The Battle of Aubers Ridge and the
Subsequent Munitions Scandal (London: Leo Cooper, 2005), p. 80; Prior and
Wilson, Command on the Western Front, p. 90; BOH 1915 Volume II, pp. 7–43.
10. BEF losses were finally calculated at 11,500. German casualties were probably
fewer than 2,500. Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege, pp.
231–2; GOH Volume VIII, pp. 58–9; BOH 1915 Volume II, pp. 7–43 (casualty
statistics on p. 39); Bristow, A Serious Disappointment, passim and p. 172; entries,
10, 21 May 1915, Rupprecht MS diary, GHA NL KPR 702, pp. 759–60, 814;
Harris, Douglas Haig, p. 141.
11. Harris, Douglas Haig, pp. 133–7; 142–4; BOH Volume II, p. 41; Prior and Wilson,
Command on the Western Front, pp. 83–8, 91–3; Holmes, The Little Field Marshal,
pp. 285–94; Sheffield, Forgotten Victory, p. 115.
12. Jonathan Krause, Early Trench Tactics in the French Army:The Second Battle of Artois,
May–June 1915 (Farnham: Ashgate, 2013), pp. 70, 19.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

302 Not e s to Page s 74–80

13. French maps marked the spot as Côte (Hill) 140. The British called it Hill 145.
Today a stone memorial to the Moroccans stands near the huge Canadian
monument.
14. Krause, Early Trench Tactics, pp. 70–3, 5–6; FOH Volume III, p. 43.
15. Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege, pp. 227–8; Krause, Early
Trench Tactics, pp. 70–2, 151–6; Sheldon, The German Army on Vimy Ridge, footnote
3, pp. 42–3; I Bayerische Reservekorps (BRK) Ia No. 3735.g. ‘Erfahrungen aus
den September-Kämpfen, verglichen mit dem Erfahrungen der 3. Armee in der
Champagne’, 22 January 1916, After-action Reports, BKA AOK 6 Vorl. 164, p. 3;
entry 10 May 1915, Rupprecht MS diary, GHA NL KPR 702, pp. 759–62;
Rupprecht, Kriegstagebuch Volume III, pp. 70–1.
16. Rupprecht, Kriegstagebuch Volume I, pp. 336–41.
17. Krause, Early Trench Tactics, pp. 77–9.
18. Krause, Early Trench Tactics, pp. 79, 105, 108; Rupprecht, Kriegstagebuch Volume I,
pp. 345–7; Sheldon, The German Army on Vimy Ridge, p. 57; GOH Volume VIII,
p. 66; Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege, pp. 229–30. By the
time the 5th Bavarian Reserve Division was finally pulled out of the line at
Vimy during the night of 15/16 May, it had lost 184 officers and 6,600 men.
19. GOH Volume VIII, pp. 56, 64, 68, 72–3.
20. Entries, 13, 15 May 1915, Rupprecht MS diary, GHA NL KPR 702, pp. 789–90;
entry, 12 May, Krafft diary, BKA NL Krafft 290, pp. 1615–16; entries, 13, 15 May,
Xylander diary, BKA NL R. Xylander 12, pp. 308–9; Rupprecht, Kriegstagebuch
Volume I, pp. 348–9, 354, 356–8.
21. Rupprecht’s letter to the Kaiser, his reply, and Falkenhayn’s apology to
Rupprecht: see GHA NL KPR 495; Rupprecht, Kriegstagebuch Volume I,
pp.  361–2; entry 19 May 1915, Krafft diary, BKA NL Krafft 290, pp. 1682–6;
entry 19 May 1915, Rupprecht MS diary, GHA NL KPR 702, p. 808.
22. GOH Volume VIII, pp. 71–2; Wynne, If Germany Attacks, pp. 60–2; BOH 1915
Volume II, pp. 45–79. VII Corps lost 86 officers and 4,700 men in the period
9–20 May, but a proportion of these must have been lost on 9 May, when it
suffered 1,100 wounded alone: entry 25 May 1915,War Diary, BKA AOK 6 Vorl.
2 Bd 1 Akt. 2, n.p.; entry 21 May, Rupprecht MS diary, GHA NL KPR 702,
p. 814; Prior and Wilson, Command on the Western Front, pp. 94–9.
23. FOH Volume III, p. 58; Krause, Early Trench Tactics, pp. 106–9.
24. GOH Volume VIII, p. 198; Rupprecht, Kriegstagebuch Volume I, pp. 364–5;
entries 4, 7, 11, and 15 June 1915, Rupprecht MS diary, GHA NL KPR 702,
pp. 846, 853–4, 862, 870–1.
25. BOH 1915 Volume II, pp. 98–102; FOH III, pp. 99–100; Krause, Early Trench
Tactics, pp. 132, 134–7.
26. Known to the Germans as the Gießler Heights and to the British, later, as The
Pimple.
27. GOH Volume VIII, pp. 86–90; Rupprecht, Kriegstagebuch Volume I, pp. 366–8;
entry 22 June 1915, GHA NL KPR 702, pp. 894–7; FOH III, pp. 71, 74–97;
Krause, Early Trench Tactics, pp. 134–40.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 80 –87303

28. In all, French Tenth Army lost 102,500 men between 9 May and 18 June. Of
those, 16,703 were killed, 20,635 missing, and 65,162 wounded. British losses in
the same period were some 32,000. Rupprecht’s Sixth Army suffered casualties
of 1,560 officers and 71,512 other ranks. FOH III, p. 100; BOH 1915 Volume 2,
pp. 39, 76, 107; Humphries and Maker (eds), Germany’s Western Front Volume II:
1915, p. 210.
29. Foch to GPN, n.d. but 3 February 1915, Annexe 792, FOH II:2 Annexe Volume
2, pp. 24–9; GQG 923, ‘Note pour les armées’, 2 January 1915, Annexe 530,
FOH II:2 Annexe Volume 1, pp. 746–8.
30. GQG Note 5779, ‘But et conditions d’une action offensive d’ensemble’,
16 April 1915, Annexe 52, FOH III Annexe Volume 1, pp. 94–108.
31. GQG 8192, ‘Note sur les premiers enseignments à tirer des combats récents’,
20 May 1915, Annexe 297; GQG 10379, ‘Rectificatif à la note du 16 Avril 1915
sur les conditions d’une offensive d’ensemble’, 26 May 1915, Annexe 368; GQG
7148, ‘Annexe à la note du 16 Avril 1915 sur les buts et les conditions d’une
offensive d’ensemble’, 18 June 1915, Annexe 658; GQG 11286, Joffre to Foch 27
June 1915, Annexe 769: all in FOH III Annexe Volume 1, pp. 399–401, 490–1,
835–7, 992–4; GAN 1379/8, Foch to Joffre, 1 July 1915, Annexe 818, FOH III
Annexe Volume 2, pp. 5–8; Krause, Early Trench Tactics, pp. 5–9, 140–1.
32. BOH 1915 Volume II, pp. vi–ix, 10–15, 77–9, 92–7, 393–9.
33. AOK 6 Ia J. No. 9690, 16 March 1915, and AOK 6 Ia No. 17105, ‘Erfahrungen
aus dem letzten Kämpfen’, 28 June 1915; CGS M.J. No. 1740, ‘Erfahrungen die
sich im Laufe des Krieges für den Ausbau unserer Stellungen ergeben haben’,
29 May 1915: all in BKA AOK 6 Vorl. 164; Humphries and Maker (eds),
Germany’s Western Front Volume II: 1915, pp. 215–16.

Chapter 8
1. See Chapter 5.
2. ‘My Job to Hold Joffre’s Hosts, Says Rupprecht’, New York Times, 29 June 1915,
pp. 1–2.
3. Rupprecht, Kriegstagebuch Volume I, p. 375; entry 22 August 1915, Rupprecht
MS diary, GHA NL KPR 703, p. 997.
4. Rupprecht, Kriegstagebuch Volume I, pp. 375–8; entries 28 August–14 September,
Rupprecht MS diary, GHA NL KPR 703, pp. 1005–30; GOH Volume IX,
pp. 21–2, 26–9.
5. BOH Volume II, pp. 135–8, 149–50; Prior and Wilson, Command on the Western
Front, pp. 112–13.
6. Entry 22 September 1915, War Diary, BKA AOK 6 Vorl. 2 Bd 1 Akt. 2, n.p.;
GOH Volume IX, pp. 29, 44–6.
7. GOH Volume IX, pp. 53–60; BOH Volume II, pp. 171–267; FOH III, pp. 433–42;
Rupprecht, Kriegstagebuch Volume I, pp. 382–5.
8. Report of General Kunze, quoted in entry 1 November 1915, Rupprecht MS diary,
GHA NL KPR 703, p. 1158; BOH Volume II, pp. 139–40, 308–335, 339, 341–5.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

304 Not e s to Page s 87– 94

9. FOH III, pp. 442–50; Rupprecht, Kriegstagebuch Volume I, p. 387; GOH Volume
IX, p. 74.
10. FOH III, pp. 457–60.
11. Entries 28–9 September 1915, Rupprecht MS Diary, GHA NL KPR 703,
pp. 1077, 1082; entry 29 September 1915, Xylander diary, BKA NL R. Xylander 12,
Volume II, p. 424; Rupprecht, Kriegstagebuch Volume I, p. 388; Rupprecht,‘Meine
Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 38.
12. Rupprecht, Kriegstagebuch Volume I, pp. 390.
13. Entry 12 October 1915, GHA NL KPR 703, p. 1116.
14. Nick Lloyd, Loos 1915 (Stroud: History Press, 2008 [2006]), pp. 192–201.
15. BOH 1915 Volume II, p. 388; Lloyd, Loos, pp. 203–12.
16. The French lost 48,230 men in Artois, relatively few compared with the 143,567
casualties in Champagne. The British lost 50,380 men and the Germans 51,100:
FOH III, p. 540; BOH 1915 Volume II, pp. 391–3; GOH Volume IX, pp. 89,
106–7.
17. GQG 11239, ‘Instruction sur l’emploi de l’artillerie lourde’, 20 November 1915;
GQG 2481, ‘Instruction sur le combat offensive des petites unités’, 8 January
1916; GQG 9585, ‘Instruction du 16 janvier 1916 visant le but et les conditions
d’une action offensive d’ensemble’, 16 January 1916; GQG 15350, ‘Instruction
sur le combat offensif des grandes unités’, 26 January 1916: all in FOH IV:1,
pp. 44–51; GQG 2641,‘Note pour les généraux commandants de GA’, 5 December
1915, FOH III, pp. 560–3.
18. AOK 6 Ia No. 26525, ‘Erfahrungen aus den letzten Kämpfen’, 7 October 1915
and AOK 6 Ia No. 27516, ‘Erfahrungen aus den letzten Kämpfen’, 18 October
1915, both in BKA AOK 6 Vorl. 164; AOK 6 Ib No. 58385, ‘Die Herbstschlacht
bei La Bassée und Arras vom 25. September bis 13. Oktober 1915’, 30 November
1915, Operations, GHA NL KPR 483; Humphries and Maker (eds), Germany’s
Western Front Volume II: 1915, pp. 331–7; GOH Volume IX, pp. 99–107.
19. Rupprecht, Kriegstagebuch Volume I, p. 394.
20. Entries 1, 11 December 1915, Xylander diary, BKA NL R. Xylander 12,Volume II,
pp. 472, 479.

Chapter 9
1. Groß, Mythos und Wirlichkeit, p. 123.
2. Foley, German Strategy and the Path to Verdun, pp. 178–208; Afflerbach, Falkenhayn,
pp. 351–75, 543–5; entry 11 February 1916, Kuhl diary, Bundesarchiv-Militärarchiv
(BAMA) RH 61/970, pp. 7–8; Rupprecht, Kriegstagebuch Volume I, p. 427; entry,
12 February 1916, Rupprecht MS diary, GHA NL KPR 703, pp. 1348–9.
3. Entries, 25, 27 December 1915, 10 January 1916, Rupprecht MS diary, GHA NL
KPR 703, pp. 1244–5, 1247, 1269; Rupprecht, Kriegstagebuch Volume I, pp. 398,
409, 412.
4. Rupprecht, Kriegstagebuch Volume I, p. 427; entry, 12 February 1916, Rupprecht
MS diary, GHA NL KPR 703, pp. 1348–9.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 94–101305

5. Foley, German Strategy and the Path to Verdun, pp. 197–200; AOK 6 No. 41494,
‘Der Durchbruch’, 26 February 1916, BKA AOK 6 Vorl. 417; Cron, Imperial
German Army 1914–18, p. 145; entry 8 March, Rupprecht MS diary, GHA NL
KPR 703, pp. 1397–9.
6. GOH Volume X, p. 271; Foley, German Strategy and the Path to Verdun, pp. 204–5,
217; entries 13 January, 2, 5 February 1916, Rupprecht MS diary, GHA NL KPR
703, pp. 1281, 1330, 1338; Rupprecht, Kriegstagebuch Volume I, p. 420.
7. Entries 28 February, 1 March 1916, Rupprecht, Kriegstagebuch Volume I, pp. 432–4.
8. Entry 27 March 1916, Kuhl diary, BAMA RH 61/970, p. 10; entry 29 April 1916,
Rupprecht MS diary, GHA NL KPR 704, p. 1515; Rupprecht, Kriegstagebuch
Volume I, pp. 438–9.
9. Rupprecht, Kriegstagebuch Volume I, pp. 459, 477; GOH Volume X, pp. 271–2, 275.
10. Rupprecht, Kriegstagebuch Volume I, pp. 415, 418, 424, 428, 431, 433–4, 454–6,
463, 465 and Volume III pp. 79–82, 84–7.
11. Rupprecht, Kriegstagebuch Volume I, p. 413; entries 9 January, 8 March, 20 June
1916, Rupprecht MS diary, GHA NL KPR 704, pp. 1268, 1400, 1681.
12. FOH IV:1, pp. 532–6.
13. Peter Chasseaud, ‘Field Survey in the Salient: Cartography and Artillery Survey
in the Flanders Operations in 1917’, in Peter H. Liddle (ed.), Passchendaele in
Perspective: The Third Battle of Ypres (Barnsley: Pen & Sword, 2013 [1997]),
pp. 117–39: 117–21; Greenhalgh, The French Army, pp. 120–6; BOH 1915 Volume
II, p. 19 footnote 2; Philpott, Bloody Victory, pp. 98–9, 113–15; Greenhalgh, Foch in
Command, p. 146; entry, 8 April 1916, Rupprecht MS Diary, GHA NL KPR 704.
14. CGS No. 27956 op., ‘Eininge Erfahrungen aus den Kämpfen im Maasgebiet
(Februar und März 1916)’, 15 May 1916, BKA AOK 6 Vorl. 164; Rupprecht,
Kriegstagebuch Volume I, pp. 467–8.
15. One of the sources was described as Sir Ian Hamilton’s ‘sister’. Hamilton had
only brothers. However, as Lady Hamilton’s biographer Celia Lee has sug-
gested, this may refer to his sister-in-law, Betty Moncrieffe. I am grateful to
John and Celia Lee for discussing this with me; entries 15 May, 8, 14, 19 June
1916, Rupprecht MS diary, GHA NL KPR 704, pp. 1581, 1649, 1667, 1679;
Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 39;
entries 1–22 June 1916, Kuhl diary, BAMA RH 61/970, pp. 14–16.
16. Rupprecht, Kriegstagebuch Volume I, pp. 481, 484–5, 496; GOH Volume X,
pp. 318–20; BOH 1916 Volume I, pp. 316–19; entries 2 and 3 July 1916, Kuhl
diary, BAMA RH 61/970, pp. 17–18; Duffy, Through German Eyes, p. 130.

Chapter 10
1. Sheffield, The Chief, p. 170; Prior and Wilson, Command on the Western Front,
p. 185; BOH 1916 Volume I, p. 483.
2. Goya, La Chair et L’Acier, p. 265; Philpott, Bloody Victory, pp. 197–9; Prior and
Wilson, Command on the Western Front, p. 184; Sheffield, The Chief, pp. 171–2.
3. Duffy, Through German Eyes, pp. 165–9.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

306 Not e s to Page s 102–108

4. Philpott, Bloody Victory, pp. 189–92, 204–8; Duffy, Through German Eyes, pp. 165–9;
BOH 1916 Volume I, pp. vi–ix, 484–92; Travers, The Killing Ground, pp. 161–6;
Prior and Wilson, Command on the Western Front, pp. 171–82; Sheffield, The
Somme, pp. 135–40; Prior and Wilson, The Somme, pp. 112–18; J. P. Harris, Douglas
Haig and the First World War, pp. 234–7.
5. GOH Volume X, pp. 382–3.
6. Jim Beach, Haig’s Intelligence: GHQ and the German Army, 1916–18 (Cambridge:
Cambridge University Press, 2013), pp. 199–203; Philpott, Bloody Victory,
pp. 219–43: p. 236; Prior and Wilson, Command on the Western Front, pp. 185–202:
p. 189.
7. Duffy, Through German Eyes, p. 173.
8. Rupprecht had sent seven divisions to Verdun by the end of June. Rupprecht,
Kriegstagebuch Volume I, pp. 499–501, 503; Philpott, Bloody Victory, p. 227;
Reichsarchiv, GOH Volume X, pp. 355, 409–10, 407, 275.
9. Rupprecht, Kriegstagebuch Volume I, pp. 493–5, 503; Afflerbach, Falkenhayn,
pp. 419–20; entries 3, 4 July 1916, Rupprecht MS diary, GHA NL KPR 704,
pp. 1732–7.
10. Afflerbach, Falkenhayn, pp. 424–36; Foley, German Strategy and the Path to Verdun,
pp. 256–8; entry 18 July 1916, Kuhl diary, BAMA RH 61/970, p. 19.
11. Rupprecht, Kriegstagebuch Volume I, pp. 494, 497–9, 501, 508; Rudolf von
Valentini, Kaiser und Kabinettschef (Oldenburg: Gerhard Stalling, 1931), pp. 234–7;
entry 9 July 1916, Rupprecht MS diary, GHA NL KPR 704, p. 1753.
12. Afflerbach, Falkenhayn, pp. 426–38; Afflerbach, Kaiser Wilhelm II. als Oberster
Kriegsherr im Ersten Weltkrieg: Quellen aus der militärischen Umbegung des Kaisers
1914–1918 (Munich: R. Oldenbourg, 2005), p. 49.
13. Afflerbach, Falkenhayn, pp. 436–44.
14. Rupprecht, Kriegstagebuch Volume I, pp. 520–1, 523–4; GOH Volume X,
pp. 422–3.
15. Afflerbach, Falkenhayn, pp. 447–50.
16. Entry 29 August 1916, Kuhl diary, BAMA RH 61/970, p. 22.
17. 6 September 1916, Letter to King Ludwig III, GHA NL KPR 428.

Chapter 11
1. On the role of the staff, see David T. Zabecki (ed.), Chief of Staff: The Principal
Officers behind History’s Great Commanders Volume I: Napoleonic Wars to World War
I (Annapolis, MD: Naval Institute Press, 2008); Thomas Müller, Konrad Krafft
von Dellmensingen (1862–1953): Porträt eines bayerischen Offiziers (Munich: Kommission
für bayerische Landesgeschichte, 2002), pp. 308–9; Foley, ‘A Case Study in
Horizontal Military Innovation’, pp. 816–17; Foley, German Strategy and the Path
to Verdun, p. 85; Groß, Mythos und Wirklichkeit, p. 116. More generally on the skill
and flexibility of the German army, see Dupuy, A Genius for War; Lupfer, The
Dynamics of Doctrine; Samuels, Command or Control? The most recent and
comprehensive treatment of Auftragstaktik is Marco Sigg, Der Unterführer als
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 109 –114307

Feldherr im Taschenformat: Theorie und Praxis der Auftragstaktik im deutschen Heer

1869 bis 1945 (Paderborn: Ferdinand Schöningh, 2014).
2. Letter to author from Rupprecht’s grandson, Herzog Franz von Bayern, 3 July
2015; Sendtner, Rupprecht von Wittelsbach, p. 129; entry, 26 October 1914, Krafft
Diary, BKA NL Krafft 48, p. 124a; Harris, Douglas Haig and the First World War,
p. 196; Gary Sheffield, The Chief, p. 136; Xylander Diary Volume II, BKA NL R.
Xylander 12, pp. 455–6.
3. Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, p. 74;
Sven Hedin, With the German Armies in the West (H. G. de Walterstorff, trans.)
(London: John Lane, The Bodley Head, 1915), pp. 344–7.
4. Kuhl, Der deutsche Generalstab, p. 186.
5. Rupprecht, Kriegstagebuch Volume I, pp. 52–3; AOK 6 Order of Battle, 6 August
1914, Miscellaneous Orders, GHA NL KPR 476; HG KPR ‘Vorläufige
Stellenbesetzung’, 19 September 1916, and ‘Stellenbesetzung’, 5 September
1918, both in Operations, GHA NL KPR 601; Cowan, ‘Genius for War?’, p. 187;
Administrative Staff Miscellaneous Orders, BKA HG KPR 375. Today known
as Chateau Gendebien, Chateau Hardenpont is an elegant mansion which has
kept its military connection: it is the residence of the NATO Supreme Allied
Commander Europe.
6. Administrative Staff Miscellaneous Orders, BKA HG KPR 375; entry, 26 January
1915, R. Xylander Diary Volume I, BKA NL R. Xylander 12, p. 212.
7. Entry 18 February 1915, Rupprecht MS Diary, GHA NL KPR 701, pp. 621–2;
Krafft Diary, BKA NL Krafft 290, pp. 1522–4, 1557–60.
8. Rupprecht, Kriegstagebuch Volume II, p. 53; entry 27 October 1916, Rupprecht
MS Diary, GHA NL KPR 705, pp. 2050–1, 2364–5.Within months, the Germans
blew up Coucy-le-Chateau during their retreat to the Hindenburg Line. HG
KPR IIb No. 12785 27 March 1917, BKA HG KPR 481; entries, 5 August, 24
December 1915, R. Xylander War Diary, BKA NL R. Xylander 12, Volume I, pp.
373–4 and Volume II, p. 488; letter, Rupprecht to Ludwig III, 27 December
1915, Correspondence with King, GHA NL KPR 428.
9. Müller, Konrad Krafft von Dellmensingen, pp. 308–9, 356–9; entry, 5 August 1914,
Krafft Diary, BKA NL Krafft 145, p. 9; entry 18 November 1914, Krafft Diary,
BKA NL Krafft 289, p. 753 and 48, p. 92b; entry, 27 September, Krafft Diary,
BKA NL Krafft 48, pp. 33a, 71a; Krafft, ‘Kurzer Überblick über die Vorgänge
beim Oberkommando der 6. Armee im August 1914 bis zur Schlacht in
Lothringen am 20.8.1914’, 25 February 1916, Miscellaneous Orders, GHA NL
KPR 476, p. 24; Pöhlmann, Kriegsgeschichte und Geschichtspolitik, pp. 284–321;
Rupprecht, Kriegstagebuch Volume I, pp. 8–10, 20–3, 37–40, 80–1, 103–7, 169,
210–11; entries 28 August, 5 September, 14 October 1914, Rupprecht MS Diary,
GHA NL KPR 699, pp. 110–11, 115, 172 and NL KPR 700, p. 308; entries, 14, 15
January 1915, Xylander Diary Volume I, BKA NL R. Xylander 12, pp. 196–7.
10. Rupprecht, Kriegstagebuch Volume I, p. 362; entries, 25 May, 8, 23, 25, 30 June, 3,
27 August, 29 September, 16 November 1915, Xylander Diary Volume I, BKA
NL R. Xylander 12, pp. 319, 330, 344, 346, 350, 373, 383, 424, 461–2; entries,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

308 Not e s to Page s 115–121

24 June, 4 August, 19 October, 14, 16, 19, 22 November 1915, Rupprecht MS

Diary, GHA NL KPR 702, p. 903 and 703, pp. 976, 1133–4, 1180–1, 1183, 1190,
1194; letter, Rupprecht to Nagel, 22 November 1915, Correspondence with
Bavarian Military Representative at GHQ, GHA NL KPR 494.
11. Entry, 28 November 1915, Kuhl Diary, BA-MA RH 61/970, p. 4; entry,
1 December 1915, Xylander Diary Volume II, BKA NL R. Xylander 12, p. 472.
12. GOH Volume IV, pp. 533–7 and Volume I, pp. 187, 257–8;‘Aufmarschanweisung
für Oberkommando der 6. Armee’, 2 August 1914, Miscellaneous Orders, GHA
NL KPR 476, paragraph 21; Mombauer, Helmuth von Moltke, pp. 254–5; Herwig,
The Marne, 1914, pp. 171–2; Strachan, The First World War Volume I: To Arms, pp.
233–4; entry, 18 September 1914, Krafft Diary, BKA NL Krafft 48, p. 1a;
Rupprecht, Kriegstagebuch Volume I, pp. 6–21, 37–61; Krafft, Die Führung des
Kronprinzen Rupprecht von Bayern, pp. 17–21.
13. Rupprecht, Kriegstagebuch Volume I, pp. 126–7, 206, 232–6, 321, 348–52, 361–2,
369; entries 18, 23 September, 14 October 1914, Krafft Diary, BKA NL Krafft
48, pp 1a–2a, 15b–16b, 81b–82b, 84a–87a; entry, 5 March 1915, Krafft Diary
Appendix 2, BKA NL Krafft 189; Foley, German Strategy and the Path to Verdun,
pp. 99–102, 113–24; entries, 26 October 1914, 2 March, 15, 19 May, 15, 22, 25
June 1915, Rupprecht MS Diary, GHA NL KPR 701, pp. 356–7 and 702, pp.
643–4, 789–91, 808, 870–1, 894–7, 903–5; Afflerbach, Falkenhayn pp. 325–36.
Note, however, that in his published diary Rupprecht seems to have played up
the extent of his friction with Falkenhayn in an effort to distance himself from
him. I am grateful to Holger Afflerbach for pointing this out: Afflerbach,
‘Kronprinz Rupprecht’, footnote 11, p. 24. Rupprecht, ‘Meine Teilnahme am
Weltkrieg 1914–18’, GHA NL KPR 716, p. 33; letter, Minister Leopold Krafft
von Dellmensingen to Privy Councillor Otto von Dandl, 25 May 1915, Dandl-
Krafft Correspondence, GHA NL KPR 436.
14. Rupprecht, Kriegstagebuch Volume I, p. 521, Volume II, pp. 11–13, 15–16 and
Volume III, pp. 108–11; entry, 26 August 1916, Rupprecht MS Diary, GHA NL
KPR 704, pp. 1895–6; GOH Volume X, pp. 422–3 and Volume XI, pp. 106–9;
Rupprecht,‘Meine Teilnahme am Weltkrieg 1914-18’, GHA NL KPR 716, pp. 41–2.
15. Entry, 28 August 1914, Rupprecht MS Diary, GHA NL KPR 699, pp. 110–11;
Gebsattel, Von Nancy bis zum Camp des Romains 1914, pp. 19–36, 59, 61–2.

Chapter 12
1. Heeresgruppe Kronprinz Rupprecht (HG KPR) Ic ‘Liste des Einsatzes und der
Verluste der Divisionen bei 1. und 2. Armee von Beginn der Somme-Schlacht
bis 30.11.1916’, n.d., GHA NL KPR 588; Rupprecht, Kriegstagebuch Volume II,
pp. 4–6; entry 29 August 1916, Rupprecht MS diary, GHA NL KPR 704,
pp. 1908; GOH Volume X, p. 384; GOH Volume XI, pp. 41–2.
2. One train could carry 26,800 rounds for the 77-mm field artillery cannon or
6,000 for heavy howitzers. GOH Volume X, pp. 360–1 and p. 219 footnote;
GOH Volume XI, p. 16, especially footnote; Humphries and Maker (eds),
Germany’s Western Front Volume II: 1915, pp. 150–3.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 122–133309

3. Rupprecht, Kriegstagebuch Volume II, pp. 3–8 and Volume III, pp. 94–6; entries
26, 29, 31 August 1916, Rupprecht MS diary, GHA NL KPR 704, pp. 1895–6,
1908, 1913; Duffy, Through German Eyes, p. 208.
4. Rupprecht, Kriegstagebuch Volume II, pp. 7–11.
5. Rupprecht, Kriegstagebuch Volume II, pp. 9–11; BOH 1916 Volume II, pp. 250–87;
Philpott, Bloody Victory, pp. 347–9.
6. Rupprecht, Kriegstagebuch Volume II, pp. 9–13; Philpott, Bloody Victory, pp. 349–54;
GOH Volume XI, pp. 59–62.
7. Rupprecht, Kriegstagebuch Volume II, pp. 6–7, 12, 18.
8. Rupprecht, Kriegstagebuch Volume II, pp. 14–16; entry 8 September 1916,
Rupprecht MS diary, GHA NL KPR 704, p. 1940.
9. Watson, Ring of Steel, pp. 323–5; Rupprecht, Kriegstagebuch Volume II, pp. 13, 15, 17.
10. Greenhalgh, Victory though Coalition, p. 65; Philpott, Bloody Victory, pp. 354–5,
357–9.
11. Philpott, Bloody Victory, pp. 353–7; Rupprecht, Kriegstagebuch Volume II,
pp. 18–28; entry 19 September 1916, Rupprecht MS diary, GHA NL KPR 704,
p. 1970; GOH Volume XI, pp. 65–6, 73–4.
12. See discussion of this in Philpott, Bloody Victory, pp. 361–3, 368–9.
13. GOH Volume XI, pp. 68–70.
14. Rupprecht, Kriegstagebuch Volume II, pp. 21–2; GOH Volume XI, pp. 68–70;
BOH Volume II, p. 354.
15. Rupprecht, Kriegstagebuch Volume II, p. 25.
16. Rupprecht, Kriegstagebuch Volume II, pp. 23–5 and Volume III, pp. 102, 104–5.
17. Rupprecht, Kriegstagebuch Volume II, pp. 24–6; entry, 23 September 1916,
Rupprecht MS diary, GHA NL KPR 704, p. 1982; entries, 14, 21, 28 September,
4, 21, 26, 28 October 1916, Kuhl Diary, BAMA RH 61/970, pp. 24–8, 31, 34.
18. James S. Corum, ‘Air War Over the Somme: The Evolution of British and
German Air Operations’, in Matthias Strohn (ed.), The Battle of the Somme
(Oxford: Osprey, 2016), pp. 75–91; GOH Volume XI, p. 73; Rupprecht,
Kriegstagebuch Volume II, pp. 27–9; entry 30 September 1916, Rupprecht MS
diary, GHA NL KPR 705, p. 1997.
19. Rupprecht, Kriegstagebuch Volume II, p. 33; GOH Volume XI, pp. 75–7.
20. Duffy, Through German Eyes, pp. 236, 239–42; BOH 1916 Volume II, p. 354;
Rupprecht, Kriegstagebuch Volume III, p. 111; Philpott, Bloody Victory, p. 377;
Robin Prior and Trevor Wilson, The Somme (New Haven, CT:Yale University
Press, 2005), pp. 259, 262–3; Christopher Pugsley, ‘Trial and Error: The
Dominion Forces on the Somme in 1916’, in Strohn (ed.), The Battle of the
Somme, pp. 113–29.
21. Rupprecht, Kriegstagebuch Volume II, pp. 34, 37; GOH Volume XI, p. 77; Jim
Beach, Haig’s Intelligence, pp. 214–15.
22. GOH Volume XI, pp. 77–9; Rupprecht, Kriegstagebuch Volume II, pp. 34, 36,
39–40; entry 30 September 1916, Rupprecht MS diary, GHA NL KPR 705,
p. 1996; HG KPR Ic No. 694, Erfahrungen, 1 October 1916, GHA NL KPR 606.
23. HG KPR Ia No. 609, ‘Betrifft die Lage an der Somme’, 27 September 1916,
reproduced in Rupprecht, Kriegstagebuch Volume III, pp. 106–11.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

310 Not e s to Page s 134–14 2

Chapter 13
1. See the useful collection of essays on British, French, and German tactics and
operations in Matthias Strohn (ed.), The Battle of the Somme (Oxford: Osprey,
2016).
2. GOH Volume XI, pp. 62–3; Erich Ludendorff, My War Memories: 1914–1918
Volume I (London: Hutchinson, 1919), pp. 265–74.
3. Rupprecht, Kriegstagebuch Volume II, p. 18.
4. Rupprecht, Kriegstagebuch Volume II, p. 26; Volume III, pp. 103–4 and HG KPR
Ia No. 609, ‘Betrifft die Lage an der Somme’, 27 September 1916, reproduced
pp. 106–11.
5. Philpott, Bloody Victory, pp. 374–5.
6. Philpott, Bloody Victory, pp. 386–92; BOH 1916 Volume II, pp. 457–8; Ian
Malcolm Brown, British Logistics on the Western Front 1914–1919 (Westport, CT:
Praeger, 1998), pp. 109–38.
7. Rupprecht, Kriegstagebuch Volume II, pp. 42, 44, 61.
8. Rupprecht, Kriegstagebuch Volume II, pp. 39–40; entries 2, 4 October 1916, Kuhl
Diary, BAMA RH 61/970, pp. 27–8.
9. Rupprecht, Kriegstagebuch Volume II, pp. 37, 40, 42, 44, 61; entry 8 October 1916,
Kuhl Diary, BAMA RH 61/970, p. 28; GOH Volume XI, pp. 80–3; Philpott,
Bloody Victory, pp. 395–400.
10. Entry, 13 October 1916, Kuhl Diary BAMA RH 61/970, p. 30; GOH Volume
XI, p. 84.
11. BOH 1916 Volume II, pp. 440–3; Prior and Wilson, Command on the Western
Front, pp. 252–4, 256–60; Gary Sheffield, The Somme (London: Cassell, 2004
[2003]), p. 142.
12. GOH Volume XI, pp. 87, 92; entries 30 October, 10 November 1916, Rupprecht,
Kriegstagebuch Volume II, pp. 54, 63, 95, 111.
13. Rupprecht, Kriegstagebuch Volume II, pp. 44, 47, 49, 53–4, 56, 58, 62.
14. Prior and Wilson, The Somme, pp. 296–7.
15. BOH 1916 Volume II, pp. xii–xvii; GOH Volume XI, p. 186; Sheffield, The
Somme, pp. 151–2 and The Chief, p. 432, footnote 92; Prior and Wilson, The
Somme, pp. 300–1; James McRandle and James Quirk,‘The Blood Test Revisited:
A New Look at German Casualty Counts in World War I’, Journal of Military
History 70 (2006), pp. 667–702; Philpott, Bloody Victory, pp. 600–3; Watson, Ring
of Steel, p. 324.
16. J. H. Boraston, (ed.), Sir Douglas Haig’s Despatches (December 1915–April 1919)
(London: J. M. Dent and Sons, 1919).
17. Rupprecht, Kriegstagebuch Volume I, p. 438.
18. Philpott, Bloody Victory, p. 447; Watson, Ring of Steel, p. 323.
19. Sheffield, The Chief, pp. 165–9, 194–8; Rupprecht, Kriegstagebuch Volume II, p. 34;
letter to King Ludwig III, 25 October 1916, Correspondence, GHA Munich
NL KPR 428; Alexander Watson, Enduring the Great War: Combat, Morale and
Collapse in the German and British Armies, 1914–1918 (Cambridge: Cambridge
University Press, 2008), pp. 165–8.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 143–150311

20. Rupprecht, Kriegstagebuch Volume I, p. 493 and Volume II, p. 63; IV Armeekorps
(AK) Ic No. 185 31 July 1916, Operations File Lessons Learnt Reports, August-
October 1916, BKA NL KPR 216; 56 Infanteriedivision (ID) Ia No. 2497,
15 September 1916, Operations File Lessons Learnt Reports, August–October
1916, BKA NL KPR 216; Hauptmann Grimm, im (sachs.) Inf. Rgt. No. 133
‘Bericht auf befehl der OHL gem. B.K.O. 416 I v. 12.9.’ 17 September 1916,
BKA AOK 6 Vorl. 164 Lessons Learnt 1914–1917.
21. Philpott, Bloody Victory, pp. 454–63; Rupprecht, KriegstagebuchVolume II, pp. 69–70.

Chapter 14
1. GOH Volume XII, p. 1; Watson, Ring of Steel, pp. 378–84; Rupprecht,
Kriegstagebuch Volume II, pp. 76, 91.
2. Rupprecht, Kriegstagebuch Volume I, pp. 453–4, 463 and Volume II, pp. 81–2,
104–5; Afflerbach, ‘Kronprinz Rupprecht’, pp. 43–4; Afflerbach, Kaiser Wilhelm
II. als Oberster Kriegsherr, pp. 23–5; GOH Volume XII, pp. 4–9, 52–9; Anthony
Cowan, ‘Genius for War? German Operational Command on the Western
Front in Early 1917’ (unpublished PhD thesis, King’s College London, 2016),
chapter 8.
3. GOH Volume XII, pp. 26–32; GOH Volume X, pp. 382–9.
4. GOH Volume XI, pp. 107–16; GOH Volume XII, pp. 32–7; CGS Ia/II No. 175,
‘Erfahrungen der Somme-Schlacht and Lehren’, 25 September 1916, Lessons
Learnt, BKA HG KPR 216; HG KPR Ia No. 609, ‘Betrifft die Lage an der
Somme’, 27 September 1916: Rupprecht, Kriegstagebuch Volume III, pp. 106–11;
Lessons Learnt, BKA HG KPR 215–17; Foley, ‘Learning War’s Lessons’,
pp. 471–504.
5. Max Bauer, Der große Krieg in Feld und Heimat (Tübingen: Osiander’sche
Buchhandlung, 1922), pp. 118–19; GOH Volume XII, pp. 38–51.
6. I am grateful to Dr Tony Cowan for his help with these publications. GOH
Volume XII, pp. 50–1; AOK1 Ia No. 2122, ‘Erfahrungen der 1. Armee in der
Sommeschlacht’, 30 January 1917, Operations, GHA NL KPR 581; Foley,
‘Learning War’s Lessons’, pp. 502–3.
7. Rupprecht, Kriegstagebuch Volume II, p. 100; GOH Volume XII, pp. 34, 52–60;
HG KPR Ic 10103, 19 February 1917, Lessons Learnt, BKA HG KPR 220.
8. Rupprecht, Kriegstagebuch Volume II, pp. 76–8, 83–6; HG KPR Ia No. 2026
‘Vorschlag für die Operationen auf dem französischen Kriegsschauplatz im
Früjahr 1917’, 15 January 1917, Operations: Alberich, October 1916–March 1917,
GHA NL KPR 586; Kuhl Diary, BAMA RH 61/970, p. 39.
9. Rupprecht, Kriegstagebuch Volume III, p. 126 and Volume I, p. 70. Also HG KPR
Id No. 1496, 2 November 1916, Operations: Siegfried Stellung, October 1916–
February 1917, GHA NL KPR 585; Rupprecht, Kriegstagebuch Volume II, p. 70.
10. Kuhl Diary, BAMA RH 61/970, p. 39; Rupprecht, Kriegstagebuch Volume II,
pp. 83–4.
11. Kuhl Diary, BAMA RH 61/970, pp. 40–6; Rupprecht, Kriegstagebuch Volume II,
pp. 13, 17, 22, 30, 88–94 and Volume III, pp. 133–4; entry, 4 February 1917,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

312 Not e s to Page s 151–155

Rupprecht MS Diary, GHA NL KPR 705, p. 2248; Kuhl Diary, BAMA RH

61/970, pp. 35–6, 45; HG KPR Id ‘Vergleich zwischen jetziger Front und
Siegfried’, 16 February 1917, and HG KPR to OHL Iad No. 2104 ‘Ausnützung
der großen rückw. Stellungen im Frühjahr 1917: die augenblickliche Lage’,
28 January 1917, both in Operations: Siegfried Stellung, October 1916–February
1917, GHA NL KPR 585; Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’,
GHA NL KPR 716, pp. 45–6.
12. Rupprecht, KriegstagebuchVolume II, pp. 22, 33, 46, 98, 115–16; entries, 17 October
1916, 5 March 1917, Rupprecht MS Diary, GHA NL KPR 705, pp. 2031–2, 2312;
Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914-18’, GHA NL KPR 716, pp.
45–6; GOH Volume XII, pp. 121–33.
13. BOH 1917 Volume I, p. 93.
14. Rupprecht, Kriegstagebuch Volume II, pp. 105, 110–18, 122–5; GOH Volume XII,
pp. 133–45; BOH 1917 Volume I, pp. 89–94, 129, 169–70; Beach, Haig’s Intelligence,
pp. 224–32.
15. GOH Volume XII, pp. 145–6; Beach, Haig’s Intelligence, p. 233; BOH 1917
Volume I, pp. 171–2.
16. GOH Volume XII, pp. 81–7, 184–203; Rupprecht, Kriegstagebuch Volume II,
pp. 103, 125–31.
17. GOH Volume XII, p. 34.
18. Goya, La Chair et L’Acier, pp. 261–9; Greenhalgh, The French Army and the First
World War, pp. 162–9, 176–7; FOH Volume V:1, pp. 161–3, 175–6; GQG
‘Instruction visant le but et les conditions d’une action offensive d’ensemble’
summarized in GQG 24141, ‘Note pour les groupes d’armées, les armées
d’attaque et autres armées’, 28 December 1916, Annexe 358, in Annexes Volume
I (1932), pp. 587–8.
19. IV Armee-Korps Ic No. 185, ‘Beurteilung der englischen Truppen während der
Kämpfe an der Somme’, 31 July 1916, Lessons Learnt, BKA HG KPR 216;
Paddy Griffith, Battle Tactics of the Western Front, pp. 55–60, 76–9; John Lee, ‘Some
Lessons of the Somme:The British Infantry in 1917’, in Brian Bond (ed.), ‘Look
to Your Front’: Studies in the First World War by The British Commission for Military
History (Staplehurst: Spellmount, 1999), pp. 79–88; Jim Beach, ‘Issued by the
General Staff: Doctrine Writing at British GHQ, 1917–18’, War in History 19:4
(2012), pp. 464–91.
20. Simon Robbins, British Generalship on the Western Front 1914–18: Defeat into
Victory (London: Routledge, 2005), pp. 94–7; Peter Simkins, From the Somme to
Victory, pp. 46–9; Lee, ‘Some Lessons of the Somme’, pp. 80–6.
21. John Lee, ‘Some Lessons of the Somme’, pp. 86–7; SS 135, ‘Instructions for the
Training of Divisions for Offensive Action’, IWM EPH 1553, pp. 9–10.
22. Jonathan Boff, Winning and Losing on the Western Front: The British Third Army
and the Defeat of Germany in 1918 (Cambridge: Cambridge University Press,
2012), pp. 59–63.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 156 –164313

Chapter 15
1. BOH 1917 Volume I, pp. 180–6, 201, 302–6, 312–16.
2. GOH Volume XII, pp. 210–12.
3. GOH Volume XII, pp. 212–21; Rupprecht, Kriegstagebuch Volume II, pp. 135–6;
HGr KPR to OHL, Ia No. 2857, Preliminary Report on 9 April, dated 21 April
1917, Operations File Arras April 1917 GHA NL KPR 587; BOH 1917 Volume
I, sketch 11 (facing p. 241); Sheldon, The German Army on Vimy Ridge, p. 302.
4. Rupprecht, Kriegstagebuch Volume II, p. 136.
5. Rupprecht, Kriegstagebuch Volume II, pp. 138–41; GOH Volume XII, pp. 224–32;
BOH 1917 Volume I, pp. 244–81, 340–56.
6. BOH 1917 Volume I, pp. 201, 273–6; GOH Volume XII, p. 234.
7. Extract from letter, General von Hartz (Bavarian Military Representative at
GHQ), 15 April 1917, GHA NL KPR 587; letters, Councillor Leopold Krafft
von Dellmensingen to Prime Minister Otto von Dandl, 14, 19, 24 April 1917,
Correspondence, GHA NL KPR 436; Rupprecht, Kriegstagebuch Volume II,
pp. 140–1, 149.
8. HG KPR Iac No. 2774 to OHL, 12 April 1917; HG KPR Ia No. 2857 to OHL,
21 April 1917; I BRK No. 15290, 12 April; 1st Bavarian Reserve Division (BRD)
I No. 2710, 12 April; 1st Bavarian Reserve Infantry Brigade (BRIB) No. 7752, 13
April; 17th Reserve Division (RD) I No. 910, 14 April; 14th Bavarian Infantry
Division (BID) No. 5340/Ia, 15 April 1917; all in BKA HG KPR 221; see Cowan,
‘Genius for War?’, pp. 217–27, for a less negative view of the German after-
action process at Arras. I am grateful to Dr Cowan for directing my attention
to this passage.
9. BOH 1917 Volume I, pp. 239–41, 552–3.
10. GOH Volume XII, pp. 235–9; HG KPR Ia No. 2583, ‘Erfahrungen und
Folgerungen aus den Kämpfen bei Arras am 9.4.17’, 21 April 1917, BKA HG
KPR 221; Franz Behrmann, Die Osterschlacht bei Arras 1917 Part I: Zwischen Lens
und Scarpe (Gerhard Stalling: Oldenburg, 1929), pp. 95–6, 112, 121.
11. Gary Sheffield, Command and Morale:The British Army on the Western Front, 1914–
1918 (Barnsley: Pen & Sword, 2014), chapter 7.
12. BOH 1917 Volume I, p. 280.
13. BOH 1917 Volume I, pp. 282–4, 378; Harris, Douglas Haig and the First World
War, pp. 314–18; Sheffield, The Chief, pp. 216–18.
14. Greenhalgh, The French Army and the First World War, pp. 175–99; Rupprecht,
Kriegstagebuch Volume II, pp. 145, 147; Harris, Douglas Haig and the First World
War, pp. 319–21.
15. Rupprecht, Kriegstagebuch Volume II, pp. 144, 148, 150–1, 154; BOH 1917 Volume I,
pp. 378–402, 412–26; GOH Volume XII, pp. 248–53; HG KPR Ic No. 2910,
28 April 1917, BKA HG KPR 221.
16. Rupprecht, Kriegstagebuch Volume II, pp. 159–62, 164; GOH Volume XII,
pp. 257–60; BOH 1917 Volume I, pp. 427–54.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

314 Not e s to Page s 165–173

17. In April and May 1917, the British suffered nearly 158,660 casualties, 29,505 of
them dead with another 20,876 missing. German losses were 85,000 at Arras
versus 158,000 on the Aisne and in Champagne, where French casualties may
have been as high as 271,000. OAD 258, GHQ to army commanders, 2 January
1917, BOH 1917 Volume I, Appendix 6 and pp. 556–61, 535; GOH Volume XII,
pp. 277–8, 403 and Appendix 27; Cowan, ‘Genius for War?’, p. 245.
18. Fayolle quoted in William Philpott, War of Attrition: Fighting the First World War
(New York: Overlook Press, 2014), p. 290; BOH 1917 Volume I, pp. 294–7,
544–52.
19. Beach, Haig’s Intelligence, pp. 232–8; AOK 6 CoS No. 730, 13 May 1917; HG
KPR Ic No. 3023, 13 May 1917: both in BKA HG KPR 222.
20. HG KPR 1c No. 15818, 3 May 1917, BKA HG KPR 222.
21. CGS I No. 54448, 4 May 1917; OHL Ia No. 3124, 7 May 1917; HG KPR Ic 16261,
8 May 1917; AOK 2 report for CGS 11 May 1917; all in BKA HG KPR 222.
22. AOK 6 Ia/Ie No. 9000, Armeebefehl, 1 May 1917, Lessons Learnt, BKA HG
KPR 222.
23. AOK 6 Ia/Ie No. 9000, 1 May 1917, BKA HG KPR 222; entry, 12 May 1917,
Rupprecht MS Diary GHA NL KPR 706, p. 2494; Rupprecht, Kriegstagebuch
Volume II, p. 164.

Chapter 16
1. BOH 1917 Volume I, pp. 428–30; Rupprecht, Kriegstagebuch Volume II, p. 165.
2. GOH Volume XII, pp. 429–36, 438–42; Rupprecht, Kriegstagebuch Volume II,
pp. 167, 169, 172–4, 177–8, 182–4; Kuhl Diary, BA-MA RH 61/970, pp. 65–6,
67–9.
3. GOH Volume XII, pp. 425–7, 448–52; BOH 1917 Volume, pp. 41, 44–9;
Rupprecht MS Diary, GHA NL KPR 706, pp. 2604–5.
4. GOH Volume XII, pp. 453–4; Rupprecht MS Diary, GHA NL KPR 706,
pp. 2604–5; 35 ID Ia No. 1756 12 June 1917, Lessons Learnt, BKA HG KPR 235;
Rupprecht, Kriegstagebuch Volume III, pp. 165–70.
5. GOH Volume XII, pp. 454–61; BOH 1917 Volume II, pp. 32–95; Rupprecht,
Kriegstagebuch Volume III, pp. 165–70; 35 ID Ia No. 1756, 12 June 1917, Lessons
Learnt, BKA HG KPR 235;Wynne, If Germany Attacks, pp. 272–81; Robin Prior
and Trevor Wilson, Passchendaele: The Untold Story (New Haven, CT: Yale
University Press, 1996), pp. 61–6.
6. Rupprecht, Kriegstagebuch Volume II, pp. 187–95 and Volume III, pp. 163–5;
Rupprecht MS Diary, GHA NL KPR 706, pp. 2588–9, 2592, 2604–5; Kuhl
Diary, BA-MA RH 61/970, pp. 70–1.
7. GOH Volume XII, p. 471; BOH 1917 Volume II, p. 87; Rupprecht, Kriegstagebuch
Volume II, pp. 191–3.
8. Kuhl Diary, BA-MA RH 61/970, pp. 71–4; entries, 25 January, 12 June, 6 August,
3 September 1917, Rupprecht MS Diary, GHA NL KPR 705, p. 2221 and 706,
pp. 2610, 2782, 2893; Rupprecht, Kriegstagebuch Volume II, p. 257.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 173–180315

9. Rupprecht, Kriegstagebuch Volume II, pp. 198, 204–9, 213–20, 229–30; entries 19,
30 July, Rupprecht MS Diary, GHA NL KPR 706, pp. 2743, 2760; GOH Volume
XII, Beilage 27 and Volume XIII, pp. 41, 56–7, Beilage 28b; BOH 1917 Volume
II, p. 134; Prior and Wilson, Passchendaele, pp. 71–3.
10. BOH 1917 Volume II, p. 138 footnote 2; GOH Volume XIII, pp. 60–3.
11. Rupprecht, Kriegstagebuch Volume II, pp. 231–2.
12. Prior and Wilson, Passchendaele, pp. 86–96; GOH Volume XIII, pp. 64–5, 96,
Beilage 28b; BOH 1917 Volume II, pp. 164–79.
13. Rupprecht, Kriegstagebuch Volume II, pp. 232, 211; Kuhl Diary, BA-MA RH
61/970, p. 84.
14. Entry, 2 August 1917, Rupprecht MS Diary, GHA NL PR 706, p. 2770; Fusilier
Regiment 73 B No. 4733, 9 August 1917; 111th ID I No. 1149, 11 August 1917;
Lehr Infantry Regiment (no number or date); 3rd Guards Division (GD) I No.
7142, 9 August 1917: all in Lessons Learnt, BKA HG KPR 236; OHL Liaison
Officer with AOK 4 to HG KPR, 6 August 1917, Lessons Learnt, BKA HG
KPR 225.
15. Rupprecht, Kriegstagebuch Volume II, pp. 235–6; entry, 4 August 1917, Kuhl
Diary, BA-MA RH 61/970, p. 84.
16. BOH 1917 Volume II, pp. 208–10; Rupprecht, Kriegstagebuch Volume II, pp. 244–9,
257, 259–60; 26th ID Ia No. 7074, 6 September 1917, Lessons Learnt, BKA HG
KPR 236; AOK 4, 29 August 1917, Lessons Learnt, BKA HG KPR 244; Strength
and Casualty Returns, BKA HG KPR 438; entries, 21 August and 5 September
1917, War Diary, BKA HG KPR 8; GOH Volume XIII, Beilage 28b and
pp. 70–1; entry, 20 August 1917, Kuhl Diary, BA-MA RH 61/970, p. 86.
17. Rupprecht, Kriegstagebuch Volume II, pp. 259–62; GOH Volume XIII, p. 71.

Chapter 17
1. BOH 1917 Volume II, pp. 237–44, 250–79; HG KPR Ic No. 31500, 30 October
1917, Lessons Learnt, BKA HG KPR 217; 2nd GRD Ia 281/9, 24 September
1917 and 9th RD, 23 September 1917, both in Lessons Learnt BKA HG KPR
244; Prior and Wilson, Passchendaele, pp. 113–23; GOH Volume XIII, pp. 74–6.
2. GOH Volume XIII, p. 76; BOH 1917 Volume II, pp. 280–95.
3. GOH Volume XIII, pp. 77–8, Appendix 28b.
4. The Germans lost 38,500 men against 36,000 for the British. GOH Volume
XIII, p. 76; Prior and Wilson, Passchendaele, pp. 125–6, 131.
5. Rupprecht, Kriegstagebuch Volume II, pp. 264–5 and Volume III, pp. 173–4.
6. Entries, 26, 27, 29 September, 1, 2 October 1917, Rupprecht MS Diary, GHA
NL KPR 707, pp. 2979–80, 2983–6, 2989, 2997; entries 30 September, 3 October
1917 Kuhl Diary, BA-MA RH 61/970, p. 90; OHL Iac No. 3996, 28 September
1917, War Diary, BKA HG KPR 8; AOK4 Ia/g No. 791/Sept., 30 September
1917, Lessons Learnt, BKA HG KPR 244.
7. GOH Volume XIII, pp. 78–80; 4th GD report in Rupprecht, Kriegstagebuch
Volume III, pp. 179–81; Rupprecht MS Diary, GHA NL KPR 707, pp. 3001,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

316 Not e s to Page s 181–187

3003–5, 3010; BOH 1917 Volume II, pp. 296–322; Prior and Wilson, Passchendaele,
pp. 133–9; OHL Ia No. 4071, 7 October 1917,War Diary, BKA HG KPR 8; AOK
4 Ia/g 276.Oktober, 7 October 1917, Lessons Learnt, BKA HG KPR 244.
8. GOH Volume XIII, pp. 77–8; Matthias Strohn, The German Army and the
Defence of the Reich: Military Doctrine and the Conduct of the Defensive Battle 1918–39
(Cambridge: Cambridge University Press, 2010).
9. GOH Volume XIII, pp. 77–8, Appendix 28b; Strohn, The German Army and the
Defence of the Reich; 4th GD and 195th ID reports in Rupprecht, Kriegstagebuch
Volume III, pp. 179–81 and 174–6 respectively; see also entry 5 October 1917,
Volume II, pp. 267–8; 6th Bavarian Infantry Division Ia No. 7660, 15 October
1917; 119th Infantry Division Ia No. 2027, n.d.; 233rd Infantry Division Ia
No. 4045, 21 October 1917, all in Lessons Learnt, BKA HG KPR 236; Rupprecht
MS Diary, GHA NL KPR 707, pp. 3001, 3003–4.
10. OHL Iab No. 4153, 14 Oct 1917, War Diary, BKA HG KPR 8; entries, 2, 5, 9
October 1917, Rupprecht MS Diary, GHA NL KPR 707, pp. 2997–8, 3003–4,
3022; Rupprecht, Kriegstagebuch Volume II, pp. 269–70 and Volume III, p. 174;
entries 3, 5 October 1917, Kuhl Diary, BA-MA RH 61/970, p. 90.
11. BOH 1917 Volume II, pp. 323–38; Prior and Wilson, Passchendaele: The Untold
Story, pp. 159–65; AOK 4 Ia/g No. 420/Okt, 11 October 1917, Operations, GHA
NL KPR 589; GOH Volume XIII, pp. 82–3; Rupprecht, Kriegstagebuch Volume
II, p. 270; Strength and Casualty Returns, BKA HG KPR 438; entries, 11 and 13
October 1917, Rupprecht MS Diary, GHA NL KPR 707, pp. 3029, 3032–3;
entries 6, 8, 10 October 1917, Kuhl Diary, BA-MA RH 61/970, pp. 91–2.
12. GOH Volume XIII, pp. 83–5; BOH 1917 Volume II, pp. 339–45; Prior and
Wilson, Passchendaele, pp. 165–9; Strength and Casualty Returns, BKA HG KPR
438; Rupprecht, Kriegstagebuch Volume III, pp. 174–9; 22nd RD I No. 10303,
Lessons Learnt, BKA HG KPR 236.
13. Rupprecht, Kriegstagebuch Volume II, pp. 271–3; entry, 13 October 1917,
Rupprecht MS Diary, GHA NL KPR 707, pp. 3032–3; Albrecht von Thaer
(Siegfried A. Kaehler, ed.), Generalstabsdienst an der Front und in der O.H.L.
(Göttingen: Vandenhoeck & Ruprecht, 1958), pp. 137–9, 142–5.
14. BOH 1917 Volume II, pp. 345–60; Rupprecht, Kriegstagebuch Volume II,
pp. 277–83; entries, 31 October, 7 November 1917, Rupprecht MS Diary, GHA
NL KPR 707, pp. 3127, 3164–5.
15. Rupprecht, Kriegstagebuch Volume II, p. 287; GOH Volume XIII, Appendix 28b.
16. GOH Volume XIII, pp. 95–9; BOH 1917 Volume II, pp. 360–5; Greenhalgh,
The French Army and the First World War, p. 235; Stevenson, With Our Backs to the
Wall, p. 27; Ludendorff, War Memories, Volume II, pp. 491–2; Bayerisches
Kriegsarchiv, Die Bayern im Großen Kriege, p. 410.
17. BOH 1917 Volume II, pp. xii–xv, 366–87.
18. Entry 5 December 1917, War Diary, BKA HG KPR 8; GOH Volume XIII,
pp. 32, 95–9, 146,Appendix 28b; GOH Volume XI, pp. 186–7; Watson, Enduring the
Great War, pp. 164–7, 153–5; Rupprecht, Kriegstagebuch Volume II, pp. 281, 284–8;
entry, 11 November 1917, Rupprecht MS Diary, GHA NL KPR 707, p. 3181.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 188–197317

19. Sheffield, The Chief, pp. 247–8; Harris, Douglas Haig and the First World War,
pp. 381–2; Greenhalgh, The French Army and the First World War, pp. 201–16;
Beach, Haig’s Intelligence, pp. 239–61.
20. Prior and Wilson, Passchendaele, p. 199; entry, 16 November 1917, Rupprecht MS
Diary, GHA NL KPR 707, p. 3194.
21. Prior and Wilson, Passchendaele, pp. 171–81, 200.
22. Stevenson, With Our Backs to the Wall, p. 29.

Chapter 18
1. Rupprecht, Kriegstagebuch Volume II, pp. 288, 290; Rupprecht MS diary, GHA
KPR 707, pp. 3208–9; GOH Volume XIII, pp. 124–6. The best modern account
is Bryn Hammond, Cambrai 1917: The Myth of the First Great Tank Battle (London:
Weidenfeld & Nicolson, 2008).
2. BOH 1917 Volume III, pp. 1–49.
3. Rupprecht, Kriegstagebuch Volume II, pp. 291–2; Rupprecht MS diary, GHA
KPR 707, p. 3211.
4. ‘Victory Peals at St Paul’s’, The Times, 24 November 1917, p. 7; Hammond,
Cambrai, pp. 429–37.
5. HG KPR Ic No. 4561 24 November 1917, Lessons Learnt BKA HG KPR 244;
HG KPR Ic No. 4703, 4 December 1917, Lessons Learnt, BKA HG KPR 234;
entry 22 November 1917, Kuhl Diary, BA-MA RH 61/970, p. 97; Rupprecht,
Kriegstagebuch Volume II, pp. 291–2; Rupprecht MS diary, GHA KPR 707, p. 3211.
6. BOH 1917 Volume III, pp. 120–61; Sheffield, The Chief, pp. 248–57.
7. Rupprecht, Kriegstagebuch Volume II, pp. 295–6; Kuhl Diary, BA-MA RH
61/970, pp. 97–8.
8. Rupprecht, Kriegstagebuch Volume II, pp. 294–8, 301; Rupprecht MS diary, GHA
KPR 707, p. 3250; GOH Volume XIII, pp. 133–42.
9. Rupprecht, Kriegstagebuch Volume III, pp. 192–222; Hammond, Cambrai,
pp. 371–7; BOH 1917 Volume III, pp. 257–61.
10. BOH 1917 Volume III, pp. 273–5; GOH Volume XIII, pp. 143–5; Rupprecht,
Kriegstagebuch Volume II, p. 299.
11. BOH 1917 Volume III, pp. v, 165–6, 168, 294–305; Hammond, Cambrai, pp. 437–40;
Beach, Haig’s Intelligence, pp. 262–72: p. 265.
12. Ludendorff, War Memories Volume II, p. 497; Rupprecht, Kriegstagebuch Volume
II, pp. 294–8; Rupprecht MS Diary, GHA NL KPR 707, p. 3275; Rupprecht,
Kriegstagebuch Volume III, pp. 190, 192–222.
13. BOH 1917 Volume III, p. iii.
14. HG KPR Ia No. 4692, 4 December 1917, Rupprecht, Kriegstagebuch Volume III,
pp. 188–9; HG KPR Ibd No. 4781, 11 December 1917, Lessons Learnt, BKA HG
KPR 245.
15. HG KPR Ia No. 4692, ‘Erfahrungen beim Angriff am 30. November’,
4 December 1917; HG KPR Iad No. 4812,‘Erfahrungen aus den Angriffskämpfen
bei Cambrai für die “Angriffsschlacht” ’, 14 December 1917; ‘1. Fortsetzung zu
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

318 Not e s to Page s 197–207

Erfahrungen aus den Angriffskämpfen bei Cambrai für die “Angriffsschlacht” ’,

18 December 1917; ‘2. Fortsetzung zu Erfahrungen aus den Angriffskämpfen
bei Cambrai für die “Angriffsschlacht” ’, 24 December 1917: all in Rupprecht,
Kriegstagebuch Volume III, pp. 188–9, 192–222.
16. Bruce I. Gudmundsson, Stormtroop Tactics: Innovation in the German Army, 1914–
1918 (New York: Praeger, 1989), pp. 148–51; Samuels, Command or Control?,
pp. 241–5; GOH Volume XIV, pp. 43–50; CGS Ia/II No. 6608, 16 February 1918,
Lessons Learnt, BKA HG KPR 244; Rupprecht, Kriegstagebuch Volume II,
pp.  317–19 and Volume III, pp. 263, 282–3; entries, 18–23 January 1918, War
Diary, BKA HG KPR 9.

Chapter 19
1. Stevenson, With Our Backs to the Wall, pp. 31–4; Zabecki, The German 1918
Offensives, pp. 93–6.
2. Entry, 24 October 1917, Kuhl Diary, BA-MA RH 61/970, pp. 92–3; Rupprecht,
Kriegstagebuch Volume II, pp. 274–5, 278–9; GOH Volume XIV, pp. 50–1.
3. GOH Volume XIV, pp. 53–5; Kuhl Diary, BA-MA RH 61/970, pp. 94–5;
Rupprecht, Kriegstagebuch Volume II, pp. 284–7; Rupprecht MS Diary, GHA
NL KPR 707, p. 3181; Zabecki, The German 1918 Offensives, pp. 97–100.
4. GOH Volume XIV, pp. 55–8; HG KPR Ia 4501, 20 November 1917, Rupprecht,
Kriegstagebuch Volume III, pp. 222–38, also in Operations, GHA NL KPR 591.
5. GOH Volume XIV, pp. 58–67; HG KPR Ia No. 4835, 15 December and HG
KPR Ia No. 4877, 19 December 1917, GHA NL KPR 591; the former is repro-
duced also in Rupprecht, Kriegstagebuch Volume III, pp. 238–45; entry, 19
December 1917, Kuhl Diary, BA-MA RH 61/970, pp. 100–1; Zabecki, The
German 1918 Offensives, pp. 101–7.
6. GOH Volume XIV, pp. 67–9; Rupprecht, Kriegstagebuch Volume II, pp. 304–11;
BOH 1918 Volume I, pp. 95–7; Beach, Haig’s Intelligence, pp. 283–4; entry
4 September 1917, Rupprecht MS Diary, GHA NL KPR 706, p. 2893; entry 29
December 1917, Rupprecht MS Diary, GHA NL KPR 707, pp. 3309–10; Kuhl
Diary, BA-MA RH 61/970, p. 102.
7. Rupprecht, Kriegstagebuch Volume II, pp. 305–9.
8. HG KPR Iaf No. 5002, 3 January 1918, War Diary, BKA HG KPR 8; Georg
Mayer (ed.), Generalfeldmarschall Wilhelm Ritter von Leeb: Tagebuchaufzeichnungen
und Lagebeurteilungen aus zwei Weltkriegen (Stuttgart: Deutsche Verlags-Anstalt,
1976), p. 103; Rupprecht, Kriegstagebuch Volume II, pp. 313–14 and Volume III,
pp. 248–59.
9. GOH Volume XIV, pp. 72-9; Rupprecht, Kriegstagebuch Volume II, pp. 317–24
and Volume III, pp. 263–4; entry, 22 January 1918, Kuhl Diary, BA-MA RH
61/970, pp. 105–7; entry 23 January 1918, War Diary, BKA HG KPR 9;
Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914-18’, GHA NL KPR 716,
pp.  51–2; Ludendorff, War Memories Volume II, pp. 590–2; Zabecki, The
German 1918 Offensives, pp. 101–7.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 208–214319

10. HG KPR Ia No. 5423, 26 January 1918, Operations, GHA NL KPR 591; entry,
3 February 1918,War Diary, BKA HG KPR 9; Rupprecht, Kriegstagebuch Volume
III, pp. 277–8, 282–3; Zabecki, The German 1918 Offensives, pp. 113–16; minutes
of meeting, 21 January 1918, BA-MA PH 5 I/45; Ludendorff, War Memories
Volume II, pp. 590–2.
11. Rupprecht, Kriegstagebuch Volume II, pp. 326, 330–2, 336; Correspondence with
King Ludwig III, GHA NL KPR 428; entries, 6, 8 February 1918, Kuhl Diary,
BA-MA RH 61/970, pp. 109–11.
12. Rupprecht, Kriegstagebuch Volume II, pp. 326, 330–2, 336; Correspondence with
King Ludwig III, GHA NL KPR 428; entry, 6 February 1918, entries 6,
8 February 1918, Kuhl Diary, BA-MA RH 61/970, pp. 109–11; GOH Volume
XIV, pp. 92–3.
13. HG KPR Ia No. 6702 and AOK2 Ia No. 532, both 6 March 1918, Operations,
GHA NL KPR 593; GOH Volume XIV, pp. 82–4; Zabecki, The German 1918
Offensives, pp. 116–18; Rupprecht, Kriegstagebuch Volume II, p. 335.
14. OHL Ia No. 7070, 10 March 1918 and HG KPR Ia No. 6263, 16 March 1918, both
in Operations, GHA NL KPR 593; see also Rupprecht, Kriegstagebuch Volume III,
pp. 299–302 and Volume II, pp. 337–43; GOH Volume XIV, pp. 84–7.
15. Rupprecht, Kriegstagebuch Volume II, pp. 343; Rupprecht, ‘Meine Teilnahme am
Weltkrieg 1914–18’, GHA NL KPR 716, p. 53.

Chapter 20
1. GOH Volume XIV, p. 104.
2. Rupprecht, Kriegstagebuch Volume II, pp. 344–5; GOH Volume XIV, pp. 105–32.
3. BOH 1918 Volume I, pp. 166–217, 240–59; GOH Volume XIV, pp. 105–32;
Zabecki, The German 1918 Offensives, pp. 138–42.
4. BOH 1918 Volume I, pp. 264–326; GOH Volume XIV, pp. 132–47; Zabecki,
The German 1918 Offensives, pp. 142–3.
5. Rupprecht, Kriegstagebuch Volume III, pp. 305–6 and Volume II, pp. 346–50.
6. BOH 1918 Volume I, pp. 327–98; GOH Volume XIV, pp. 147–60, 162–9;
Zabecki, The German 1918 Offensives, pp. 143–5; Rupprecht, Kriegstagebuch
Volume II, pp. 350–1.
7. BOH 1918 Volume I, pp. 399–495; GOH Volume XIV, pp. 169–89; Zabecki,
The German 1918 Offensives, pp. 145–9; Rupprecht, Kriegstagebuch Volume II,
pp. 353–5; HG KPR Ia No. 6427, 25 March 1918, Operations, GHA NL KPR 593.
8. Douglas Haig, War Diaries and Letters 1914–18 (Gary Sheffield and John Bourne,
eds), (London: Phoenix, 2006 [2005]), pp. 392–3; Harris, Douglas Haig and the
First World War, pp. 453–6; Sheffield, The Chief, pp. 272–6; Greenhalgh, The
French Army and the First World War, pp. 274–6; Greenhalgh, ‘Myth and Memory:
Sir Douglas Haig and the Imposition of Allied Unified Command in March
1918’, Journal of Military History 68:2 (2004), pp. 771–820; Zabecki, The German
1918 Offensives, pp. 148–9; BOH 1918 Volume I, pp. 538–44; Stevenson, With Our
Backs to the Wall, pp. 62–5.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

320 Not e s to Page s 215–225

9. Rupprecht, Kriegstagebuch Volume II, pp. 355–60; Zabecki, The German 1918
Offensives, pp. 149–54; BOH 1918 Volume I, pp. 496–536 and Volume II,
pp. 12–41; GOH Volume XIV, pp. 189–202, 206–13; entries 25 and 27 March 1918,
Kuhl Diary, BA-MA RH 61/970, pp. 115–16; HG KPR Ia No. 6438, 26 March
1918, Operations, GHA NL KPR 593.
10. GOH Volume XIV, pp. 206–13; Rupprecht, Kriegstagebuch Volume II, pp. 358–
61; Zabecki, The German 1918 Offensives, pp. 153–5.
11. GOH Volume XIV, pp. 213–29; Rupprecht, Kriegstagebuch Volume II, pp. 360–2;
entry, 6 April 1918, Kuhl Diary, BA-MA RH 61/970, pp. 128–9; HG KPR Ia No.
6468, HG KPR Ia No. 6478 and OHL Ia 7380, all 28 March 1918, Operations,
GHA NL KPR 593.
12. GOH Volume XIV, pp. 224–36; Rupprecht, Kriegstagebuch Volume II, pp. 362–5;
Kuhl Diary, BA-MA RH 61/970, p. 119; Rupprecht MS Diary, GHA NL KPR
707, pp. 3595–6; Mayer (ed.), Generalfeldmarschall Wilhelm Ritter von Leeb, p. 112.
13. GOH Volume XIV, pp. 236–41; Rupprecht, Kriegstagebuch Volume II, pp. 367–9.
14. GOH Volume XIV, pp. 241–54; Zabecki, The German 1918 Offensives, pp. 157–9.
15. GOH Volume XIV, pp. 254–9; BOH 1918 Volume II, pp. 113, 456–93; Zabecki,
The German 1918 Offensives, pp. 160–73.

Chapter 21
1. GOH Volume XIV, pp. 270–1.
2. Rupprecht, Kriegstagebuch Volume II, p. 372; entries 4, 6 April, Kuhl Diary,
BA-MA RH 61/970, pp. 125, 128–9; Georg Mayer (ed.), Generalfeldmarschall
Wilhelm Ritter von Leeb, p. 113; Untersuchungsausschuss der Deutschen
Verfassunggebenden Nationalversammlung und des Deutschen Reichstages
1919–26, Die Ursachen des Deutschen Zusammenbruchs im Jahre 1918, Volume III
(Berlin: Deutsche Verlagsgesellchaft für Politik und Geschichte, 1925), pp. 156–7;
Hermann von Kuhl, Der Weltkrieg 1914–1918 Volume II (Berlin: Tradition
Wilhelm Kolk, 1929), p. 331; GOH Volume XIV, pp. 265–71.
3. Rupprecht, Kriegstagebuch Volume II, pp. 375–9; BOH 1918 Volume II, pp. 156–283,
512; Zabecki, The German 1918 Offensives, pp. 186–90; GOH Volume XIV, pp. 53,
272–80; entry, 6 May 1918, Rupprecht MS Diary, GHA NL KPR 708, p. 3728.
4. Rupprecht, Kriegstagebuch Volume II, pp. 381–7; entry, 15 April 1918, Rupprecht
MS Diary, GHA NL KPR 707, pp. 3662–4; entries 2, 13, 15 April, 11 June 1918,
Kuhl Diary, BA-MA RH 61/970, pp. 124–5, 138–41, 152.
5. Rupprecht, Kriegstagebuch Volume II, pp. 381–7, 393; entry, 15 April, Rupprecht MS
Diary, GHA NL KPR 707, pp. 3662–4; BOH 1918 Volume II, pp. 284–455; GOH
Volume XIV, pp. 280–98, 301–10; Zabecki, The German 1918 Offensives, pp. 190–8.
6. 325,800 German and 328,304 allied soldiers were dead, wounded, or missing.
British casualties were 236,300 and French 92,004. GOH Volume XIV,
pp.  299–301; BOH 1918 Volume II, pp. 488–93; Zabecki, The German 1918
Offensives, pp. 198–205.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 225–232321

7. FOH VI:1, pp. 521–4; BOH 1918 Volume III, pp. 4–5.
8. Zabecki, The German 1918 Offensives, pp. 206–17, 280–7; OHL Ia No. 7960,
2 May 1918, Operations, GHA NL KPR 594.
9. CGS Ia/II No. 7745 ‘Angriffserfahrungen’, 17 April 1918, Operations, GHA
NL KPR 593; 6 BID Ia No. 2500, 24 April 1918, Lessons Learnt, BKA HG KPR
236; HG KPR Iaf No. 7454 ‘Einzelerfahrungen’, 28 May 1918, Operations, GHA
NL KPR 594; Gerhard P. Groß, ‘Das Dogma der Beweglichkeit: Überlegungen
zur Genese der deutschen Heerestaktik im Zeitalter der Weltkriege’, in Bruno
Thoß and Hans-Erich Volkmann (eds), Erster Weltkrieg, Zweiter Weltkrieg: Ein
Vergleich: Krieg, Kriegserlebnis, Kriegserfahrung in Deutschland (Paderborn: Ferdinand
Schöningh, 2002), pp. 151–3; Rupprecht, Kriegstagebuch Volume II, p. 396; Thaer,
Generalstabsdienst, pp. 187–8.
10. Rupprecht, Kriegstagebuch Volume II, pp. 395–8, 401 and Volume III, pp. 318,
321; report, 7 May 1918, Correspondence with King Ludwig III, GHA NL
KPR 428.
11. Rupprecht, Kriegstagebuch Volume II, pp. 399–401; Lothar Machtan, Prinz Max
von Baden: Der letzte Kanzler des Kaisers: Eine Biografie (Berlin: Suhrkamp, 2013),
pp. 347–61; Rupprecht MS Diary, GHA NL KPR 708, p. 3765.
12. Entry, 21 May 1918, Kuhl Diary, BA-MA RH 61/970, p. 152; Rupprecht,
Kriegstagebuch Volume III, p. 321; Rupprecht MS Diary, GHA NL KPR 708,
p. 3760.
13. Zabecki, The German 1918 Offensives, pp. 206–32; Rupprecht, Kriegstagebuch
Volume II, pp. 402–4.
14. Zabecki, The German 1918 Offensives, pp. 287–301; Rupprecht, Kriegstagebuch Volume
III, pp. 327, 331–2; entry, 9 June 1918, Rupprecht MS Diary, GHA NL KPR 708,
p. 3807; entry, 11 June 1918, Kuhl Diary, BA-MA RH 61/970, pp. 158–9.
15. BOH 1918 Volume III, pp. 194–213; Rupprecht, Kriegstagebuch Volume II,
pp. 406, 409, 414, 418.
16. OHL Ia/II No. 9135, 6 July 1918, Operations, GHA NL KPR 596; Rupprecht,
Kriegstagebuch Volume II, pp. 419–20; Heeres-Sanitätsinspektion, Sanitätsbericht
über das Deutsche Heer Volume III, pp. 122–3.
17. Rupprecht, Kriegstagebuch Volume II, pp. 415–17; entry, 2 July 1918, War Diary,
BKA HG KPR 9; H. A. Jones, The War in the Air: Being the Story of the Part Played
in the Great War by the Royal Air Force Volume VI (Oxford: Clarendon, 1937),
pp.  409–10. The raids were suspended after a week, having proved that air
power did not yet possess the accuracy to cut railways by day or night.
18. Zabecki, The German 1918 Offensives, pp. 246–79, 302–7; Michael S. Neiberg,
The Second Battle of the Marne (Bloomington, IN: Indiana University Press,
2008); entry 20 July 1918, Kuhl Diary, BA-MA RH 61/970, pp. 169–70; HG
KPR Ia No. 8528 and No. 8542, 20 July 1918, Operations, GHA NL KPR 596;
OHL Ia 9388, 20 July 1918, BKA HG KPR 105; GOH Volume XIV, pp. 531–5;
Fritz von Loßberg, Meine Tätigkeit im Weltkriege 1914–1918 (Berlin: E. S. Mittler &
Sohn, 1939), p. 352.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

322 Not e s to Page s 233–238

Chapter 22
1. Rupprecht, Kriegstagebuch Volume II, pp. 426–30; entry, 3 August 1918, Rupprecht
MS Diary, GHA NL KPR 708, pp. 3955–6; entry, 4 August 1918, Kuhl Diary,
BA-MA RH 61/970, p. 175.
2. BOH 1918 Volume IV, pp. 16–92; GOH Volume XIV, pp. 549–57; Thilo von
Bose, Die Katastrophe des 8. August 1918 (Berlin: Gerhard Stalling, 1930), p. 196;
entry, 8 August 1918, Kuhl Diary, BA-MA RH 61/970, p. 176; Rupprecht,
Kriegstagebuch Volume III, p. 346; entry, 9 August 1918, Rupprecht MS Diary,
GHA NL KPR 708, p. 3970.
3. Entry, 9 August 1918, Rupprecht MS Diary, GHA NL KPR 708, pp. 3976–7;
entries, 10–11 August 1918, Kuhl Diary, BA-MA RH 61/970, pp. 176–8; BOH
1918 Volume IV, pp. 93–162; GOH Volume XIV, pp. 557–67. Good modern
treatments of Amiens can be found in Charles Messenger, The Day We Won the
War: Turning Point at Amiens, 8 August 1918 (London: Weidenfeld & Nicolson,
2008); Nick Lloyd, Hundred Days: The End of the Great War (London: Viking,
2013), pp. 28–70, and J. P. Harris with Niall Barr, Amiens to the Armistice:The BEF
in the Hundred Days Campaign, 8 August–11 November 1918 (London: Brassey’s,
1998), pp. 59–117.
4. Bose, Die Katastrophe des 8. August 1918, pp. 196–8; Ludendorff, War Memories:
1914–1918 Volume II, pp. 679, 684; Nebelin, Ludendorff, pp. 444–5, 451; Rupprecht,
Kriegstagebuch Volume II, p. 435.
5. Rupprecht, Kriegstagebuch Volume II, p. 438; Rupprecht MS Diary, GHA
NL KPR 708, pp. 3990–4001; Herbert Michaelis and Ernst Schraepler (eds),
Ursachen und Folgen vom deutschen Zusammenbruch 1918 and 1945 bis zur staatli-
chen Neuordnung Deutschlands in der Gegenwart Volume II: Der militärische
Zusammenbruch und das Ende des Kaiserreichs (Berlin: Herbert Wendler & Co.,
n.d.), pp. 283–4.
6. Beach, Haig’s Intelligence, pp. 307–16; entry, 23 August 1918, Kuhl Diary, BA-MA
RH 61/970, pp. 182–4.
7. Entry, 2 September 1918, Rupprecht MS Diary, GHA NL KPR 708, p. 4001;
entries, 27–8 August 1918, Kuhl Diary, BA-MA RH 61/970, pp. 186–7;
Ludendorff, War Memories Volume II, p. 696; Nebelin, Ludendorff, p. 452; report,
4 September 1918, Correspondence with King Ludwig III, GHA NL KPR 428.
8. ‘Zusammenstellung der Gesamtverpflegung-Stärken der Armeen 19 September
1916–30 October 1918’, BKA HG KPR 438; GHQ AG War Diary, TNA WO
95/26; Rupprecht, Kriegstagebuch Volume II, p. 441. For contrasting views on
morale, compare Watson, Enduring the Great War, pp. 184–232 with Boff, Winning
and Losing on the Western Front, pp. 92–122.
9. Rupprecht, Kriegstagebuch Volume II, p. 442; entry, 6 September 1918, Kuhl
Diary, BA-MA RH 61/970, pp. 190–2; Loßberg, Meine Tätigkeit im Weltkriege,
p. 357; 3rd Marine Division Ia No. 806, 6 September 1918, Lessons Learnt, BKA
HG KPR 236.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 239 –245323

10. Boff, Winning and Losing on the Western Front, pp. 45–53, 165–78; OHL II No.
10162, 4 September 1918, BKA 16 BID, Bd 10/7; GOH Volume XIV, pp. 607–8;
entries, 13, 17 September 1918, Rupprecht MS Diary, GHA NL KPR 708,
pp. 4040, 4056; Rupprecht, Kriegstagebuch Volume II, p. 356.
11. Nebelin, Ludendorff, pp. 453–6; entry, 6 September, Kuhl Diary, BA-MA RH
61/970, p. 191; entries, 9, 27 September 1918, Rupprecht MS Diary, GHA NL
KPR 708, pp. 4028–9, 4083; GOH Volume XIV, footnote 1, pp. 594–5.
12. Entry, 9 September 1918, Rupprecht MS Diary, GHA NL KPR 708, pp. 4028–9;
diary entries 11, 13–14, 17–20 September 1918, War Diary, BKA HG KPR 9;
Rupprecht, Kriegstagebuch Volume II, p. 446; entry 14 September 1918, Kuhl
Diary, BA-MA RH 61/970, p. 192; GOH Volume XIV, pp. 606–7.
13. Greenhalgh, The French Army and the First World War, pp. 336–41.
14. Rupprecht, Kriegstagebuch Volume II, p. 452 and Volume III, p. 358; OHL No.
10552, 30 September 1918, BKA HG KPR 99/101; report and letter,
30 September 1918, Correspondence with King Ludwig III, GHA NL KPR
428; Mayer (ed.), Generalfeldmarschall Wilhelm Ritter von Leeb, p. 138; entry, 30
September 1918, Kuhl Diary, BA-MA RH 61/970, pp. 193–4.
15. Ludendorff, War Memories Volume I, pp. 719–22;Thaer, Generalstabsdienst, p. 234;
entry, 5 October 1918, Kuhl Diary, BA-MA RH 61/970, p. 195; Rupprecht,
Kriegstagebuch Volume II, p. 455; Rupprecht MS Diary, GHA NL KPR 708,
pp. 4102–3; Stevenson, With Our Backs to the Wall, pp. 510–5; Watson, Ring of
Steel, pp. 533–5.
16. Rupprecht, Kriegstagebuch Volume II, p. 456; entry, 11 October 1918, War Diary,
BKA HG KPR 5.
17. Heeresgruppe Boehn (HB) Ia No. 1620, 3 October 1918, HB War Diary, BA-MA
PH 5 I/47; entry, 11 October 1918, Kuhl Diary, BA-MA RH 61/970, p. 196;
entry, 11 October 1918, Rupprecht MS Diary, GHA NL KPR 708, p. 4113;
Rupprecht, Kriegstagebuch Volume II, p. 458.
18. Wilhelm Deist, ‘The Military Collapse of the German Empire: The Reality
Behind the Stab-in-the-Back Myth’ (E. J. Feuchtwanger, trans.), War in History
3:2 (April 1996), pp. 186–207; Christoph Jahr, Gewöhnliche Soldaten: Desertion
under Deserteure im deutschen und britischen Heer 1914–1918 (Göttingen:
Vandenhoeck & Ruprecht, 1998), pp. 150–76; Watson, Enduring the Great War,
pp. 186–230; Scott Stephenson, The Final Battle: Soldiers of the Western Front and
the German Revolution of 1918 (Cambridge: Cambridge University Press, 2009),
chapters 1 and 2; Boff, Winning and Losing on the Western Front, pp. 92–122.
19. Stevenson, With Our Backs to the Wall, pp. 519–20;Watson, Ring of Steel, pp. 549–51;
Lloyd, Hundred Days, pp. 234–7; entries, 27–31 October 1918, Kuhl Diary,
BA-MA RH 61/970, pp. 198–9; entry, 27 October 1918, Rupprecht MS Diary,
GHA NL KPR 708, p. 4165.
20. Mayer (ed.), Generalfeldmarschall Wilhelm Ritter von Leeb, p. 149; HKR Ia
No. 10606, 5 November, BKA HKR Bd 99/101.
21. März, Das Haus Wittelsbach im Ersten Weltkrieg, pp. 492–500.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

324 Not e s to Page s 246 –260

22. Mayer (ed.), Generalfeldmarschall Wilhelm Ritter von Leeb, p. 154; Rupprecht,
Kriegstagebuch Volume II, pp. 472–6 and Volume III, pp. 368–71; entries, 2,
5 November 1918, Rupprecht MS Diary, GHA NL KPR 708, pp. 4182–91; entry,
4 November 1918, Rupprecht MS Diary, GHA NL KPR 709, n.p.; Mayer (ed.),
Generalfeldmarschall Wilhelm Ritter von Leeb, p. 154.

Chapter 23
1. Passport for the party of Marquis de Villalobar, Correspondence, GHA NL
KPR 255.
2. Letter to Krafft from the Supreme State Prosecutor, b.J. 353/20/12, 3 March
1923, Miscellaneous Documents regarding Crown Prince Rupprecht, BKA NL
Krafft 196; Horne and Kramer, German Atrocities 1914, pp. 63–70, 329–55;
Heather Jones, Violence against Prisoners of War: Britain, France and Germany, 1914–
20 (Cambridge: Cambridge University Press, 2011), p. 78.
3. Weiß, Kronprinz Rupprecht von Bayern, pp. 223–4.
4. Reiner Pommerin, ‘Die Ausweisung von “Ostjuden” aus Bayern 1923. Ein
Beitrag zum Krisenjahr der Weimarer Republik’, Viertelsjahrhefte für Zeitgeschichte
34:3 (July 1986), pp. 311–40; Weiß, Kronprinz Rupprecht von Bayern, pp. 211–22.
5. Weiß, Kronprinz Rupprecht von Bayern, pp. 211–22; März, Das Haus Wittelsbach,
p. 530.
6. Weiß, Kronprinz Rupprecht von Bayern, pp. 263–72.
7. This school was apparently also well known at the time as good at handling
free-spirited young ladies. I am grateful to Mr Charles Messenger for this
information.

Chapter 24
1. Rupprecht, Kriegstagebuch Volume II, pp. 33, 38; entry, 8 October 1916, Kuhl
Diary, BA-MA RH 61/970, p. 29.
2. Entries 12, 14, 16 April 1917, Kuhl Diary, BA-MA RH 61/970, pp. 55–6; entry
16 April 1917, Rupprecht MS Diary, GHA NL KPR 706, pp. 2425–6; Gebsattel
quoted in Ulrich and Ziemann (eds), German Soldiers in the Great War,
pp. 170–1.
3. Rupprecht, Kriegstagebuch Volume II, pp. 164, 211; entries, 12, 21 May, 2 July 1917,
Rupprecht MS Diary, GHA NL KPR 706, pp. 2494, 2520, 2669; entries, 1–2 July
1917, Kuhl Diary, BA-MA RH 61/970, p. 79.
4. Entries, 27 September 1917, 2, 9 October, Rupprecht MS Diary, GHA NL KPR
707, pp. 2980, 2998–9, 3022; Thaer, Generalstabsdienst, pp. 187–8; Rupprecht,
Kriegstagebuch Volume II, pp. 269–70.
5. Entries, 13–15 April, 9 June, 10, 24–30 August 1918, Kuhl Diary, BA-MA RH
61/970, pp. 138–42, 158, 177, 184–7.
6. Entries, 24 November, 25 December 1915, 21 February 1917, Rupprecht MS Diary,
GHA NL KPR 703, p. 1195 and 705, pp. 2150–2, 2281–2; entries, 24 November, 1,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 260 –265325

2, 11 December 1915, Xylander Diary Volume II, BKA NL R. Xylander 12,

pp. 466, 472–3, 479; Kuhl Diary, BA-MA RH 61/970, pp. 19–20; Rupprecht,‘Meine
Teilnahme am Weltkrieg 1914–18’, GHA NL KPR 716, pp. 38–41; Kuhl, Der
deutsche Generalstab, p. 191; Rupprecht, Kriegstagebuch Volume II, p. 266, 439.
7. See Markus Pöhlmann’s discussion of this in his Kriegsgeschichte und
Geschichtspolitik: Der Erste Weltkrieg: Die amtliche deutsche Militärgeschichtsschreibung
1914–1956 (Paderborn: Ferdinand Schöningh, 2002), pp. 249–51.
8. Entries, 28 June, 1, 4, 6, 7, 14, 15 July, 16, 26 August, 1 September 1917, Kuhl
Diary, BA-MA RH 61/970, pp. 78–87; entry, 7 July 1917, Rupprecht MS diary,
GHA NL KPR 706, p. 2686; Thaer, Generalstabsdienst, p. 126.
9. März, Das Haus Wittelsbach im Ersten Weltkrieg, pp. 56–61.
10. März, Das Haus Wittelsbach im Ersten Weltkrieg, p. 58; Weiß, Kronprinz Rupprecht
von Bayern, p. 34.
11. Weiß, Kronprinz Rupprecht von Bayern, pp. 88–96; Rupprecht, ‘Meine Teilnahme
am Weltkrieg 1914-18’, GHA NL KPR 716, p. 10; Afflerbach, ‘Kronprinz
Rupprecht von Bayern im Ersten Weltkrieg’, p. 29.
12. Rupprecht, Kriegstagebuch Volume I, pp. 356, 462–3, 469, 473, 503–5 and Volume
II, pp. 55, 142; letters, Rupprecht to War Minister General Otto Kreß von
Kreßenstein, 25 May and 3 October 1915, Correspondence with War Minister,
GHA NL KPR 430.
13. Telegram, Kaiser Wilhelm II to Crown Prince Rupprecht, 13 August 1914,
GHA NL KPR 419; Rupprecht, ‘Meine Teilnahme am Weltkrieg 1914–18’,
GHA NL KPR 716, p. 22; entries, 8 September, 21, 22, 25 October 1916, Kuhl
Diary, BA-MA RH 61/970, pp. 24, 33; letter, Counsellor Leopold Krafft von
Dellmensingen to Prime Minister Dandl, 24 October 1916, Correspondence,
GHA NL KPR 436; Rupprecht, Kriegstagebuch Volume II, p. 116; Benjamin
Ziemann, Front und Heimat: Ländliche Kriegserfahrungen im südlichen Bayern 1914–
1923 (Essen: Klartext, 1997), p. 68.
14. Entries 27 September, 19 October 1914, Krafft Diary, BKA NL Krafft 48,
pp. 33a, 100a; entries 9 December 1916, 21 February 1917, Rupprecht MS Diary,
GHA NL KPR 705, pp. 2116–7, 2281–2; Wenninger, letter, 6 September 1914,
quoted in Schulte, ‘Neue Dokumente zu Kriegsausbruch und Kriegsverlauf
1914’, p. 168. For further detail, see Tony Cowan, ‘A Picture of German Unity?
Federal Contingents in the German Army, 1916–17’, in Jonathan Krause, The
Greater War: Other Combatants and Other Fronts (Basingstoke: Palgrave Macmillan,
2014), pp. 141–60.
15. Entries 21 December 1916, 19 January 1917, Rupprecht MS Diary, GHA NL
KPR 705, pp. 2141–2, 2210.
16. Letters from Ludwig III to Rupprecht, 22 November, 7 December 1914,
28 September 1915 and letters from Rupprecht to Ludwig III, 21 July, 2 August
1915: all in Correspondence with King, GHA NL KPR 427 and 428; Rupprecht,
Kriegstagebuch Volume I, pp. 285, 405, 475–6.
17. Rupprecht, Kriegstagebuch Volume II, p. 302: entries 19 December 1917, 12, 16
May 1918, Rupprecht MS diary, GHA NL KPR 707, p. 2191 and 708, pp. 3743,
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

326 Not e s to Page s 266 –272

3748; report to Ludwig III, 31 March 1918, Correspondence with King, GHA
NL KPR 428.
18. Afflerbach, Kaiser Wilhelm II. als Oberster Kriegsherr, pp. 1–91; Röhl, Wilhelm II:
Into the Abyss of War and Exile, 1900–41, p. 1108; Walter Görlitz (ed.), The Kaiser
and his Court: The Diaries, Note Books and Letters of Admiral Georg Alexander von
Müller, Chief of the Naval Cabinet, 1914–1918 (London: Macdonald, 1961), pp. 42,
181; Holger Afflerbach,‘Wilhelm II as Supreme Warlord’, Isabel V. Hull,‘Military
Culture, Wilhelm II, and the End of the Monarchy in the First World War’,
Matthew Stibbe, ‘Germany’s “Last Card”: Wilhelm II and the Decision in
Favour of Unrestricted Submarine Warfare in January 1917’, all in Annika
Mombauer and Wilhelm Deist (eds), The Kaiser: New Research on Wilhelm II’s
Role in Imperial Germany (Cambridge: Cambridge University Press, 2003),
pp. 195–217 (quotation from p. 210), 219–34, 235–58.
19. Afflerbach, Kaiser Wilhelm II. als Oberster Kriegsherr im Ersten Weltkrieg, p. 17;
entries 13, 15 April 1918, Kuhl Diary, BA-MA RH 61/970, pp. 138–41.
20. Voltaire quoted in Archibald Wavell, Generals and Generalship (Harmondsworth:
Penguin, 1941), p. 20; entries, 5 September 1914, 18 March 1915, 10 August 1916,
25 May and 6 June 1917, Rupprecht MS Diary, GHA NL KPR 699, p. 172, NL
KPR 702, p. 667, NL KPR 704, p. 1851; Rupprecht, ‘Meine Teilnahme am
Weltkrieg 1914–18’, GHA NL KPR 716, p. 39.
21. Thomas Ertman, Birth of Leviathan: Building States and Regimes in Medieval and
Early Modern Europe (Cambridge: Cambridge University Press, 1997), p. 317;
Manfred Nebelin, Ludendorff: Diktator im Ersten Weltkrieg (Munich: Siedler,
2010), pp. 453–4; Wilhelm Heye, Lebenserinnerungen des Generaloberst W.
Heye: Teil II: ‘Wie ich den Weltkrieg erlebte’, NL Generaloberst W. Heye,
BA-MA N 18/4, pp. 66–8.

Chapter 25
1. März, Das Haus Wittelsbach im Ersten Weltkrieg, p. 172.
2. Entry, 12 February 1916, Rupprecht MS Diary, GHA NL KPR 703, pp. 1348–9;
Rupprecht, Mein Kriegstagebuch Volume II, pp. 76, 103, 219–29, 305–9, 321–6,
329–32; entry, 22 January 1918, Kuhl Diary, BA-MA RH 61/970, pp. 105–7;
Watson, Ring of Steel, pp. 458–60.
3. Entry, 14 May 1918, Rupprecht MS diary, GHA NL KPR 708, p. 3745;
Rupprecht, Mein Kriegstagebuch Volume II, pp. 399; Karl Hertling, Ein Jahr in
der Reichskanzlei: Erinnerungen an die Kanzlerschaft meines Vaters (Freiburg:
Hedersche Verlagshandlung, 1919), pp. 139–41; Watson, Ring of Steel, pp. 533–5,
547–51; März, Das Haus Wittelsbach, pp. 461–3.
4. Letters from Rupprecht, 15 August, 18 October 1918, Prince Max von Baden
Correspondence, GHA NL KPR 650–1; Letters and Reports from Rupprecht,
25 July, 4, 30 September, 7, 14, 15, 26 October, 1 November 1918, King Ludwig III
Correspondence, GHA NL KPR 428.
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Not e s to Page s 273–284327

5. März, Das Haus Wittelsbach im Ersten Weltkrieg, pp. 461–85; entry, 2 November
1918, Rupprecht MS Diary, GHA NL KPR 708, p. 4182; Letter to Rupprecht,
7 November 1918, King Ludwig III Correspondence, GHA NL KPR 435.
6. Rupprecht, Kriegstagebuch Volume II, pp. 448–9, 473.
7. This occurred in May 1915. The other occasions were October 1914, May and
July 1916, February and June 1917.
8. März, Das Haus Wittelsbach im Ersten Weltkrieg, pp. 401–9; entry, 10 January 1916,
Rupprecht MS diary, GHA NL KPR 703, pp. 1274–6; letter from Rupprecht, 13
July 1917, King Ludwig III Correspondence, GHA NL KPR 428; entry, 13 July
1917, Rupprecht MS diary, GHA NL KPR 706, p. 2707; Weiß, Kronprinz
Rupprecht, pp. 126–34, 143–5; entry, 19 January 1917, Rupprecht MS diary, GHA
NL KPR 705, p. 2210.
9. Ludendorff, War Memories Volume I, p. 275.
10. Entries, 31 May, 11, 13 July, 1917, Rupprecht MS Diary, GHA NL KPR 706,
pp. 2557, 2698, 2707; letter, Leopold Krafft von Dellmensingen to Otto Dandl,
1 June 1917, GHA NL KPR 436; letter from Rupprecht, 13 July 1917, King
Ludwig III Correspondence, GHA NL KPR 428; Machtan, Max von Baden,
pp. 355–66; Weiß, Kronprinz Rupprecht, pp. 148–9; März, Das Haus Wittelsbach,
pp. 373–6; 404–10.
11. März, Das Haus Wittelsbach, pp. 390, 440–8, 456–65; Dieter Storz, ‘ “Aber was
hätte anders geschehen sollen?” Die deutschen Offensiven an der Westfront’, in
Jörg Duppler and Gerhard Groß (eds), Kriegsende 1918: Ereignis, Wirkung,
Nachwirkung, pp. 51–95; Watson, Ring of Steel, pp. 523–37, 547–55.

Chapter 26
1. Jonathan Steinberg argues that Bismarck deliberately arranged power to this
end: Bismarck: A Life (Oxford: Oxford University Press, 2011).
2. Hans von Below memoir, 27 February 1919, Nachlass Otto von Below, BA-MA
N 87/2, p. 14.
3. Rupert Brooke, ‘Peace’ at: http://projects.oucs.ox.ac.uk/jtap/tutorials/intro/
brooke/ipeace.html#peace, accessed 10 February 2017; J. Glenn Gray, The
Warriors: Reflections on Men in Battle (New York: Harper & Row, 1967 [1959]),
pp. 28–52.
4. Three battles of Artois in 1915; the Somme in 1916; Artois, Flanders, and
Cambrai in 1917.
5. Samuel Hynes, The Soldier’s Tale: Bearing Witness to Modern War (London:
Pimlico, 1998), p. 11.
6. W. B.Yeats, The Oxford Book of Modern Verse (Oxford: Oxford University Press,
1936), p. xxxiv.
7. Eksteins, Rites of Spring, p. 184.
8. See Robert Gerwarth, The Vanquished: Why the First World War Failed to End
(London: Allen Lane, 2016).
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Bibliography

Unpublished Primary Material

Geheimes Hausarchiv, Abteilung III, Bayerisches Hauptstaatsarchiv,
Munich (GHA)
Nachlass Kronprinz Rupprecht (NL KPR)
17, 41: correspondence, family
140: correspondence, miscellaneous
164–7: correspondence, Pappenheim
251, 255: correspondence, miscellaneous
393, 416, 427–8, 434–5: correspondence, King Ludwig III
419, 495: correspondence, Kaiser Wilhelm II
430, 614: correspondence, War Minister
436: correspondence, Leopold Krafft von Dellmensingen
with Dandl
476–93, 561, 578–601, 606–10: military operations
494: correspondence, Bavarian military representative
602–3: correspondence, Generals 1917–18
626: correspondence, wartime
639, 647: correspondence, miscellaneous
650–1: correspondence, political
699–709: manuscript diary
716: memoir
723: correspondence, Konrad Krafft von Dellmensingen,
1920–51

Bayerisches Kriegsarchiv, Abteilung IV, Bayerisches Hauptstaatsarchiv,

Munich (BKA)
Armeeoberkommando 6 (AOK 6): war diary, orders, operations and after-action
reports
Vorl. 1–2, 6, 9, 11–12, 24–5, 34, 61, 164, 169, 206, 220–1, 270–1, 417, 419–20, 587, 611,
653, 774
Heeresgruppe Kronprinz Rupprecht (HG KPR)
1–12: war diary and annexes
178, 198, 286, 375, 381, 391–2: miscellaneous
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

330 bi bliogr a ph y

214–63: after-action reports

438–44: strength returns and replacements
477: morale
478–82: miscellaneous
Nachlass Frauenholz
1: memoirs
Nachlass Krafft von Dellmensingen
70–4: notebooks
145–52, 290: diary
183, 186–9, 196, 202, 391: post-1918 correspondence regarding the war
260: letters to wife
288–9, 324: operations
Nachlass R. Xylander
12: diary
16: Lorraine
Bundesarchiv-Militärarchiv, Freiburg-im-Breisgau (BAMA)
N 18/4, 6 Nachlass Generaloberst W. Heye
N 46/129, 157 Nachlass Wilhelm Groener
N 87/2, 65 Nachlass Otto von Below
PH 5 I/42 Heeresgruppe Kronprinz Rupprecht war diary
PH 5 I/47–8 Heeresgruppe Boehn war diary
PH 5 II/124–7 Armeeoberkommando 2 war diary
RH 61/970 Hermann von Kuhl diary

Official Printed Sources

Bayerisches Kriegsarchiv, Die Bayern im Großen Kriege (Munich:Verlag des Bayerischen
Kriegsarchivs, 1923)
Beach, James (intro.), The German Army Handbook of 1918 (Barnsley: Frontline, 2008
[1918])
Behrmann, Franz, Die Osterschlacht bei Arras 1917 Part I: Zwischen Lens und Scarpe
(Oldenburg: Gerhard Stalling, 1929)
Behrmann, Franz, Die Osterschlacht bei Arras 1917 Part II: Zwischen Scarpe und Bullecourt
(Oldenburg: Gerhard Stalling, 1929)
Beumelburg, Werner, Loretto (Oldenburg: Gerhard Stalling, 1927)
Beumelburg, Werner, Flandern 1917 (Oldenburg: Gerhard Stalling, 1928)
Bose, Thilo von, Die Katastrophe des 8. August 1918 (Berlin: Gerhard Stalling, 1930)
Cron, Hermann, Imperial German Army 1914–18: Organisation, Structure, Orders-of-
Battle (C. F. Colton, trans.) (Solihull: Helion, 2002 [original published in German,
1937])
Edmonds, James E. (ed.), Military Operations, France and Belgium, 14 volumes (London:
HMSO and Macmillan & Co., 1922–48)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y331

Gebsattel, Ludwig von, Von Nancy bis zum Camp des Romains 1914 (Oldenburg: Gerhard
Stalling, 1924)
German General Staff, Ypres, 1914:An Official Account Published by Order of the German
General Staff (translation of Die Schlacht an der Yser und bei Ypern im Herbst 1914
(Oldenburg: Gerhard Stalling, 1918) by G. C. W[ynne]) (Nashville, TN: Battery
Press, 1994 [1919])
Heeres-Sanitätsinspektion des Reichskriegsministeriums, Sanitätsbericht über das
Deutsche Heer (Deutsches Feld- und Besatzungheer) im Weltkriege 1914/1918, 3 vol-
umes (Berlin: E. S. Mittler & Sohn, 1934–8)
Humphries, Mark Osborne and Maker, John (eds), Germany’s Western Front:Translations
from the German Official History of the Great War Volume II: 1915 (Waterloo, Ontario:
Wilfred Laurier University Press, 2010)
Raleigh, Sir Walter and Jones, H. A., The War in the Air: Being the Story of the Part
Played in the Great War by the Royal Air Force, 6 volumes (Oxford: Clarendon,
1922–37)
Reichsarchiv, Der Weltkrieg 1914 bis 1918, 14 volumes (Berlin: E. S. Mittler & Sohn,
1925–56)
Service Historique, État-Major de l’Armée, Ministère de la Guerre, Les Armées
Françaises dans la Grande Guerre, 11 volumes (Paris: Imprimerie Nationale,
1922–36)
Stosch, Albrecht von, Somme-Nord, 2 volumes (Oldenburg: Gerhard Stalling, 1927)
Strutz, Georg, Die Tankschlacht bei Cambrai 20–29 November 1917 (Oldenburg: Gerhard
Stalling, 1929)
Untersuchungsausschuss der Deutschen Verfassunggebenden Nationalversammlung
und des Deutschen Reichstages 1919–26, Die Ursachen des Deutschen Zusammenbruchs
im Jahre 1918, Volume III (Berlin: Deutsche Verlagsgesellchaft für Politik und
Geschichte, 1925)
War Office, Statistics of the Military Effort of the British Empire during the Great War
1914–1920 (London: HMSO, 1922)

Primary Published Sources

Balck, W., Entwicklung der Taktik im Weltkriege (Berlin: R. Eisenschmidt, 1922)
Bauer, Max, Der große Krieg in Feld und Heimat (Tübingen: Osiander’sche Buchhandlung,
1922)
Boraston, J. H. (ed.), Sir Douglas Haig’s Despatches (December 1915–April 1919)
(London: J. M. Dent and Sons, 1919)
Brooke, A. F., ‘The Evolution of Artillery in the Great War’, The Journal of the Royal
Artillery Volume LI (1924), pp. 250–68, 359–72; Volume LII (1925–6), pp. 37–51,
369–87;Volume LIII (1927), pp. 76–93, 233–51, 320–39, 469–82
Churchill,Winston S., The World Crisis, 6 volumes (London:Thornton Butterworth,
1923–31)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

332 bi bliogr a ph y

Condell, Bruce and Zabecki, David T. (eds and trans.), On the German Art of War:
Truppenführung (Boulder, CO: Lynne Rienner, 2001)
Deuringer, Karl, Die Schlacht in Lothringen und in des Vogesen 1914: Die Feuertaufe der
Bayerischen Armee, 2 volumes (Munich: Max Schick, 1929)
Foch, Ferdinand, The Memoirs of Marshal Foch (T. Bentley Mott, trans.) (London:
William Heinemann, 1931)
Görlitz, Walter (ed.), The Kaiser and his Court: The Diaries, Note Books and Letters of
Admiral Georg Alexander von Müller, Chief of the Naval Cabinet, 1914–1918 (London:
Macdonald, 1961)
Groener, Wilhelm, Der Felherr wider Willen: Operative Studien über den Weltkrieg
(Berlin: E. S. Mittler & Sohn, 1931)
Haig, Douglas (Gary Sheffield and John Bourne, eds), War Diaries and Letters 1914–18
(London: Phoenix, 2006 [2005])
Hedin, Sven, With the German Armies in the West (H. G. de Walterstorff, trans.)
(London: John Lane, The Bodley Head, 1915)
Hertling, Karl, Ein Jahr in der Reichskanzlei: Erinnerungen an die Kanzlerschaft meines
Vaters (Freiburg: Hedersche Verlagshandlung, 1919)
Hoeppner, Ernst von, Deutschlands Krieg in der Luft (Leipzig: K. F. Koehler, 1921)
Krafft von Dellmensingen, Konrad, Die Führung des Kronprinzen Rupprecht von
Bayern auf dem linken deutschen Heeresflügel bis zur Schlacht in Lothringen im August
1914 (Berlin: E. S. Mittler & Sohn, 1925)
Krafft von Dellmensingen, Konrad, Das Oberkommando in den Reichslanden im
Sommer 1914 (Munich: Max Schick, 1931)
Kuhl, Hermann von, Der deutsche Generalstab in Vorbereitung und Durchführung des
Weltkrieges (Berlin: E. S. Mittler & Sohn, 1920)
Kuhl, Hermann von, Der Weltkrieg 1914–1918, 2 volumes (Berlin: Tradition Wilhelm
Kolk, 1929)
Leeb, Wilhelm von (Georg Mayer, ed.), Generalfeldmarschall Wilhelm Ritter von Leeb:
Tagebuchaufzeichnungen und Lagebeurteilungen aus zwei Weltkriegen (Stuttgart: Deutsche
Verlags-Anstalt, 1976)
Lloyd George, David, War Memoirs, 6 volumes (London: Nicholson & Watson,
1933–6)
Loßberg, Fritz von, Meine Tätigkeit im Weltkriege 1914–1918 (Berlin: E. S. Mittler &
Sohn, 1939)
Ludendorff, Erich, My War Memories: 1914–1918, 2 volumes (London: Hutchinson,
1919)
Ludendorff, Erich, Urkunden der Obersten Heeresleitung über ihre Tätigkeit 1916/18
(Berlin: Ernst Siegfried Mittler und Sohn, 1920)
Marwitz, Georg von der (Erich von Tschischwitz, ed.), General von der Marwitz:
Weltkriegsbriefe (Berlin: Ernst Steiniger, 1940)
Morgen, Curt von, Meiner Truppen Heldenkämpfe (Berlin: Ernst Siegfried Mittler
und Sohn, 1920)
Rupprecht, Kronprinz von Bayern (Eugen von Frauenholz, ed.), Mein Kriegstagebuch,
3 volumes (Berlin: E. S. Mittler & Sohn, 1929)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y333

Schulte, Bernd F., ‘Neue Dokumente zum Kriegsausbruch und Kriegsverlauf 1914’,
Militärgeschichtliche Mitteilungen 25:1 (1979), pp. 123–85
Thaer, Albrecht von (Siegfried A. Kaehler, ed.), Generalstabsdienst an der Front und in
der O.H.L. (Göttingen:Vandenhoeck & Ruprecht, 1958)
William, Crown Prince of Germany, My War Experiences (London: Hurst and
Blackett, n.d.)
Xylander, Rudolf von, Deutsche Führung in Lothringen 1914:Wahrheit und Kriegsgeschichte
(Berlin: Junker und Dünnhaupt Verlag, 1935)

Secondary Articles and Books

Afflerbach, Holger, Falkenhayn, Politisches Denken und Handeln im Kaiserreich (Munich:
R. Oldenbourg, 1994)
Afflerbach, Holger, Kaiser Wilhelm II. als Oberster Kriegsherr im Ersten Weltkrieg: Quellen
aus der militärischen Umbegung des Kaisers 1914–1918 (Munich: R. Oldenbourg, 2005)
Afflerbach, Holger, ‘Kronprinz Rupprecht von Bayern im Ersten Weltkrieg’,
Militärgeschichtliche Zeitschrift 75:1 (May 2016), pp. 21–54
Ashworth, Tony, Trench Warfare 1914–1918: The Live and Let Live System (London:
Macmillan, 1980)
Asprey, Robert B., The German High Command at War: Hindenburg and Ludendorff
Conduct World War I (New York: William Morrow, 1991)
Audoin-Rouzeau, Stephane and Becker, Annette, 1914–1918: Understanding the Great
War (Catherine Temerson, trans.) (London: Profile, 2002)
Bailey, Jonathan, The First World War and the Birth of the Modern Style of Warfare,
Strategic and Combat Studies Institute Occasional Paper No. 22 (SCSI: Camberley,
1996)
Bailey, Jonathan B. A., ‘The First World War and the Birth of Modern Warfare’, in
MacGregor Knox and Williamson Murray (eds), The Dynamics of Military Revolution
1300–2050 (Cambridge: Cambridge University Press, 2001), pp. 132–53
Beach, Jim, Haig’s Intelligence: GHQ and the German Army, 1916–18 (Cambridge:
Cambridge University Press, 2013)
Beckett, Ian F. W., Ypres:The First Battle, 1914 (Harlow: Pearson, 2004)
Beckett, Ian F. W. and Corvi, Steven J. (eds), Haig’s Generals (Barnsley: Pen & Sword,
2006)
Berghahn,Volker R., Imperial Germany, 1871–1914: Economy, Society, Culture, and Politics
(Providence, RI: Berghahn Books, 1994)
Bernhardi, Friedrich von, Deutschlands Heldenkampf 1914–1918 (Munich: J. F. Lehmann,
1922)
Biddle, Stephen, Military Power: ExplainingVictory and Defeat in Modern Battle (Princeton,
NJ: Princeton University Press, 2004)
Bidwell, Shelford and Graham, Dominick, Fire-Power: British Army Weapons and
Theories of War 1904–1945 (London: George Allen and Unwin, 1982)
Binneveld, Hans, From Shell Shock to Combat Stress: A Comparative History of Military
Psychiatry (Amsterdam: Amsterdam University Press, 1997)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

334 bi bliogr a ph y

Boff, Jonathan, Winning and Losing on the Western Front:The British Third Army and the
Defeat of Germany in 1918 (Cambridge: Cambridge University Press, 2012)
Bond, Brian and Cave, Nigel (eds), Haig: A Reappraisal 70 Years On (Barnsley: Leo
Cooper, 1999)
Bourke, Joanna, Dismembering the Male: Men’s Bodies, Britain and the Great War
(London: Reaktion, 1996)
Bourke, Joanna, An Intimate History of Killing: Face-to-Face Killing in Twentieth-Century
Warfare (London: Granta, 2000 [1999])
Breg, Josef, Kronprinz Rupprecht von Bayern: Ein Lebensbild (Munich: Max Kellerer,
1918)
Bristow,Adrian, A Serious Disappointment:The Battle of Aubers Ridge and the Subsequent
Munitions Scandal (London: Leo Cooper, 2005)
British Commission for Military History, ‘Look to your Front’: Studies in the First
World War (Staplehurst: Spellmount, 1999)
Brose, Eric Dorn, The Kaiser’s Army: The Politics of Military Technology in Germany
during the Machine Age, 1870–1918 (Oxford: Oxford University Press, 2001)
Brown, Ian M., ‘Not Glamorous, But Effective: The Canadian Corps and the Set-
Piece Attack, 1917–1918’, Journal of Military History 58 ( July 1994), pp. 421–44
Brown, Ian M., British Logistics on the Western Front 1914–1919 (Westport, CT: Praeger,
1998)
Campbell, Frederick Francis, ‘The Bavarian Army 1870–1918: The Constitutional
and Structural Relations with the Prussian Military Establishment’ (unpublished
PhD thesis, Ohio State University, 1972)
Chasseaud, Peter, ‘Field Survey in the Salient: Cartography and Artillery Survey in
the Flanders Operations in 1917’, in Peter H. Liddle (ed.), Passchendaele in
Perspective: The Third Battle of Ypres (Barnsley: Pen & Sword, 2013 [1997]),
pp. 117–39
Chickering, Roger, Imperial Germany and the GreatWar, 1914–1918 (2nd edn, Cambridge:
Cambridge University Press, 2004)
Citino, Robert M., The German Way of War: From the Thirty Years’ War to the Third
Reich (Lawrence, KS: University Press of Kansas, 2005)
Clark, Alan, The Donkeys (London: Hutchinson, 1961)
Clark, Christopher, Kaiser Wilhelm II: A Life in Power (London: Penguin, 2009)
Clayton, Anthony, Paths of Glory: The French Army 1914–18 (London: Cassell, 2005
[2003])
Corrigan, Gordon, Sepoys in the Trenches:The Indian Corps on the Western Front 1914–
1915 (Staplehurst: Spellmount, 1999)
Corum, James S., ‘Air War over the Somme: The Evolution of British and German
Air Operations’, in Matthias Strohn (ed.), The Battle of the Somme (Oxford:
Osprey, 2016), pp. 75–91
Cowan, Anthony, ‘Genius for War? German Operational Command on the Western
Front in Early 1917’ (unpublished PhD thesis: King’s College London, 2016)
Creveld, Martin van, Fighting Power: German and U.S. Army Performance, 1939–1945
(Westport, CT: Greenwood Press, 1982)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y335

Creveld, Martin van, Command in War (Cambridge, MA: Harvard University Press,
1985)
Creveld, Martin van, Supplying War: Logistics from Wallenstein to Patton (Cambridge:
Cambridge University Press, 2004 [1977])
Deist,Wilhelm, Militär und Innenpolitik im Weltkrieg 1914–1918, 2 volumes (Dusseldorf:
Droste, 1990)
Deist, Wilhelm, ‘Verdeckter Militärstreik im Kriegsjahr 1918?’, in Wolfram Wette
(ed.), Der Krieg des kleinen Mannes: Eine Militärgeschichte von unten (Munich: Piper,
1992), pp. 146–67
Deist, Wilhelm, ‘The Military Collapse of the German Empire: The Reality Behind
the Stab-in-the-Back Myth’ (E. J. Feuchtwanger, trans.), War in History 3:2 (April
1996), pp. 186–207
Dennis, Peter and Grey, Jeffrey (eds), 1918: Defining Victory: Proceedings of the Chief of
Army’s History Conference held at the National Convention Centre, Canberra 29
September 1998 (Canberra: Army History Unit, Department of Defence, 1999)
Doughty, Robert, Pyrrhic Victory: French Strategy and Operations in the Great War
(Cambridge, MA: Belknap Press, 2005)
Duffy, Christopher, Through German Eyes:The British and The Somme 1916 (London:
Weidenfeld & Nicolson, 2006)
Dupuy, Trevor N., A Genius for War: The German Army and General Staff, 1807–1945
(London: Macdonald and Jane’s, 1977)
Echevarria II, Antulio J., After Clausewitz: German Military Thinkers before the Great
War (Lawrence, KS: University of Kansas Press, 2000)
Ehlert, Hans, Epkenhans, Michael and Groß, Gerhard P. (eds), Der Schlieffenplan:
Analyse und Dokumente (Paderborn: Ferdinand Schöningh, 2006)
Einem, Karl von (Junius Alter, ed.), Ein Armeeführer erlebte den Weltkrieg (Leipzig:Von
Hase & Koehler, 1938)
Ekins, Ashley (ed.), 1918 Year of Victory: The End of the Great War and the Shaping of
History (Auckland: Exisle Publishing, 2010)
Eksteins, Modris, The Rites of Spring: The Great War and the Birth of the Modern Age
(London: Bantam, 1989)
Erdmann, Karl Dietrich, Der Erste Weltkrieg (Munich: Deutscher Taschenbuch, 1980)
Ertman, Thomas, Birth of Leviathan: Building States and Regimes in Medieval and Early
Modern Europe (Cambridge: Cambridge University Press, 1997)
Ferguson, Niall, The Pity of War (London: Allen Lane, 1998)
Foley, Robert T., German Strategy and the Path to Verdun: Erich von Falkenhayn and
the Development of Attrition 1870–1916 (Cambridge: Cambridge University Press,
2005)
Foley, Robert T., ‘Learning War’s Lessons: The German Army and the Battle of the
Somme 1916’, Journal of Military History 75 (April 2011), pp. 471–504
Foley, Robert T., ‘A Case Study in Horizontal Military Innovation: The German
Army 1916–1918’, Journal of Strategic Studies 35:6 (2012), pp. 799–827
Foley, Robert T., ‘Dumb Donkeys or Cunning Foxes? Learning in the British and
German Armies during the Great War’, International Affairs 90:2 (2014), pp. 279–98
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

336 bi bliogr a ph y

Fox-Godden, Aimée, ‘ “Putting Knowledge in Power”: Learning and Innovation in

the British Army of the First World War’ (unpublished PhD thesis, University of
Birmingham, 2015)
Fox-Godden, Aimée, ‘Beyond the Western Front: The Practice of Inter-Theatre
Learning in the British Army during the First World War’, War in History 23:2
(2016), pp. 190–209
French, David, Raising Churchill’s Army:The British Army and the War against Germany
1919–1945 (Oxford: Oxford University Press, 2000)
French, David, ‘Doctrine and Organisation in the British Army, 1919–32’, Historical
Journal 44 (2001), pp. 497–515
French, David, Military Identities: The Regimental System, the British Army, and the
British People, c.1870–2000 (Oxford: Oxford University Press, 2005)
Fussell, Paul, The Great War and Modern Memory (London: Oxford University Press,
1977)
Gardner, Nikolas, ‘The Beginning of the Learning Curve: British Officers and the
Advent of Trench Warfare, September–October 1914’ (ESRI Working Papers:
Salford, 2003)
Gardner, Nikolas, Trial by Fire: Command and the British Expeditionary Force in 1914
(Westport, CT: Praeger, 2003)
Geddes, Alistair, ‘Solly-Flood, GHQ, and Tactical Training in the BEF, 1916–1918’
(unpublished MA dissertation, University of Birmingham, 2007)
Gerwarth, Robert, The Vanquished: Why the First World War Failed to End (London:
Allen Lane, 2016)
Geyer, Michael, ‘Insurrectionary Warfare: The German Debate about a Levée en
Masse in October 1918’, Journal of Modern History 73:3 (September 2001),
pp. 459–527
Goetz,Walter Wilhelm, Rupprecht: Kronprinz von Bayern 1869–1955 (Munich: Bayerische
Akademie für Wissenschaften, 1956)
Goya, Michel, La Chair et l’Acier: L’Invention de la Guerre Moderne (1914–1918) (Paris:
Tallandier, 2004)
Greenhalgh, Elizabeth, ‘Myth and Memory: Sir Douglas Haig and the Imposition
of Allied Unified Command in March 1918’, Journal of Military History 68:2
(2004), pp. 771–820
Greenhalgh, Elizabeth, Foch in Command: The Forging of a First World War General
(Cambridge: Cambridge University Press, 2011)
Greenhalgh, Elizabeth, The French Army and the First World War (Cambridge:
Cambridge University Press, 2014)
Griffith, Paddy, Battle Tactics of the Western Front:The British Army’s Art of Attack 1916–18
(New Haven, CT:Yale University Press, 1994)
Griffith, Paddy (ed.), British Fighting Methods in the Great War (London: Frank Cass,
1996)
Groß, Gerhard P., ‘Das Dogma der Beweglichkeit: Überlegungen zur Genese der
deutschen Heerestaktik im Zeitalter der Weltkriege’, in Bruno Thoß and
­Hans-Erich Volkmann (eds), Erster Weltkrieg, Zweiter Weltkrieg: Ein Vergleich: Krieg,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y337

Kriegserlebnis, Kriegserfahrung in Deutschland (Paderborn: Ferdinand Schöningh,

2002), pp. 143–66
Groß, Gerhard P., Mythos und Wirklichkeit: Geschichte des operativen Denkens im deutschen
Heer von Moltke d.Ä bis Heusinger (Paderborn: Ferdinand Schöningh, 2012)
Gudmundsson, Bruce I., Stormtroop Tactics: Innovation in the German Army, 1914–1918
(New York: Praeger, 1989)
Gudmundsson, Bruce I., On Artillery (Westport, CT: Praeger, 1993)
Hammond, Bryn, Cambrai 1917: The Myth of the First Great Tank Battle (London:
Weidenfeld & Nicolson, 2008)
Harris, J. P., Men, Ideas and Tanks: British Military Thought and Armoured Forces, 1903–
1939 (Manchester: Manchester University Press, 1995)
Harris, J. P., Sir Douglas Haig and the First World War (Cambridge: Cambridge
University Press, 2008)
Harris, J. P. with Barr, Niall, Amiens to the Armistice: The BEF in the Hundred Days
Campaign, 8 August–11 November 1918 (London: Brassey’s, 1998)
Harris, Paul and Marble, Sanders, ‘The “Step-by-Step” Approach: British Military
Thought and Operational Method on the Western Front, 1915–1917’, War in
History 15:1 (January 2008), pp. 17–42
Hartcup, Guy, The War of Invention: Scientific Developments, 1914–18 (London: Brassey’s,
1988)
Herwig, Holger H., The First World War: Germany and Austria-Hungary 1914–1918
(London: Arnold, 1997)
Herwig, Holger H., The Marne, 1914: The Opening of World War I and the Battle that
Changed the World (New York: Random House, 2009)
Heyck,Thomas William,‘Myths and Meanings of Intellectuals in Twentieth-Century
British National Identity’, Journal of British Studies 37:2 (April 1998), pp. 192–221
Hirschfeld, G., Krumeich G., and Renz, I. (eds) Die Deutschen an der Somme 1914–
1918 (Essen: Klartext, 2006)
Holmes, Richard, The Little Field-Marshal: Sir John French (London: Jonathan Cape,
1981)
Horne, John and Kramer, Alan, German Atrocities 1914: A History of Denial (New
Haven, CT:Yale University Press, 2001)
Howard, Michael, ‘Men against Fire: Expectations of War in 1914’, International
Security 9:1 (Summer 1984), pp. 41–57
Hughes, Matthew and Seligmann, Matthew (eds), Leadership in Conflict 1914–1918
(Barnsley: Leo Cooper, 2000)
Hull, Isabel V., The Entourage of Kaiser Wilhelm II 1888–1918 (Cambridge: Cambridge
University Press, 1982)
Hull, Isabel V., Absolute Destruction: Military Culture and the Practices of War in Imperial
Germany (Ithaca, NY: Cornell University Press, 2005)
Hull, Isabel V., A Scrap of Paper: Breaking and Making International Law during the Great
War (Ithaca, NY: Cornell University Press, 2014)
Hynes, Samuel, A War Imagined: The First World War and English Culture (London:
Pimlico, 1992 [1990])
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

338 bi bliogr a ph y

Hynes, Samuel, The Soldier’s Tale: Bearing Witness to Modern War (London: Pimlico,
1998)
Jahr, Christoph, Gewöhnliche Soldaten: Desertion und Deserteure im deutschen und
britischen Heer 1914–1918 (Göttingen:Vandenhoeck & Ruprecht, 1998)
Johnson, Hubert C., Breakthrough! Tactics, Technology and the Search for Victory on the
Western Front in World War I (Novato, CA: Presidio, 1994)
Jones, Heather, Violence against Prisoners of War: Britain, France and Germany, 1914–20
(Cambridge: Cambridge University Press, 2011)
Kirke, Charles, Red Coat, Green Machine: Continuity in Change in the British Army
1700 to 2000 (London: Continuum, 2009)
Kitchen, Martin, The German Officer Corps 1890–1914 (Oxford: Clarendon Press, 1968)
Kitchen, Martin, The Silent Dictatorship: The Politics of the German High Command
under Hindenburg and Ludendorff, 1916–1918 (London: Croom Helm, 1976)
Kolshorn, Otto, Kronprinz Rupprecht von Bayern: Ein Lebens- und Charakterbild (Munich:
R. Piper & Co., 1918)
Krause, Jonathan, Early Trench Tactics in the French Army: The Second Battle of Artois,
May–June 1915 (Farnham: Ashgate, 2013)
Krause, Jonathan,‘The French Battle for Vimy Ridge, Spring 1915’, Journal of Military
History 77 ( January 2013), pp. 91–113
Laffin, John, British Butchers and Bunglers of World War One (Stroud: Sutton, 1988)
Liddell Hart, Basil, The Real War (London: Faber & Faber, 1930)
Lloyd, Nick, Loos 1915 (Stroud: History Press, 2008 [2006])
Lloyd, Nick, Hundred Days:The End of the Great War (London:Viking, 2013)
Lupfer, Timothy T., The Dynamics of Doctrine: The Change in German Tactical Doctrine
during the First World War (Leavenworth Paper No. 4: Combat Studies Institute,
U.S. Army Command and General Staff College, Fort Leavenworth, Kansas, 1981)
Luvaas, Jay, The Education of an Army: British Military Thought, 1815–1940 (London:
Cassell, 1965)
Machtan, Lothar, Prinz Max von Baden: Der letzte Kanzler des Kaisers: Eine Biografie
(Berlin: Suhrkamp, 2013)
McRandle, James and Quirk, James, ‘The Blood Test Revisited: A New Look at
German Casualty Counts in World War I’, Journal of Military History 70 (2006),
pp. 667–702
März, Stefan, Das Haus Wittelsbach im Ersten Weltkrieg: Chance und Zusammenbruch
monarchischer Herrschaft (Regensburg: Friedrich Pustet, 2013)
Messenger, Charles, Call to Arms: The British Army 1914–18 (London: Weidenfeld &
Nicolson, 2005)
Messenger, Charles, The Day We Won the War: Turning Point at Amiens, 8 August 1918
(London: Weidenfeld & Nicolson, 2008)
Michaelis, Herbert and Schraepler, Ernst (eds), Ursachen und Folgen vom deutschen
Zusammenbruch 1918 and 1945 bis zur staatlichen Neuordnung Deutschlands in der
Gegenwart Volume II: Der militärische Zusammenbruch und das Ende des Kaiserreichs
(Berlin: Herbert Wendler & Co., n.d.)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y339

Millett,Allan R. and Murray,Williamson (eds), Military Effectiveness, 3 volumes (Boston,

MA: Allen & Unwin, 1988)
Mombauer, Annika, Helmuth von Moltke and the Origins of the First World War
(Cambridge: Cambridge University Press, 2001)
Mombauer, Annika and Deist, Wilhelm (eds), The Kaiser: New Research on Wilhelm
II’s Role in Imperial Germany (Cambridge: Cambridge University Press, 2003)
Müller, Thomas, Konrad Krafft von Dellmensingen (1862–1953): Porträt eines bayerischen
Offiziers (Munich: Kommission für bayerische Landesgeschichte, 2002)
Murray, Williamson, Military Adaptation in War: With Fear of Change (Cambridge:
Cambridge University Press, 2011)
Nebelin, Manfred, Ludendorff: Diktator im Ersten Weltkrieg (Munich: Siedler, 2010)
Neiberg, Michael S., The Second Battle of the Marne (Bloomington, IN: Indiana
University Press, 2008)
Palazzo, Albert, ‘The British Army’s Counter Battery Staff Office and Control of
the Enemy in World War I’, Journal of Military History 63 ( January 1999), pp. 55–74
Palazzo, Albert, Seeking Victory on the Western Front: The British Army and Chemical
Warfare in World War I (Lincoln, NE: University of Nebraska Press, 2000)
Philpott, William, Bloody Victory: The Sacrifice on the Somme and the Making of the
Twentieth Century (London: Little, Brown, 2009)
Philpott, William, War of Attrition: Fighting the First World War (New York: Overlook
Press, 2014)
Pöhlmann, Markus, Kriegsgeschichte und Geschichtspolitik: Der Erste Weltkrieg: Die amtliche
deutsche Militärgeschichtsschreibung 1914–1956 (Paderborn: Ferdinand Schöningh, 2002)
Pöhlmann, Markus, ‘Yesterday’s Battles and Future War: The German Official
Military History, 1918–1939’, in Roger Chickering and Stig Förster (eds), The
Shadows of Total War: Europe, East Asia, and the United States, 1919–1939 (Cambridge:
Cambridge University Press, 2003), pp. 223–38
Pommerin, Reiner, ‘Die Ausweisung von “Ostjuden” aus Bayern 1923: Ein Beitrag
zum Krisenjahr der Weimarer Republik’, Viertelsjahrhefte für Zeitgeschichte 34:3
( July 1986), pp. 311–40
Prior, Robin and Wilson, Trevor, ‘Paul Fussell at War’, War in History 1:1 (1994),
pp. 63–80
Prior, Robin and Wilson, Trevor, Passchendaele: The Untold Story (New Haven, CT:
Yale University Press, 1996)
Prior, Robin and Wilson, Trevor, Command on the Western Front: The Military Career
of Sir Henry Rawlinson (Barnsley: Pen & Sword, 2004 [1992])
Prior, Robin and Wilson, Trevor, The Somme (New Haven, CT: Yale University
Press, 2005)
Pugsley, Christopher, ‘Trial and Error: the Dominion forces on the Somme in 1916’,
in Matthias Strohn (ed.), The Battle of the Somme (Oxford: Osprey, 2016),
pp. 113–29
Pyta, Wolfram, Hindenburg: Herrschaft zwischen Hohenzollern und Hitler (Munich:
Siedler, 2007)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

340 bi bliogr a ph y

Rawling, Bill, Surviving Trench Warfare: Technology and the Canadian Corps 1914–1918
(Toronto: University of Toronto Press, 1992)
Retallack, James (ed.), Imperial Germany 1871–1918 (Oxford: Oxford University Press,
2008)
Ritter, Gerhard, The Sword and the Sceptre: The Problem of Militarism in Germany,
4 volumes (Heinz Norden, trans.) (London: Allen Lane, 1973)
Robbins, Simon, British Generalship on the Western Front 1914–18: Defeat into Victory
(London: Frank Cass, 2005)
Röhl, John C. G., The Kaiser and his Court:Wilhelm II and the Government of Germany
(Terence F. Cole, trans.) (Cambridge: Cambridge University Press, 1994)
Röhl, John C. G., Wilhelm II: Into the Abyss of War and Exile 1900–1941 (Sheila de
Bellaigue and Roy Bridge, trans.) (Cambridge: Cambridge University Press, 2014)
Samuels, Martin, Doctrine and Dogma: German and British Infantry Tactics in the First
World War (Westport, CT: Greenwood Press, 1992)
Samuels, Martin, Command or Control? Command,Training and Tactics in the British and
German Armies, 1888–1918 (London: Frank Cass, 1995)
Sendtner, Kurt, Rupprecht von Wittelsbach, Kronprinz von Bayern (Munich: Richard
Pflaum Verlag, 1954)
Sheffield, Gary, Forgotten Victory: The First World War: Myths and Realities (London:
Headline, 2001)
Sheffield, Gary, The Somme (London: Cassell, 2004 [2003])
Sheffield, Gary, The Chief: Douglas Haig and the British Army (London: Aurum,
2011)
Sheffield, Gary, Command and Morale:The British Army on the Western Front, 1914–1918
(Barnsley: Pen & Sword, 2014)
Sheffield, Gary and Gray, Peter, Changing War: The British Army, the Hundred Days
Campaign and the Birth of the Royal Air Force, 1918 (London: Bloomsbury, 2013)
Sheffield, Gary and Todman, Dan (eds), Command and Control on the Western Front:
The British Army’s Experience 1914–1918 (Staplehurst: Spellmount, 2004)
Sheldon, Jack, The Germany Army on the Somme 1916 (Barnsley: Leo Cooper, 2005)
Sheldon, Jack, The German Army at Passchendaele (Barnsley: Pen & Sword, 2007)
Sheldon, Jack, The German Army on Vimy Ridge 1914–1917 (Barnsley: Pen & Sword,
2008)
Sheldon, Jack, The German Army at Cambrai (Barnsley: Pen & Sword, 2009)
Sheldon, Jack, The German Army At Ypres, 1914 (Barnsley: Pen & Sword, 2010)
Showalter, Dennis, Tannenberg: Clash of Empires, 1914 (Hamden: Archon, 1991)
Sigg, Marco, Der Unterführer als Feldherr im Taschenformat: Theorie und Praxis der
Auftragstaktik im deutschen Heer 1869 bis 1945 (Paderborn: Ferdinand Schöningh, 2014)
Simkins, Peter, From the Somme to Victory:The British Army’s Experience on the Western
Front 1916–1918 (Barnsley: Pen & Sword, 2014)
Simpson, Andy, Directing Operations: British Corps Command on the Western Front
1914–18 (Stroud: Spellmount, 2006)
Stachelbeck, Christian, Militärische Effektivität im Ersten Weltkrieg: Die 11. Bayerische
Infanteriedivision 1915 bis 1918 (Paderborn: Ferdinand Schöningh, 2010)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

bi bliogr a ph y341

Stachelbeck, Christian, Deutschlands Heer und Marine im Ersten Weltkrieg (Munich:

Oldenbourg, 2013)
Stephenson, Scott, The Final Battle: Soldiers of the Western Front and the German
Revolution of 1918 (Cambridge: Cambridge University Press, 2009)
Stevenson, David, 1914–1918: The History of the First World War (London: Allen Lane,
2004)
Stevenson, David, With Our Backs to the Wall:Victory and Defeat in 1918 (London: Allen
Lane, 2011)
Storz, Dieter, ‘ “Aber was hätte anders geschehen sollen?” Die deutschen Offensiven
an der Westfront’, in Jörg Duppler and Gerhard Groß (eds), Kriegsende 1918:
Ereignis,Wirkung, Nachwirkung (Munich: R. Oldenbourg, 1999), pp. 51–95
Storz, Dieter, ‘ “Dieser Stellungs- und Festungskrieg ist scheußlich!” Zu den
Kämpfen in Lothringen und in den Vogesen im Sommer 1914’, in Hans Ehlert,
Michael Epkenhans, and Gerhard P. Groß (eds), Der Schlieffenplan: Analyse und
Dokumente (Paderborn: Ferdinand Schöningh, 2006), pp. 162–204
Storz, Dieter, ‘Kronprinz Rupprecht von Bayern—dynastische Heeresführung im
Massenkrieg’, in Winfried Heinemann and Markus Pöhlmann (eds), Monarchen
und ihr Militär (Potsdam: Militärgeschichtliches Forschungsamt, 2010), pp. 45–57
Strachan, Hew, ‘The Morale of the German Army 1917–18’, in Hugh Cecil and
Peter Liddle (eds), Facing Armageddon: The First World War Experienced (London:
Leo Cooper, 1996), pp. 383–98
Strachan, Hew, The First World War Volume I: To Arms (Oxford: Oxford University
Press, 2001)
Strohn, Matthias, The German Army and the Defence of the Reich: Military Doctrine and
the Conduct of the Defensive Battle 1918–39 (Cambridge: Cambridge University Press,
2010)
Strohn, Matthias (ed.), The Battle of the Somme (Oxford: Osprey, 2016)
Strong, Paul and Marble, Sanders, Artillery in the Great War (Barnsley: Pen & Sword
Military, 2011)
Taylor, A. J. P., The First World War: An Illustrated History (London: Hamish Hamilton,
1963)
Terraine, John, To Win a War: 1918 The Year of Victory (London: Sidgwick and Jackson,
1978)
Terraine, John, White Heat: The New Warfare 1914–18 (London: Leo Cooper, 1992
[1982])
Thoß, Bruno and Volkmann, Hans-Erich, Erster Weltkrieg, Zweiter Weltkrieg: Ein
Vergleich: Krieg, Kriegserlebnis, Kriegserfahrung in Deutschland (Paderborn: Ferdinand
Schöningh, 2002)
Travers, T. H. E., ‘The Offensive and the Problem of Innovation in British Military
Thought, 1870–1915’, Journal of Contemporary History 13 (1978), pp. 531–53
Travers, Tim, ‘The Hidden Army: Structural Problems in the British Officer Corps,
1900–1918’, Journal of Contemporary History 17:3 (July 1982), pp. 523–44
Travers,Tim, The Killing Ground:The British Army, the Western Front and the Emergence
of Modern Warfare 1900–1918 (London: Allen & Unwin, 1987)
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

342 bi bliogr a ph y

Travers, Tim, How the War Was Won: Command and Technology in the British Army on
the Western Front, 1917–1918 (London: Routledge, 1992)
Uhle-Wettler, Franz, Erich Ludendorff in seiner Zeit: Soldat—Stratege—Revolutionär—
Eine Neubewertung (Berg: Kurt Vowinckel, 1995)
Ulrich, Bernd and Ziemann, Benjamin (eds), German Soldiers in the Great War: Letters
and Eyewitness Accounts (Barnsley: Pen and Sword, 2010)
Unruh, Karl, Langemarck: Legende und Wirklichkeit (Koblenz: Bernard & Graefe,
1986)
Valentini, Rudolf von, Kaiser und Kabinettschef (Oldenburg: Gerhard Stalling, 1931)
Venohr, Wolfgang, Ludendorff: Legende und Wirchlichkeit (Berlin: Ullstein, 1993)
Watson, Alexander, ‘Self-Deception and Survival: Mental Coping Strategies on the
Western Front, 1914–18’, Journal of Contemporary History 41:2 (2006), pp. 247–68
Watson,Alexander, Enduring the Great War: Combat, Morale and Collapse in the German
and British Armies, 1914–1918 (Cambridge: Cambridge University Press, 2008)
Watson, Alexander, Ring of Steel: Germany and Austria-Hungary at War, 1914–1918
(London: Allen Lane, 2014)
Wavell, Archibald, Generals and Generalship (Harmondsworth: Penguin, 1941)
Weiß, Dieter J., Kronprinz Rupprecht von Bayern: Eine politische Biografie (Regensburg:
Verlag Friedrich Pustet, 2007)
Winter, Denis, Haig’s Command: A Reassessment (London:Viking, 1991)
Wynne, G. C., If Germany Attacks: The Battle in Depth in the West (Westport, CT:
Greenwood Press, 1976 [1940])
Zabecki, David T., The German 1918 Offensives: A Case Study in the Operational Level
of War (New York: Routledge, 2006)
Zabecki, David T. (ed.), Chief of Staff: The Principal Officers behind History’s Great
Commanders Volume I: Napoleonic Wars to World War I (Annapolis, MD: Naval
Institute Press, 2008)
Ziemann, Benjamin, Front und Heimat: Ländliche Kriegserfahrungen im südlichen Bayern
1914–1923 (Essen: Klartext, 1997)
Ziemann, Benjamin, War Experiences in Rural Germany 1914–1923 (Alex Skinner,
trans.) (Oxford: Berg, 2007)
Zuber, Terence, The Mons Myth: A Reassessment of the Battle (Stroud: History Press,
2010)
 OUP CORRECTED PROOF – FINAL, 18/12/17, SPi

Picture Acknowledgements

Note: BayHStA = Bayerisches Hauptstaatsarchiv;GHA = Geheimes Hausarchiv;

W.B. = Wittelsbacher Bildersammlung; AWM = Australian War Memorial
Frontispiece AWM H12371
2.1 BayHStA, GHA, W.B. 249/242e
5.1 Bayerisches Hauptstaatsarchiv (BayHStA), Geheimes Hausarchiv
(GHA), Wittelsbacher Bildersammlung (W.B.) III 3/229a
8.1 BayHStA, GHA, W.B. 149/151a
11.1 BayHStA, GHA, W.B. 243/245d (author: Franz Grainer)
12.1 BayHStA, GHA, W.B. 112/114e
13.1 AWM H12354
14.1 BayHStA, GHA, W.B. 2996
15.1 AWM H11990
20.1 AWM H13220
22.1 Bayerische Staatsbibliothek München/Bildarchiv
23.1 BayHStA, GHA, W.B. 67/68 (author: Franz Grainer)
OUP CORRECTED PROOF – FINAL, 18/12/17, SPi
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

Index

Abbeville  212, 225 Amsterdam 247

Ablain-Saint-Nazaire  74–5, 78 Ancre Heights, Battle of (1916)  140
Administration  19, 113 Ancre River  37, 122, 131, 140, 149,
aeroplanes  98, 129, 132, 140, 173 213, 216
see also under reconnaissance Anglophobia 43
British  173, 196, 230–1, 234 Angres  85, 87
see also Royal Air Force anti-semitism 249
(RAF), the Antonie of Luxembourg, Princess
Entente  102, 129, 132, 140, 229–30, (Rupprecht’s second wife)
234 235–6, 241, 247, 249, 253
French 234 Antwerp 41
German  36, 129–30, 132, 136, 140, Antwerp–Meuse Line, the  243
145–6, 151–2, 221 Ardennes, the  24
Afflerbach, Holger  92, 266, 292n.15, Argonne Forest  240
308n.13 aristocracy 5
Aisne River  29, 151–2, 224, 228, 232, 261, Arleux 164
314n.17 Armentières  41–3, 202–3, 222
Albert  37–8, 98, 102, 131, 214, 236 Armistice, the  2, 244, 273
Albrecht, Prince of Bavaria (Rupprecht’s Arras  37–42, 61, 68, 73, 75, 85, 88, 93, 95,
son)  27, 105, 112, 235–6, 244, 99, 104, 121, 125, 151–2,
249, 251 172–3, 175, 203–5, 207, 211,
Allenby, General Sir Edmund  156, 162–3 214, 216, 225–6, 236–7,
Alsace  18, 21, 84, 239, 264 260, 263
see also Alsace-Lorraine Arras, Battle of (1917)  156, 163–4, 258,
Alsace-Lorraine  18, 22, 24, 27, 29–30, 99, 313n.8, 314n.17
227, 239, 264–5, 275 Arras–Cambrai Road  237
Amiens  37, 137, 151, 203, 209, 212, arson 33
214–16, 218–19, 221, 223, artillery  27, 38, 46, 61, 65–6, 80–1, 88, 92,
225, 228, 234–5 96–7, 102, 132, 164, 168,
Amiens, Battle of (1918)  234, 322n.3 172–4, 180, 184 see also
ammunition  136, 174 see also supplies ammunition and artillery
American 83 obervers
British  63, 65, 101, 123, 156, 174 British  38, 46, 63, 65, 71–2, 82, 85–6,
shortages of  46, 72 97, 101–2, 128, 132, 137, 145,
Entente 123 151, 153–4, 156, 161, 163–4,
French  123, 137 166, 169–75, 178, 180, 183–5,
shortages of  88 190, 196, 210, 212, 231,
German  57, 85, 121, 140, 196 234, 279
shortages of  39, 46–7, 49, 59, 123, creeping barrages of the  138, 140,
185, 195, 203, 214, 271 156, 170, 174
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

346 I n de x

artillery (cont.) attacks see offensives/attacks

preparation by the  63, 65, 72, 85, 88, Attenborough, Richard
127, 154 Oh! What a Lovely War 4
lack of  64, 81, 101, 190 attrition  95, 98, 101, 121, 137, 141–2, 145,
Canadian 156 152, 204, 274, 278
Entente  66, 74, 94, 130, 132, 152, Aubers Ridge  43, 63, 66–7, 70–4, 78
157–8, 177 Auftragstaktik see command systems,
French  23, 25–6, 39, 70, 79, 84–5, German, flexible
90, 97–8, 102, 132, 137, 145, Augusta Victoria (Kaiserin)  58, 105
153, 156 Australian forces  158, 229, 234, 237, 240
best manner of using  52, 96 Austria  245–6, 250, 262, 265, 272
preparation by the  73, 78, 82, 85, 96 see also Austria-Hungary
lack of  25 Austria-Hungary  15, 47, 68–9, 92, 106,
rolling barrages  73, 97 111, 265 see also Austria and
support of the  90 Habsburgs, the and Hungary
German  26, 28, 38, 46, 49, 63, 65, 68, authority  76, 113, 115–16, 152, 260
78, 81, 85, 97, 99, 102, 121, see also leadership
124–6, 129, 132–3, 135, delegation of  4, 116, 146, 260, 268
139–40, 144–6, 152, 159, see also responsibility,
163–4, 171–4, 178, 182–4, delegation of
192–3, 205, 216, 220–1, 228, undercutting of  97, 114
231, 238 Avesnes 207
best manner of using  51, 132, Avre River  215
135–6, 145
creeping barrages of the  210, Baden  35, 262, 264
222, 231 Baden, Prince Max von  227, 235, 241–2,
preparation by the  193, 210, 216 271–2
lack of  66 Badonviller  21–2, 30, 50
support of the  22, 40, 43, 50–1, Bailleul  157, 202, 205, 222
70, 197, 206, 231 see also Balfourier, General Maurice  100
coordination/cooperation, Bamberg 14
German, combined arms Banteux 190
lack of  49 Bapaume  37–8, 102, 126, 136, 138,
artillery observers 141, 205–6, 211, 213, 218,
British 164 236–7, 260
French 26 barbed wire  26, 60, 63, 65, 71, 74, 101,
German  51, 97, 131, 146, 164, 169, 174 140, 156, 190
Artois  60, 66, 75, 77, 80, 84–6, 88, 106, Barbot, General Ernest  74
116, 162, 260, 304n.16, Bauer, Lieutenant-Colonel Max  28, 105
327n.1 Bavaria  11–13, 16, 20, 49, 76–7, 235,
Artois, Second Battle (1915) of  73, 80, 241, 244–51, 253, 261–6,
82, 118, 327n.1 269–70, 272–4
Artois, Third Battle of (1915)  89, 93, 116, army of see Bavarian army
327n.1 culture of  13, 114
Asquith, Herbert  89 as a part of Germany  12
atrocities government of  12, 15–16, 245–7, 265,
British 69 269, 273–4, 292n.1
German  96, 150, 297n.6 see also history of  8, 12–13
scorched earth policy and people of  18, 20, 28, 77, 104, 112–13,
war crimes 244, 246, 252, 262, 264, 266
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x347

and Prussia  12–13, 18, 77, 150, Bethmann Hollweg, Theobald von  16,
261–5, 272–4 57–8, 104–5, 227, 264, 266,
revolution/unrest in  244–8 270, 274–5
royal family of  2, 11–13, 15, 236, Bihucourt 127
244–51, 262, 265, 273–4 Bismarck, Otto von  327n.1
see also Wittelsbachs, the Bissing, General Moritz von  104
soldiers of see German Armies, Sixth, Bixschoote 47
Bavarians in the Blackadder 4
and German army, Bavarian Blair, Tony  279
formations of the blame see also explanations and war guilt
and soldiers, Bavarian ascription of  50, 66, 126, 160, 175, 181,
War College of  14, 18, 77 194–5, 234–5, 237, 240
War Ministry of  19, 244 Blondat, General Ernest  74
Bavarian army  106, 264 see also German Blue Max see Pour le Mérite
army, Bavarian formations Boehn, General Max von  233, 236–7,
of the and German army, 239, 241–2, 263
Armies of the, Sixth Army, Boelcke, Oswald  140
Bavarians in the Boer War, the  191
archives of the  1–2, 6–8 Bois Grenier  220
divisions of the 1st  14 Bois de St Pierre Vaast  130–1, 138, 218
General Staff of the  17 Bolsheviks  2, 202, 227, 240, 246 see also
mobilization of the  16, 19 Russia, revolution in
official history of the  186 Bonaparte, Napoleon  214, 278
Regiment of Foot Guards  13, 21–2, 50 Bonneau, General Louis  21
size of the  19 Bouchavesnes  126, 129–31
Die Bayern im Großen Kriege 1914–1918 8 Boulogne 213
Bazentin Ridge  102 Bourlon Wood  192–4, 218
Beach, Jim  151, 166, 188, 194 Braganzas, the  11
Beaumont 37 breakthroughs  212, 240 see deadlock/
Beaumont Hamel  100, 140 stalemate, attempts to break
Beer Hall Putsch, the  250–1 hopes of achieving  81, 87, 90, 94,
Behr, Major Max von  230 100–1, 141, 155–6, 189,
Belgium  2, 18, 29, 31, 34–5, 45, 59, 74, 192, 202, 204–5, 207, 211,
104, 121, 134, 160, 168, 186, 216, 229
201, 241–2, 265, 275 giving up  148, 223, 229
army of  44, 48, 173, 239 see also possible  73, 205
soldiers, Belgian preventing  136, 166, 240 see also
control of  3, 241 defending actions
people of  111, 134, 230 Bremen 104
sovereignty of  201, 227 Britain  6–7, 20, 43, 104, 149–50, 186,
violation of neutrality of  19 201–3, 214, 227, 247–8,
Bellewaarde 79 252, 277, 279 see also
Below, General Fritz von  38, 98–9, 103, England
122–5, 130, 149, 160, 167, academics of  3–4
172, 205, 211, 242 army of see British army and British
Berchtesgaden  112, 250–1 Expeditionary Force (BEF)
Berlin  15, 104, 113, 242, 244, 248, 250, 261, artillery of see artillery, British
264, 269, 284 see also cavalry of see cavalry, British
Germany, government of defending actions of see defending
University of  13 actions, British
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

348 I n de x

Britain (cont.) under movements/

goals of see goals/objectives, British manoeuvres and offensives/
government of  194 attacks and retreats and
infantry of see also infantry, British withdrawals
lessons learned by see learning, attacks of see offensives/attacks, British
British battalions of  229
as of little concern to the Germans  20 command of  62, 89, 211, 214
offensives of see offensives/attacks, commanders in  154, 162, 164–5, 176,
British 194, 214 see also officers,
operations of see operations, British British
people of  14 corps of  162
point of view of  2, 8 I  42, 44, 60
royal family of  11 II 44
Secretary of State for War  72 IV  62–3, 71
strategy of see strategy, British XIII 100–1
tactics of see tactics, British XIV 139
British Armies  155 defeats of/setbacks for see defeats/
First  71, 84, 156 setbacks, British
Second  183, 221 Directorate of Training  154–5
Third  156, 190, 210–12, 237 divisions of  89, 95, 100, 102, 132, 138,
XVIII Corps  162 148, 156, 170, 178, 180, 183,
Fourth  100, 122, 127, 130, 138, 234, 210–12, 219, 225, 228
236, 240 1st 71
Fifth  140, 174, 210–12 2nd 62
Reserve  127, 131, 138, 140 see also 4th 157
British Armies, Fifth 7th  52, 77
British army, the  62, 88, 97, 132, 151, 280 9th 163
see also British Expeditionary 18th 131
Force (BEF), the 21st 86
official history of  36, 42, 62, 67, 100, 24th 86
126, 152, 186 51st 140
British Expeditionary Force (BEF), the  Guards 193
1, 5, 41, 43, 45, 52, 61–2, Meerut 71
65–6, 81–2, 87–8, 93–4, 100, engagement of  40
102, 115, 125–6, 134, 143, 145, ideas about  4–5
152–5, 158, 162, 165–6, 178, General Headquarters of (GHQ) 
186–7, 189, 196, 202–3, 205, 154–5, 158, 162, 188, 214
213–14, 216, 219–20, 222, organization of  154
225, 228, 279–80, 282, positions held by  52 see also territory,
299n.28 see also Australian under British control
forces and British Armies regiments of
and Canadian forces Devonshire
and Indian forces and 2nd Battalion  72
New Zealand forces and London
Portugese forces and 1st/14th Battalions (‘London
Scottish forces and South Scottish’) 46
African forces Oxfordshire and Buckinghamshire
actions of  42, 58–9, 61–4, 73–4, 77–9, Light Infantry
86, 89, 98, 100, 141, 180, 205, 2nd Battalion  47
234, 240, 260, 266 see also Territorial 46
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x349

Worcestershire Belgian 48
2nd Battalion  45 British  48, 67, 72, 78, 86–7, 89, 100,
Royal Flying Corps  169, 173 102, 121–2, 128, 140–1, 159,
see also Royal Air Force 165, 170, 172, 174, 176,
(RAF), the 179–80, 183, 186, 189, 193,
size of  52, 81, 148 211, 219, 225, 234, 237,
state of  95, 213 303n.28, 304n.16, 314n.17,
study of  4–5 320n.6
successes of the  5, 62, 64, 280 see also Canadian 185–6
exploitation of success, civilian see civilians, death of
British and victories, British Entente  89, 134, 165
tying down of  47 French  25, 30, 48, 60–1, 75, 80, 89,
weaknesses of the  89 121, 141, 153, 163, 174, 186,
Bronsart von Schellendorff, Colonel 219, 234, 303n.28, 304n.16,
Bernhard  125, 129 314n.17, 320n.6
Broodseinde, Battle of  180–3, 258 German  20–2, 24, 28–30, 36, 39, 43,
Brooke, Rupert 48–50, 61, 67, 75, 78, 83, 89,
‘Peace’ 281 95–6, 103, 121, 125, 132, 141,
Bruges  112, 242 159, 165, 174, 176–7, 179–80,
Brusilov offensive, the  99, 104 184, 186–7, 193–4, 214, 216,
Brussels  2, 35, 41, 160, 243, 246 219, 225, 227–9, 234, 237,
Spanish Embassy in  246–7 242, 295n.7, 297n.13,
Bucharest 149 298n.19, 303n.28, 304n.16,
Bulgaria 240 314n.17, 320n.6
Bullecourt  158–9, 164 Russian 92
Bülow, Colonel-General Karl von  45, 58 cavalry 21–2
Butte de Warlencourt  139 British  44, 100, 157, 210
Byng, General Sir Julian  190–1, 193–4, French  36, 43
211, 214 German  22, 25, 36–7, 39–40, 144
Indian 191
Calais  202, 220, 225 Cavan, General the Lord  139
Cambrai  37, 124, 134, 145, 150, 156, 165, Central Powers, the  92, 106, 142, 165, 227
190–1, 204, 209, 217–18, 228, Chales de Beaulieu, General Martin  175
234, 237, 239–41, 247 Champagne  60, 66, 80, 84–6, 88–9, 94,
Cambrai, Battle of (1917)  189–91, 193–6, 152, 163, 304n.16, 314n.17
257, 327n.1 Champagne, First Battle of (1914–15)  60
Canadian forces  61, 157, 169 changes  8, 82, 89–90, 143, 155 see also
actions of  129, 156, 162, 185–6, 234, learning and warfare, new
237, 241 style of, adaptation to
Canadian National Monument/Visitor failure to make  181–2 see also
Centre, the  87, 157, 302n.13 warfare, new style of, failure
Canal du Nord  190, 239 to adapt to
Caporetto, Battle of (1917)  195, 197, 205 made by the British  178, 196
Carency  39, 73–5, 118 made by the French  188
Cassel  225, 229 made by the Germans  233
Castelnau, General Noël de  22–3, 36, managing 53 see also learning
39, 84 need for  3, 53, 90, 233
casualties  30, 48, 69, 73, 81–2, 95, 101, Charles I (King of England)  11
165, 186, 189, 233, 293n.1, Charmes gap, the  25, 27–8
298n.18 see also attrition Charteris, Brigadier-General John  194
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

350 I n de x

Chateau Gendebien see Mons, Chateau communication  21, 36, 53, 110, 179,
Hardenpont 183–4 see also coordination/
Chateau-Thierry Salient  231 cooperation and liaison
Chaulnes 138 British 166
Chemin des Dames  163, 165, 225, 228 breakdowns of/difficulties with  64,
Christmas Day  61, 111–12 72, 211, 214
Churchill, John (1st Duke of Entente  70, 158
Marlborough) 266 French 52
Churchill, Winston  4, 191 breakdowns of/difficulties with  75
Cirey  22, 50 German  21, 23, 76, 97, 115, 118, 129,
Citino, Robert M.  17 150–1, 196, 235, 294n.10
civilians 4 breakdowns of/difficulties with  21,
death of  96, 247, 284 24, 35–6, 94, 117, 138, 171,
mistreatment of  30–3, 247 179, 183, 185, 195, 238
see also war crimes cutting of  71, 157
resistance put up by  32–3 protecting lines of  18
Claer, General Eberhard von  64–6, 76, Compiègne  228, 235, 243
117, 263 concentration camps  253 see also Dachau
Cléry 218 conditions  61, 64, 87–8, 126, 175–7,
combined arms operations 183, 186–7, 189, 210–11,
Anglo-French 50 222, 234, 238 see also mud
British  72, 127, 131, 174, 196 and weather
French 73 conferences
German  22, 50, 125, 130, 132, 136, 146, Entente  212, 214
183, 193, 197, 206 French 168
Combles  127, 129, 131 German  124, 134, 148, 166, 177, 197,
command see also command systems/ 203, 205, 207
structures and under Congreve, Lieutenant-General
individual armies Walter 100
(i.e. German army, the, conservatism  13, 274
command of ) Contalmaison 102
difficulties faced by  50, 162 Convent of the Sacred Heart of
mistakes made by  50 Roehampton 252
nature of  8, 17, 101 coordination/cooperation  78, 80, 132
command systems/structures  53 see also Anglo-French  49, 88–9, 100–1, 130,
command and 212–14, 216, 239
responsibilities lack of  126, 163, 214
British  162, 166 British  163, 234
Entente 101 poor  63, 78, 176, 279
German  17, 33, 45, 103, 108, 113–18, German  151, 196, 209
125, 130, 146, 160–1, 208, poor  21–2, 43–4, 47, 78, 117, 183
217, 233, 258–61, 267–8, Côte 140 see Hill 145
277–8 Coucy-le-Chateau  111, 307n.8
flexible/decentralized  4, 108, 115, counter-attacks  65–6, 82, 90, 178
117, 260, 268, 277 British  45–7, 49, 94, 193, 195, 211
rigidity in the  167, 182–3, 226, Entente  1, 49
238, 268 French  45, 49, 92, 130, 204, 231, 237
strained  21, 117, 121, 133, 216–17, German  52, 66, 76, 86, 88, 98, 117–18,
223–4, 259–61, 264, 267–8 126, 128–9, 131, 134, 136,
Commonwealth, the  4, 62, 132 146, 157–8, 166, 171–2,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x351

174–5, 179–80, 185, 187, 190, British  101, 193–4, 202, 209, 211, 213,
192, 194–5, 241, 277 216, 218, 220
counter-battery fire  50 overrunning/destruction of  210–12,
Courcelette  127–9, 131 215–16
Courland 265 Entente  216–17, 222
Courtrai  112, 168, 181, 242 French  96, 101, 213, 216, 218, 224, 231
Cowan, Tony  311n.6, 313n.8 German  50–1, 59–60, 62–6, 70, 73,
Croisilles 210 79–82, 85–6, 88, 90, 94, 97–8,
Crozat Canal  211–12 100–1, 106, 117–18, 127,
Cuinchy 61 130–2, 134–6, 138–41,
145–50, 152, 156, 159–61,
Dachau  251, 253 165–6, 168–9, 171–2, 174–6,
Daily Mail (newspaper)  43 178–80, 182, 184, 229, 231–2,
Dandl, Otto Ritter von  273 233, 237–42, 246
Dardanelles, the  62 Forward Zone of  229
deadlock/stalemate  26–7, 30, 40, 48, holding of  72, 80, 83, 87, 94, 149,
195–6 see also lines/ 175–6, 185, 190, 192
fortifications, stabilization of overrunning /destruction of  87,
attempts to break  41–2, 68, 90, 94, 126, 135, 139–40, 145, 156,
101, 127, 156, 189, 207 158, 161–2, 166, 172, 176,
see also breakthroughs 179–80, 185, 190, 234, 236–7,
failure of  57, 80, 141, 195, 205 241–2
Debeney, General Marie-Eugène  216, 234 poor  62, 128, 156, 159–61, 168, 238
decision-making 196 see also mistakes strength of  64, 82, 90, 101, 135, 141,
and Rupprecht, Maria 149–50, 152
Luitpold Ferdinand, supplementary lines of  94, 135, 137,
decisions of 149–50, 169, 171, 174, 180
British 194 types of
poor 177 elastic and in depth  65–6, 90,
German  4, 65, 113–18, 146, 149–50, 98, 134–6, 140, 146–7, 160–1,
205, 236, 257–8, 260, 267–70, 166–7, 169, 172, 179–81,
281 184, 259
poor  4, 8, 27, 115, 226, 238, 268, 278 rigid and forward  65–6, 97, 134,
defeats/setbacks  50, 233, 282 see also 148, 179–81, 238
retreats and withdrawals defending actions  80, 90, 194, 206, 278
Austro-Hungarian 68 see also defences
British  44–5, 72–3, 87, 99, 210–12, 222, British  46, 49, 65, 194, 210–11, 213, 216,
224, 228 218, 222–3, 225, 228–9, 231
Entente  70, 94, 187 Entente  48, 49, 216–17
French  23–5, 50, 85, 202 French  24, 28, 36–7, 39–40, 49, 96, 212,
German  2–3, 25, 50–1, 57, 64, 66, 74, 214–16, 218, 225, 228, 231
85–6, 95, 126, 130, 141, 145, German  23–6, 30, 37, 39–42, 45, 48,
149, 156–60, 162, 180, 183, 59, 63–6, 71, 73–5, 78–80, 83,
185, 191, 195, 207, 225, 229, 85–8, 90, 94, 98, 123, 126,
232, 234–7, 240–3, 258, 282 128–9, 131, 134, 136–7,
see also Germany, defeat of 140–1, 146, 148–50, 152, 156,
Italian 195 158–9, 161, 163–4, 166, 171,
defences  90, 206 see also defending 174–6, 179–80, 183–5, 187,
actions and lines/ 190–2, 194, 209, 229, 232,
fortifications 234, 236–40, 242, 277
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

352 I n de x

defensive lines see lines/fortifications efficiency 261

Delville Wood  122–4 Eichhorn, Colonel-General Hermann
demobilization 243 von 18
democracy  272, 274–5 Einem, General Karl von  58
demography 12 Eisenhower, General Dwight  253
desertion/mutiny  242–3, 297n.6 Eisner, Kurt  244–5
Dieuze 31–2 engineers/sappers
‘digging in’ see also trenches British 85
British  43–4, 64 French  71, 84
French 84 German 82
German  36, 43, 47, 59, 62, 136 England  14, 248 see also Britain
Dinant 30 Entente/Allied Powers, the  34, 69, 79,
diplomacy  7, 57, 233, 270 90, 93–4, 101, 123, 129, 142,
discipline 148, 188–9, 203, 208, 213,
French  42, 188 225, 232, 237, 239, 252–3,
German  2, 33, 113, 147, 150, 187 275, 279, 282 see also Britain
disease/illness  27, 61, 96, 229, 246 and France and Russia and
disengagement 29 United States, the
diversions 99 command of the  204, 224–5
British 100 demands of  241
German  94, 204 offensives of see offensives, Entente
Dixmuide  45, 47 resistance of  57, 96, 214, 216–17
doctrine  159–60, 226, 237 strength of  49, 69
codification of  146, 180 under-estimation of  50, 213,
disagreements about  97, 101–2, 136 275, 278
dissemination of  53, 97–8, 117, 154, 166 tying down of  47
see also memoranda/manuals Épehy 239
formulation of  53, 146, 155, 180, 182 Epinal 24–5
see also learning equipment  97–8, 121, 127
published  146–7, 154 British  156, 191
Dohna-Schlobitten, General Alfred French 137
Graf zu  182, 260 German  132–3, 135, 144–5, 160, 169
Dommes, Lieutenant-Colonel Ertman, Thomas  267
Wilhelm von 23 Essen 12
Douai  37–8, 42, 73, 88, 96, 111, 157, 161, Estaires 40
231, 242 Ettal Abbey  252
Douai Plain, the  61, 79, 162 explanations  50, 66, 195, 229, 280
Douaumont  92, 139 see also blame and learning
Doullens  209, 212, 214, 225 and reports
Dubail, General Auguste  21–3, 50 exploitation of success  65, 80, 196
Duffy, Christopher  102, 131 British  154, 158
Dunkirk 220 lack of  64, 157, 162, 172
French 100
East Prussia  22 lack of  153
Eastern Front, the  47–8, 57–9, 68, 78, 91–3, German  24, 78, 212, 228
99, 104–5, 142, 202, 208 lack of  195, 216, 224
Ebert, Friedrich  244
economics 248 Fabeck, General Max von  44–7, 76, 91,
Écurie 61 114, 297n.13
Edmonds,Brigadier-General Sir Falkenhausen, Colonel-General Ludwig
James  49, 101 von  106, 121, 152–3, 157, 160
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x353

Falkenhayn, General Erich von  30, 41, army of see French Armies and French
57–8, 64, 69, 91–2, 99, 104–7, army, the
114–16, 124, 160, 260, 267, artillery of see artillery, French
274–5, 277 cavalry of see cavalry, French
actions of  36, 44–5, 47, 49, 51, 68, 71, defending actions of see defending
76–8, 82, 85, 103, 105, 116, actions, French
145, 270, 273 goals see goals/objectives, French
command of  97, 104 infantry of see also infantry, French
opinions/ideas of  47, 66, 86, 94–5, 97, intelligence of see intelligence, French
134, 136, 238 lessons learned by see learning, French
opposition to  103–6, 269–70 Northern  18, 45
plans of  41–2, 44–5, 47, 93–4, 104, offensives of see offensives/attacks,
141–2 French
promotion of  35 people of  31–3, 96–7, 134
reputation of  57–8, 102 society of  172
resignation of  106–7, 270 tactics of see tactics, French
and Rupprecht  44–5, 76–7, 91, 93–5, territory of  20 see also territory,
97, 102–4, 106–7, 116, 273, controlled by the French
275, 281, 308n.13 Franco-Prussian War  20, 32
Falls, Cyril  151, 165–6 Franz Ferdinand (Archduke of
Fasbender, General Karl von  40, 76, Austria-Hungary) 15
85, 263 Franz Josef (Habsburg Emperor)  262
Faupel, Wilhelm  129 Frederick the Great (King of Prussia)  36,
Fayolle, General Émile  25, 100, 123, 126, 250, 261
130, 153, 165, 214 Freiburg-im-Breisgau military archives  7
Festubert  52, 60, 77, 222, 224 Frelinghien 220
Ficheux 85 French, Field Marshal Sir John  41, 61–2,
Field War Schools  147 64, 72, 89
firepower  49–50, 145 see also artillery French Armies
Flandern-Stellung see Flanders line First  21–4, 216, 234
Flanders  41, 45–6, 48–9, 57–8, 61, 70, 75, Second  22–3, 36
112, 116, 164, 168, 172–3, 176, Sixth  100, 126, 130, 138
183, 186–7, 190–2, 201–5, Tenth  79, 84, 231, 303n.28
207, 218–20, 225–6, 228, actions of the  60, 73–4, 78, 85, 87–8,
239–41, 258, 266, 282, 327n.1 123, 129
Flanders line  169, 171 Groupe d’armées du Centre
Flanders Ridge  224 (GAC) 84
Flaucourt Plateau  37, 102 French army, the  19, 41, 52, 83–4,
Flers 126–9 88–9, 93–4, 97, 100, 102, 126,
Flers-Courcelette, Battle of (1916)  142, 151, 153, 168, 187–9, 194,
126–9 205, 214, 219, 279–80, 282
Flesquières  190–5, 206, 210–11 see also French Armies
Foch, General Ferdinand  23, 38–9, 80–2, actions of  18, 58–9, 71, 78–9, 95,
86–7, 97–8, 129, 137, 142, 98, 212, 243 see also under
153, 212, 214, 231, 239–40, movements/manoeuvres and
243 offensives/attacks and
actions of  126, 143, 212, 228, 236, 240 retreats and withdrawals
France  2–3, 7, 18, 30–4, 48, 57, 59, 61, 75, attacks of see offensives/attacks, French
82, 89, 104, 134, 145, 150, caution of the  22–3
201–2, 214, 219, 241, 247–8, Chausseurs Alpins  38, 42
252, 275, 279 command of  21, 151, 153, 214
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

354 I n de x

French army, the (cont.) Gerbéviller 30

commanders in  50, 214 see also Gerke, Captain Ernst  230
officers, French German Armies  146
corps of  100 First  52, 91, 103, 122–3, 125, 128–9,
VIII 52 132, 139, 145, 147, 158
XX Corps (‘Iron Corps’)  23, 37, actions of the  35, 126
98–100 command of the  58, 103, 106
divisions of the 39th  98 Second  58, 98–9, 102–4, 123–5, 129,
criticism of  22 132, 136, 167, 190–2, 204,
defeats of/setbacks for see defeats/ 206, 214–18, 223–4, 234–5,
setbacks, French and under 239, 243
French Armies actions of the  205, 209–13
divisions of  27, 79, 100, 130, 148, 163, command of the  103, 106, 233
174, 219, 225 splitting up of  103
70th 25 stripping of divisions from the  44
77th 74–5 Third 58
Moroccan  74–5, 79, 302n.13 Fourth  41–2, 45, 47, 58, 116, 168,
Grand Quartier Général (GQG)  52–3, 171–3, 175, 177, 179–83,
80–1, 90 186–7, 205, 239, 266,
headquarters of see French army, the, 297n.13
Grand Quartier Général actions of the  43–5, 47, 204, 222
(GQG) defeats of/setbacks for the  44, 180,
official history of  8 297n.13
organization of  154 weakness of the  47, 240
resistance put up by  36–8, 40, 96, 214, Fifth  17–18, 25, 29, 58
275, 278–9 Sixth  1, 15, 18–19, 22, 30, 39, 41, 47, 52,
size of the  148 58, 60–1, 70, 73, 76, 79, 82–7,
study of  5 90, 95, 103, 106, 115–17, 121,
successes of  21, 279–80 see also 123, 152, 158, 163–4, 167, 172,
exploitation of success, 205–6, 220–1, 229, 239, 260,
French and victories, French 263, 295nn.7 and 13,
Territorial troops  37–8 298n.19, 303n.28
tying down of  18–19, 23, 46–7, 104 actions of the  21, 23, 25–6, 28, 30,
unrest within  172 35–7, 39, 42–7, 86, 88, 94,
Fresnoy 164 117, 153, 204, 211, 213, 222
Fricourt  37, 100 battle reports of the  66
Fromelles 71 Bavarians in the  18–19, 21–5,
Frouard 18 28–9, 39, 46, 50, 85, 94,
Froyennes 230 96, 111, 164 see also under
Fuchs, General Georg  149 German Armies, Sixth,
corps of the
Gallwitz, General Max von  58, 103, brigades of the
123–4, 129 Landwehr 18
Gardner, Nikolas  299n.28 chief of staff of the  17–18,
Gavrelle 158 160, 223
Gebsattel, Lieutenant-General Ludwig command of the  17, 21, 45, 76–7, 87,
von  25, 117, 258 94, 106, 114–18, 160
Gegenangriffe 136 commanders of the  25, 38, 42, 52,
Gegenstoße 136 64, 87, 91, 97, 118, 160
George V (King of Great Britain)  14, 191 corps of the  18, 47, 52, 118
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x355

I Bavarian  14, 35, 37–9, 70, strength of  47–8, 61, 95, 229
88, 263 stretching of the  24
I Bavarian Reserve  40, 73, 76 stripping of divisions from the  44,
5th Bavarian Reserve 47–8, 93, 99, 103
Division 73 successes of the  23–4, 40, 44–5, 51,
II Bavarian  35, 37, 39, 45, 222 see also victories,
224, 266 German
III Bavarian  25, 39, 51, 60, supplies of  40
167, 170 Württembergers in the  46
IV Prussian  38, 42, 85 Seventh  18–19, 111, 122, 228, 295n.7
VII  64, 66, 78, 117 actions of the  21, 23, 26
XIV Reserve  35, 37–9, 60 command of the  58, 106
XXI 35–6 Twelfth 91
Guard  39, 86, 88 Seventeenth  205–6, 209, 211–12,
defeats of/setbacks for the  25, 28, 214–16, 231, 237, 242–3
30, 37, 39–40, 48, 50, 64, 66, Eighteenth  204–6, 209, 211–13,
75, 96, 157, 161, 282, 297n.13, 215–17
298n.19, 303n.28 German army, the  4, 16, 31, 58, 68, 106–7,
deployments of the  33–5, 51 121, 142, 144–8, 167, 181–2,
difficulties facing the  40, 47–8, 64 186–8, 195–6, 207, 233, 235,
divisions of the  18, 153 see also 240, 242, 246, 251, 257,
Germany Armies, Sixth, 260–1, 270, 276–80 see also
stripping of divisions German Armies
from the and German Army Groups
5th Bavarian Infantry actions of  31–4, 49, 145, 148, 150, 192,
14th Bavarian Infantry 219 see also under defending
Regiment 28 actions and movements/
Bavarian cavalry  18 manoeuvres and offensives/
headquarters of the  17–19, 35, 46, attacks and retreats and
62, 64, 84–5, 90–1, 110–11, withdrawals
114, 117, 294n.10 administration of see administration,
intelligence 52 German
memoranda issued by  66 artillery regiments of  13 see also
mission(s) of the  19, 24–5, 27–9, artillery, German
34–5, 40–2, 51, 76, 86, 93 attacks of see offensives/attacks,
organization of the  18, 21 German
Prussians in the  18, 38, 111 battalions of  144, 238
records of  7, 67 Bavarian formations of  7, 18, 31, 49,
reinforcement of  75–6, 152 128, 139, 176, 262–4 see also
reserves of the  99, 153 German Armies, Sixth,
resources of  29, 93, 117, 121 Bavarians in the
Rupprecht’s command of the  17, Bavarian Lifeguard Regiment  228
29–30, 33, 38, 42–5, 51–2, 58, as cautious  49
76, 83, 85, 87–8, 91, 97, 106, cavalry regiments of  13 see also
110–11, 114–18, 263, 273, 282 cavalry, German
see also orders, German, 1st Heavy Cavalry Regiment  14
issued by Rupprecht Chief of the General Staff of  17–18,
and Rupprecht, as a 24, 35, 57, 70, 104, 115, 266–7
commander chiefs of staff of  17, 47, 124–5, 148,
staff of the  17–18, 21, 91, 266 172–3, 234–7
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

356 I n de x

German army, the (cont.) 195th Infantry  184

command of  4, 7, 12, 18, 21, 30, 44–5, Saxon 175
47, 57–8, 75–7, 80, 94, 97, ethos of  32
103–8, 113–18, 124–5, 134, General Staff of  18, 52, 108,
159, 161, 167, 172, 175, 181–2, 181, 201, 226, 229, 259
195–6, 201, 207–8, 215–17, see also Germany, Chief of
223–4, 226, 233, 235, 238–9, the General Staff of
241, 257–61, 266–8, 277, 282 histories of  8 see also official histories,
see also authority and German
command systems, German ideas about  4–5
and conferences, German infantry brigades of see also infantry,
and decision-making, German
German and Oberste 7th Infantry Brigade  14
Heeresleitung (OHL) Jagdstaffel 140
commanders in  17–18, 33, 42, 47, 58, 91, left wing of  18, 21, 24, 58
124, 146, 149, 159, 164, 167, mobilization of  16, 19
169, 171–3, 177, 179, 181–4, official history of  67, 102, 152, 181, 187,
192, 194, 196, 223–4, 235, 234, 292n.18
239, 241, 246, 258–62, 266–8 operation of  261
see also officers, German organization of  17–18, 76–7, 94, 103,
communications of see 106–8, 113, 115, 144, 146, 148,
communication, German 173, 187, 233, 275
corps of the  33, 91, 125, 173, 193 plans/order of battle of  18–19, 65
XVII 103 problems with  74, 94, 115–16, 129, 167,
Alpenkorps 77 181–3, 187, 207–8, 221, 226,
decline of  234, 237–8, 240–1 233, 238, 258, 261, 266–8,
defeats of/setbacks for  5, 39–40 see also 278–80
defeats/setbacks, German Prussian Guard  47
and under German Armies records of  7–8
and Germany, defeat of regiments of  68, 144, 159
defending actions of see defending 8th Bavarian  161
actions, German 89th Grenadier  52
deterioration of  5 Landwehr 74
divisions of  68–9, 78, 80, 99, 103, reputation of  2, 4, 69, 150, 260,
124–5, 142, 144, 148, 152, 268, 279
156, 167, 171, 173, 176–7, 179, right wing of  18, 21, 24, 29, 35–6, 41
181, 183, 187, 190, 192, 202, study of  4–5, 7, 17
210, 215–16, 219, 224, 227, successes of  47 see also exploitation
231, 238 of success, German and
3rd Marine  237 under German Armies and
4th Guards  180 victories, German
5th Bavarian Reserve  74 training operations of see training
6th Bavarian Reserve  71, 297n.13 operations, German
13th Westphalian  71 tying down of  141
14th Infantry  60, 157 weaknesses of  4–5, 8, 94, 229, 241–3,
15th Infantry  138 260, 268, 276, 280
27th Württemberg  158 German Army Groups  58–9, 145,
111th 233 147, 159, 167, 172, 205, 207,
117th Infantry  86 218, 233, 239–40, 242, 246,
119th 181 259, 267
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x357

von Boehn  233, 237, 241–2 cavalry of see cavalry, German

Crown Prince Rupprecht  1, 59, Chancellery of  33
106, 121, 125, 129, 131, 135, Chancellor of  16, 57, 104, 144, 201,
149, 152, 167, 173, 177, 179, 227, 241–4, 266, 271, 273–5
184, 187, 203, 206–7, 217, 219, see also Baden, Max von and
221–2, 224–5, 232–3, 237, Bethmann Hollweg,
239, 241, 261, 281 Theobald von and Hertling,
actions of  64, 171, 185, 193, 205, Georg
209, 219, 225 constitution of  251, 253, 261, 265–6
chiefs of staff of  113–14, 137, defeat of  3, 8, 106, 221, 233, 235, 240–4,
205, 260 246, 253, 272, 276, 279–80,
command of  76, 115, 121, 125, 282–3 see also defeats,
172–3, 246 German
composition of  58, 205 Foreign Office of  33, 227
defeats of/setbacks for  191, 224, goals of see goals/objectives, German
236, 241 government of  12, 15–16, 104–6, 201,
difficulties for/problems facing  133, 227, 240, 242–4, 248, 250–1,
138, 183–7, 217, 229, 236, 253, 261–2, 264–6, 269–71,
238, 260 273, 275–7
headquarters of  59, 71, 110–12, 125, Imperial Headquarters  104, 106–7
129, 184, 196, 202, 206, 210, Imperial Naval Cabinet  106
215, 227, 229–31, 236, 240 infantry of see also infantry, German
records of  7 intelligence of see intelligence,
Rupprecht’s command of  58, 73, German
114–17, 121, 123, 129, 151, Kaiser of  12, 17 see also Wilhelm II
153, 171–3, 177, 180, 183–5, (Kaiser)
187, 191–4, 209, 215, 233, leadership of  264–8, 271–2, 277–8
246, 257, 260–1, 263, 281 see also Germany army,
size of  229 command of the
staff of  129, 138, 173, 185, 260 mistakes made by the  3, 265–6,
stripping of divisions from  59, 192, 270–1, 275, 278 see also
228, 231 mistakes, German
successes of  60, 222 lessons learned by see learning,
Crown Prince Wilhelm  58, 93, 147, German
206–7, 215, 225, 228 Military Cabinet of  105–6, 114, 208
Gruppe Arras  156–7 offensives of see offensives, German
Gruppe Dixmuide  184 operations of see operations, German
Gruppe Souchez  156 plans of  19, 93, 158 see also Germany
Gruppe Vimy  156 army, plans/order of battle
Gruppe Wytschaete  173, 184–5 of the
Gruppe Ypern  182 point of view of  2, 8
Heeringen 58 policy of see policy, German
Germany  3, 12–13, 43–4, 48, 69, 92, 110, preparations of see preparation,
144, 205, 247, 252–3, 257, German
261–2, 264–5, 270–2, 274–9, as a republic  244, 248, 253
283 revolution in  2, 5, 243–6, 282
army of see German Armies prospect of  201, 227–8, 233, 240,
and German army 272, 276
and German Army Groups royal family of  227, 240, 249, 251, 262,
artillery of see artillery, German 266
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

358 I n de x

Germany (cont.) 139, 141–2, 151, 157, 162–3,

society of  4, 13, 184, 261–2, 272, 274–6 166, 183, 186, 191, 193, 197,
see also public, the 214, 219, 225, 239
(re)organization of  144, 244, 275, 283 actions of  46, 64–5, 86, 95, 109, 176,
unrest within  2, 5, 166, 201, 233, 240, 194, 211, 234
243–51, 262, 264, 272, 276 command of  60, 62
strategy of see strategy, German opinions of  49, 64, 87, 100, 102,
tactics of see tactics, German 188, 222
territory of  18 see also territory, under Ham 211
German control Hamel  37, 216, 229
unity of  261–2, 265, 269 Hamilton, General Sir Ian  305n.15
War Ministry of  35, 58, 105, 117, 250, 274 Hamilton, Lady  305n.15
Gestapo, the  252–3 Hammond, Bryn  191
Geyer, Captain Hermann  197 ‘Harnier Circle’, the  252
Gheluvelt  44–5, 47 Harris, J.P.  188–9
Gheluvelt Plateau  173–6, 180 Hart, Basil Liddell  4, 191
Ghent  241, 243 Havrincourt  190, 193, 239
Gießler Heights  86–7, 91, 95, 114 Hazebrouck  202–5, 222–3, 225, 229
see also Pimple, the Hedin, Sven  46, 109
Gildemeister, Dr  104 Heeringen, General Josias von  21, 58,
Ginchy  37, 79, 122–3, 128 117, 122
Givenchy  58, 60–2, 88, 222, 224 Heinrich of Bavaria (Rupprecht’s
Givenchy-lès-La-Bassée  60, 78, 220 cousin) 282
goals/objectives  278, 282 Heinrich, Prince of Bavaria
British  171, 174, 178, 188, 282 (Rupprecht’s son) 252–3
French  30, 282 Held, Heinrich  251
German  18–19, 30, 41, 43, 45, 58, 91, Hermann Line, the  239, 241
193, 201–2, 205, 209, 213, Hertling, Georg  271, 273, 275
218–19, 221, 225, 227 Heye, Colonel Wilhelm  239
frustration of  30, 48, 232, 270, 282 High Wood  122–3, 127, 213
Gommecourt 100 Hill 70  85–6
Gorlice-Tarnow  69, 78 Hill 119  79
Gough, General Sir Hubert  77, 131, Hill 140  87–8, 157 see also Hill 145 and
138, 140, 174, 176–7, 183–4, Pimple, the
211, 214 Hill 145  157–8, 302n.13
Gouzeaucourt  211, 217 Hindenburg, Field Marshal Paul von  57,
Grand Couronné de Nancy  24, 28 104, 106–7, 116, 124–5, 135,
Grand Cross, the  44 145, 147, 160, 167, 172, 201,
Groener, General Wilhelm  243 237–8, 240, 243, 257, 266–7,
Gueudecourt  127, 131 270, 274, 277
Guillemont 123 actions of  58, 105, 144, 147, 240, 243,
Guise River  243 246, 265, 273
opinions of  106, 144
Haber, Fritz  69 and Rupprecht  124–5
Habsburgs, the  11, 68, 262 Hindenburg Line, the  125, 135, 143,
The Hague  98 148–52, 156, 159, 161, 190,
Hague Conventions, the  32 193, 232, 237, 239–41, 263,
Haig, General later Field Marshal Sir 307n.8 see also Operation
Douglas  1, 42–4, 65, 71–2, ALBERICH
84–5, 89, 101, 126, 129, 137, Hirschberg, Captain Anton von  230–1
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x359

Hitler, Adolf  5, 45, 248, 249–53, 284, French  61, 232

292n.15 German 226
Hoéville 25 intelligence  22, 164, 278 see also
Hohenborn, General Adolf Wild reconnaissance and spies
von 105 British  95, 132, 151, 161, 188, 194, 205
Hohenzollerns, the  244, 249, 272, 275 poor/sketchy  102, 162, 166, 188,
Horne, General Sir Henry  156 194
Horne, John  30 Entente  148, 151
Hull, Isabel  31–2 French  19, 225
Hulluch 84–6 poor/sketchy  75, 79
humanitarianism 150 German  7, 52, 61–2, 70, 80, 84, 98,
humanity  6, 96 113, 123, 140, 151–2, 168,
‘Hundred Days’ campaign  1 172–3, 186–7, 196, 212, 229,
Hungary 265 239, 269, 278
Hutier, General Oskar von  205, 209, 211, mistakes of  22, 49
213–14 poor/sketchy  27, 36, 49
hygiene  70, 98 Iron Cross, the  24, 44
Hynes, Samuel  284 Isar River  244
Italy  14, 77, 201, 206, 225, 252–3, 273
Independent Social Democrats, the  army of  189, 195
244
India 14 see also British Expeditionary Jews 248 see also anti-semitism
Force (BEF), the, corps of, Joffre, General Joseph  22, 30, 61–2, 77,
Indian 79, 80–1, 96, 126, 153
Indian forces  44, 60, 62, 191 actions of  21, 24–5, 27–8, 36, 50, 87,
individuals  5–7, 49–50, 160, 181 89, 97–9
industry/industrial production  12, 96, plans of  60, 84
98, 121 Julius Caesar  36, 282
British  65, 72, 101
German  69, 121, 144 Kahr, Gustav von  248, 250–1
infantry  50, 80, 97, 132 see also soldiers Kiel 243
British  63, 72, 127, 132, 143, 166, 170, Kindermord 44
175, 178, 190, 196, 210 Kitchener, Field Marshal Lord
artillery support for  63, 72, 151, 170, Herbert  61, 72, 86, 89
174–5, 183, 234 Kluck, General Alexander von  35,
French  26, 73 91, 114
artillery support for  90 Koblenz 264
German  26, 28, 30, 39, 49, 65, 68, Krafft von Dellmensingen, Konrad  17–18,
130, 132, 136, 143–4, 152, 39, 44, 51, 69, 91, 109,
160–1, 164, 169, 183, 193, 113–14, 205, 260, 262, 269,
197, 221, 237–8 297n.6
actions of  21–2, 43, 49, 228 actions of  19, 23
artillery support for  22, 50–1, 183, diary of  37–9, 49
197, 206 see also opinions/ideas of  20, 23–4, 29, 42, 65,
coordination/cooperation, 68, 262
German, combined arms orders given to  29, 35
lack of  49, 231 and Rupprecht  113–14, 205
initiative  41, 103, 183 transfer of  77
British 154 Krafft von Dellmensingen, Leopold  269
Entente  148, 188, 232 Kramer, Alan  30
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

360 I n de x

Kriegsakademie see Bavaria, War College of British  4–5, 52, 62, 65, 97, 131, 134, 143,
Kuhl, General Hermann von  91–5, 98–9, 153–5, 178, 194, 196–7,
104, 110, 113, 115, 122, 129, 279–80, 299n.28 see also
136, 138, 145, 147, 151, 169, ‘learning curve’
172–3, 175, 178, 182–5, changes in manner of  155
202–3, 205–6, 209, 212, inadequacies of  72, 82, 89, 162,
214–16, 221–3, 232, 239, 165–6, 176
242–3, 257, 259–61 French  50–2, 81–2, 89–90, 96–7,
actions of  137, 148, 168, 190, 196, 202, 143, 153
233, 242–3, 257 inadequacies of  165
diary of  173, 191, 234 German  50–2, 65, 82, 90, 97, 102–3,
opinions/ideas of  144, 158, 171, 176–7, 136, 145–7, 159–60, 166–7,
192–3, 204, 208, 216, 220, 227, 172, 193–4, 196–7, 226, 229,
229, 236–7, 240–1, 259, 261 313n.8
and Rupprecht  114–15, 137, 228, 263 inadequacies of  5, 51, 117, 182–3,
Kühlmann, Richard von  227 226, 238, 258–9, 278, 280
‘learning curve’  4–5, 65
La Bassée  39–43, 48, 85 Lee, Celia  305n.15
La Bassée Canal  58, 61, 79, 85, 202, 204 Lee, John  305n.15
La Fère  210 Leeb, Captain Wilhelm von  206, 217,
La Folie  87–8 221, 241
La Malmaison  188–9 Leigh Fermor, Patrick  5
Labyrinth redoubt, the  74–5, 78 Leipzig War Crimes Trials  32, 248
Laffert, General Maximilian von  171 Lens  39, 61, 85, 152, 204, 236
Lagarde 21 Lerchenfeld, Hugo von  104, 261
Lambsdorff Colonel Gustav von  77, Lesboeufs 131
86–7, 91, 114, 260 Leutstetten 253
Landsberg, Mr. see Rupprecht, Maria liaison see also communication and
Luitpold Ferdinand, in coordination/cooperation
disguise Anglo-French 49
Landsturm, the  230 German 224
Langemarck  44, 47 liberalism/social progressivism  13, 32
Langemarck, Battle of (1917)  176 Lihons  37, 138
Langle de Cary, General Fernand de  60 Lille  39–40, 42, 62, 71, 96, 112, 150, 242
Laon  111, 125 lines/fortifications
Le Cateau  192 British  193, 204, 209, 213, 216, 220
Le Hamel  216 breaking of  45, 47
Le Sars  138 gaps in  211–13, 216
Le Transloy  130 restoration of  45, 47
leadership  214, 266 Entente  62, 186, 218
Entente 74 breaking of  49
French 214 restoration of  49
German  196, 264, 266 see also Germany, French  213, 216, 243
leadership of German  24, 30, 42, 66, 72–3, 77, 81–2,
poor  181–3, 195, 217, 237–8, 266 84, 88, 90, 94, 126–7, 130,
learning  4–5, 8, 50–3, 61–2, 82, 96–7, 134–5, 137, 139, 149–50, 152,
196–7, 280–1, 284 see also 156, 161–2, 169–71, 174, 178,
changes and conferences and 180, 192, 234, 236, 238–40
explanations and see also Hermann Line, the
improvement and reports and Hindenburg Line, the
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x361

breaking of  62, 65–6, 127, 145, 157, bearing of  125
166, 180, 236, 238, 240–2 command of  167, 223–4, 235
gaps in  126, 242 memoirs of  194, 207, 235, 240
holding of  66, 152, 158, 166, 192, opinions/ideas of  134, 136, 140, 144,
239–40 182, 185, 192, 197, 201, 203,
moving of  125, 135, 148–50, 152, 214, 220, 229, 232, 271
169, 184, 236–7, 239 see also plans of  161, 187, 204–5, 212–13,
Operation ALBERICH 218–19, 221
plugging of gaps in  100 removal/resignation of  242–3
shortening of  135, 148–9 and Rupprecht  124–5, 132, 167, 191,
stretching/extending of  40 202, 208, 214, 217, 221, 223,
stabilization of  46 see also 226, 229, 235, 246, 258–9,
deadlock/stalemate 271, 275, 282
Linselles 46 Ludwig II (King of Bavaria)  11, 15
Linsingen, General Alexander von  47 Ludwig III (King of Bavaria, Rupprecht’s
literature  6, 284 father)  2, 11–12, 15, 27, 77,
Lithuania 265 107, 142, 208, 227, 241,
Lloyd, Nick  89 247–8, 262–3, 271–5, 283
Lloyd George, David  4 actions of  16, 245–6, 273
Lochow, General Ewald von  76–80, Luitpold (Prince of Bavaria)  11–12,
116, 260 27, 253
logistics  98, 226 Lunéville  25, 30
London  43, 98, 170, 191, 201 Luxembourg  11, 29, 235–6
Longueval 123 Lyncker, General Moritz von  105–6,
Loos  39, 84–6, 88, 94, 153, 162, 169 208, 271
Loos, Battle of  89 Lys River  42, 44, 62, 171, 202–3,
Lorette Spur, the  61, 73–4, 78, 93, 158, 213, 222
203, 211, 216
Lorraine  18–20, 22–3, 26–7, 29–30, machines/mechanism  6, 96, 121, 150,
38–9, 239, 247, 263, 282 195–6 see also industry
see also Alsace-Lorraine Mametz 100
Lorraine, Battle of (1914)  83 Mametz Wood  37, 102
Loßberg, Lieutenant-Colonel Fritz Mangin, General Charles  149, 231
von  45, 103, 128, 160, 164, Mann, Thomas  244
166, 172–3, 180–2, 184, 232, Manonviller (Fort)  24
236, 241–2, 259, 263 manpower  121, 136, 171, 186
Louvain 30 British  69, 81, 132, 171, 210, 212, 279
Löwenstein, Prince Alois  109 see also soldiers, British,
Ludendorff, General Erich  57, 104, number of
106–7, 116, 121, 124, 132, Entente  123, 148
135, 145, 147, 150, 160, 171–3, German  68, 95, 117, 123, 125, 132, 135,
175, 202, 205, 208, 212–13, 144, 148, 150, 165, 171, 173,
215–17, 222–7, 236–40, 257, 184, 195, 208, 210–11, 219,
259, 263, 266–8, 270, 275, 237–8, 240–1, 271 see also
277–8 soldiers, German, number of
actions of  105, 144, 153, 177, 191, 193, 203, Marchais 111
206–9, 211–12, 214–16, Marcoing 193
218–19, 222, 224, 228, 232, Maria Therese (Princess of Austria,
234, 235, 240–2, 246, 249–50, Rupprecht’s mother)  11,
258–9, 265, 271, 273, 276 208, 248
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

362 I n de x

Maricourt  37–8, 51 ‘Command in War and the General

Marie-Gabriele (Duchess in Bavaria, Staff ’ 147
Rupprecht’s first wife)  ‘Training Manual of Infantry’  197
15, 253 Menin 41
Marne, First Battle of the (1914)  27, Menin Road, the  44–7
29–30, 35, 38, 48–9, 57, Menin Road Ridge, Battle of
270, 275 (1917) 178–9
Marne, Second Battle of the (1918)  232, Méricourt 158
234–5 merit  261, 277
Marne River  21, 228–9, 231, 233, 270 Merville 40
Marne Salient, the  231 Messines 222
Marquillies 64 Messines, Battle of (1917)  171–2, 178,
Martinpuich 127–8 188–9, 260
Marwitz, General Georg von der  192, Messines Ridge  44, 46, 168–71, 206
194–5, 214, 234–5, 239 Méteren 229
März, Stefan  261 Metz  18–20, 25, 193, 195
Masnières  191, 217 Metz-en-Couture 193–4
Masurian Lakes, Second Battle of Meurthe  18, 239
(1915)  58, 68 Meuse River  35, 95, 239–40, 243–4
Materialschlacht see attrition Mézières  239, 243
matériel  186, 189 see also resources Michaelis, Georg  201
British 171–2 Micheler, General Joseph Alfred  123
German  171–3, 185 Middle East, the  14
Maxse, General Sir Ivor  131 mines  169, 172
Maxwell, Lieutenant-Colonel Frank  131 mistakes 50 see also learning
Mecklenburg, Grand Duke Adolphus British  86, 89, 127, 162, 164–6, 177, 194
Frederick VI of  109 Entente 165
Mediterranean, the  14, 62 French 25
memoranda/manuals German  3, 22, 24–5, 49, 52, 57, 94–5,
British 154 104, 116, 128, 171, 181–3,
‘The Attack in Position Warfare’  197 208, 238, 275, 278
‘Instructions for the Training of mobility  69, 135–6, 196, 211–12, 226
British Armies in see also warfare, mobile
France’ 155 modernity  2, 5–6, 257, 280, 283, 289n.4
‘Instructions for the Training of Moltke the Younger, General Helmuth
Divisions for Offensive von  17–18, 21, 24–5, 28–32,
Action’ 154 37, 45, 58, 115, 262–3
‘Instructions for the Training of dismissal of  35
Platoons for Offensive and Rupprecht  19, 29
Action’ 154 momentum  51, 98, 146, 188, 197
‘The Normal Formation for the see also initiative
Attack’ 154 British 192
French 153 German  44, 209, 213, 224
German  66, 94, 97–8, 148, 160, 197, Monaco, Prince of  111
202, 204, 238 Monchy-le-Preux  157–8, 206, 211
‘Collected Instructions for Trench Moncrieffe, Betty  305n.15
Warfare’ 146–7 Mons  110–12, 202–3, 209, 217, 244
‘Principles for Command of the Chateau Hardenpont  111, 307n.5
Defensive Battle in Trench Montdidier  36, 215–16, 228
Warfare’ 146–7 Monte Cassino  139
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x363

Montgomery, Field Marshal Sir Nagel, Major-General Karl von  160

Bernard 1 Namur 35
Mont St Quentin  130–1, 138, 212, 217 Nancy  24–5, 27–9 see also Grand
morale  96, 322n.8 Couronné de Nancy
British  177, 195 nationalism  248–9, 251, 262, 265
Entente  74, 93 NATO 307n.5
French  96, 168 Nazi Party, the  248–52
German  47, 70, 93, 123, 132, 134, 139, Netherlands  1–2, 244, 247, 297n.6
142, 149, 152, 172, 184, 211, Neuschwanstein 11
228, 237–8, 242, 276 Neuve Chapelle  44, 62–4, 71, 77, 82, 85
morality  6, 32–3, 144 see also Neuve Chapelle, Battle of (1915)  62–6,
responsibility, moral 117, 195
Morhange  22–3, 38 Neuville-Saint-Vaast  74–5, 78–9
Mormal Forest  242 Neuville-Vitasse 157
Morval  37, 130–1 Nevinson, C.R.W.
Moselle River  18–19, 27 ‘La Mitrailleuse’  6
motivation 4 see also under soldiers newspapers
Mount Kemmel  46, 205–6, 224, 228 British  43, 267
Mouquet Farm  122–3, 129, 131 French 111
movements/manoeuvres  49–51, 179, German 262
212, 279 see also retreats and New York Times (newspaper)  83
roulement and withdrawals New Zealand forces  236
British  41–2, 137, 211 Nicholas II (Tsar of Russia)  262
Entente 62 Nieppe Forest  222, 225
French  24, 27, 50, 87–8, 137 Nieuport 173
German  16, 29, 34–5, 75, 99, 116, ‘Night of the Long Knives’  251
123–4, 151, 177, 179, 209, Nivelle, General Robert  151, 153, 162–3,
212, 221, 231–2, 240, 279 165, 168
attempted outflanking  36, 40 No Man’s Land  65–6, 71–3, 77, 84–5,
of troops from the Eastern 128, 131, 164, 168–9, 298n.18
Front  187, 225, 227 Nomény 30
of troops to the Eastern Front  47–8, Notre Dame de Lorette  39, 46, 60, 75, 78
59–60, 68, 78, 93, 99, 103, see also Lorette Spur, the
141–2, 263 Noyon  60, 106, 122, 150, 209, 214, 216
mud  59, 88, 126, 176, 183 Nymphenberg Palace  253
Mulhouse 21
Müller, Admiral Georg Ober Ost see Eastern Front, the
Alexander von 106 Oberste Heeresleitung (OHL)  21, 23–6,
Munich  1–2, 11–12, 15–16, 18, 20, 27, 28–9, 44–5, 47, 52–3, 59, 64,
109, 112, 141, 208, 244–5, 250, 78, 83, 99, 103, 106, 110, 113,
253, 262, 264, 274 see also 116, 125, 131, 135, 144, 149–50,
Bavaria, government of 152, 161, 163, 168, 177–8, 181,
Bavarian Household Privy Archive  6 184–5, 192, 202–3, 214–15,
Bavarian State War Archive  7, 217, 221, 223, 225, 227, 235,
52, 145 237–8, 240–1, 257–8, 261,
Hotel Bayerischer Hof  17 266–8, 271, 274–5, 277
Leuchtenberg Palace  292n.1 actions of the  75–7, 80, 85, 106, 124,
Residenz (palace)  15, 272 140, 142, 144, 159–60, 179,
University of  13 186, 205, 216, 224, 229, 231,
Mussolini, Benito  284 233, 257–8, 263
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

364 I n de x

Oberste Heeresleitung (OHL) (cont.) frustration of  50, 78–9, 131, 136,
directives of the  24–9, 33, 35, 41–2, 141, 193
45, 47, 75, 106, 108, 145–7, stalling of  75, 79–80, 87, 193
205, 209, 215, 217, 223, 237, suspension of  60, 88–9
263 see also orders, German German  21, 23, 25, 27–30, 35–9, 43,
interference by the  45, 116, 197, 208, 45–6, 48–9, 52, 60, 68–70,
217, 223, 257–9, 266, 273 83, 86, 91–5, 101, 116, 123,
memoranda issued by  66 148, 150, 180, 187, 193–5,
requests made of the  73, 77, 85, 105, 118 201–26, 228–9, 231, 270
Rupprecht and the  23, 25–7, 29, 42, frustration of  49, 65, 94–5, 194–5,
44–5, 73, 75–7, 85, 93, 103, 215–16, 219, 225, 232, 271
106, 108, 110, 113, 115–16, manner of  51
118, 131, 140, 149–50, 152, stalling of  43, 46, 48, 95, 211, 213–16,
160, 166, 197, 205–6, 208, 219, 222–4, 228
215, 217, 226, 237, 241, 243 suspension of  47, 141, 193, 216, 218,
staff of the  23, 35, 37, 68, 115, 151, 202, 222, 257
217, 226, 260, 267 Russian  22, 263
offensives/attacks  80–1, 91–2, 94, officers  50, 98
134, 180, 188, 194, 197, 278 British  32, 49, 63, 67, 89, 100, 166, 194
see also counter-attacks and as inexperienced  89, 101, 162
exploitation of success training of  212
American  236, 239–40, 242–3, 252 French  22, 32, 49, 74
suspension of  240 sacking of  50
Belgian 239 German  32, 49, 57, 77, 98, 110–11, 115,
British  59, 62–4, 66–7, 71–4, 77–8, 81, 132, 147, 159, 167, 223–4,
84, 86, 88–9, 93, 95, 100, 262, 303n.28
122–3, 126–7, 130–1, 134, complaints of  47
137–40, 145, 154, 156–8, insufficiencies of  39, 43, 50, 181–2
162–4, 170–80, 183–5, relations between  21, 91, 110, 113,
187–92, 202, 229–30, 234, 117, 129, 223–4, 263–4
236, 239–43, 252, 260 training of  18, 136, 147, 212
frustration of  63, 67, 72, 86–7, 89, Oise River  35, 39, 58, 205, 233
100, 141, 175–7, 184–5, 187 Oostaverne line  170–1
stalling of  78, 128, 174, 187, 192, 234 Operation ALBERICH  149–52, 161, 172,
style of  178–80 204–5
suspension of  65, 88, 141, 234 Operation BLÜCHER  225, 228
Canadian  157, 162, 185–6, 234, Operation CASTOR and
237, 241 POLLUX 203
Entente/Allied  7, 29, 52, 74, 84–5, 88, Operation GEORG  203, 207, 215, 220
94, 98–9, 104, 118, 123–4, Operation GEORG I  205–7
126, 129–30, 134, 136–8, Operation GEORG II  205, 207
141–2, 145, 148, 152–3, 168, Operation GEORGETTE  217–18,
183, 234, 237–43 220–2, 224, 271
French  18, 21–5, 29, 37, 39, 51, 59–60, Operation GNEISENAU  228–9
66, 73–5, 78–80, 84–8, 90, Operation HAGEN  225, 229, 231–2
96, 98–100, 102, 114, 116, Operation KLEIN-GEORG  213, 220
118, 122–4, 126, 130–1, Operation MARNESCHUTZ-REIMS 
137–9, 152, 159, 162–3, 165, 229, 231, 237
174–6, 183–4, 232, 234, 236, Operation MARS  203, 205–7, 211,
239–40, 242–3 215–16
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x365

Operation MARS-NORTH  213 issued by Rupprecht  33, 37, 43, 51,

Operation MARS-SOUTH  210, 213 97, 113, 118, 153, 180, 192,
Operation MICHAEL  203–9, 212, 196, 209, 243, 247, 297n.6
215–22, 225, 257, 271 Otto (King of Bavaria)  11–12
Operation MICHAEL I  206 Owen, Wilfred  6, 284
Operation MICHAEL II  206
Operation MICHAEL III  206 Paris  29, 168, 201, 212, 214, 228
Operation NEU-GEORGE  225 Pasha, Zekki  111
Operation NEU-MICHAEL  225 Passchendaele  1, 185–6, 188, 192, 222
Operation RIDE OF THE Passchendaele, First Battle of
VALKYRIES  211, 213 (1917) 184–5
operations  3, 18, 53, 80–1, 90, 94, Passchendaele, Second Battle of
102, 134, 141, 183, 197, 202, (1917) 186
207, 242 Passchendaele Ridge  173, 180, 183, 185
‘bite-and-hold’ approach to  65, 179, patriotism 20
182, 189, 279 patronage  261, 277
British  4, 72, 162, 178–9, 182, 184, peace  58, 91, 125, 201, 241–2, 244, 271,
196–7 275–6 see also Armistice, the
ideas about  101–2, 154–5, 178, and truces
184, 279 asking for  240, 242–3
methodical  154, 189 compromise  270, 275
poor  89, 98, 166, 177 hopes of  57, 205, 208, 227, 235, 241
Entente  126, 137, 238 negotiation of  227, 241, 270, 275–6
French  60, 98, 143, 153, 155 rejection of  242
ideas about  51, 81, 89–90, 101–2, seeking of  69, 105, 201, 241–4, 270
153 Pérenchies Ridge  42–3
methodical  51, 73, 78, 90, 94, 153 Péronne  37, 102, 124, 126, 130, 136,
problems facing  51, 153 141, 153, 205, 212, 217,
German  3–4, 7, 98–9, 113, 115–17, 236–7
121, 124, 166, 182–3, 202, Pershing, General John J.  240
204, 206–8, 221, 236, 239, personalities  104, 117
258–9, 277 Pétain, Marshal Philippe  153, 214
ideas about  51, 134–6, 145, Philpott, William  132, 138
226, 278 Picardy  34, 98, 237
poor  49, 181, 224, 238, 268, 278 Pilckem Ridge, Battle of  175
problems facing  51, 183, 217, 279 Pimple, the  87–8, 157–8, 302n.13
Oppy 158 Pius XII (Pope)  252
orders see also memoranda planning
British 63–5 British  62, 72, 89, 170, 172, 234
‘Backs to the Wall’  222 Entente  74, 152
French 23 German  78, 93–4, 185, 192, 194,
German  4, 16, 29, 66, 117, 129, 180, 197, 201–3, 205–8, 212,
196, 203, 205, 212, 215, 217, 216, 226
242, 248 see also Oberste Plessen, General Hans von  105–6,
Heeresleitung (OHL), 208, 271
directives of the Plettenberg, General Karl von  47
confusion over  29, 217, 238 Ploegsteert  44, 46
given to Rupprecht  18, 25, 27, 29, Plumer, Sir Hubert  176, 178–9, 182–4,
35, 40–2, 75, 93, 106, 108, 189, 222, 279
124, 150, 217, 232, 237, 246 Poelcappelle, Battle of  183–4
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

366 I n de x

poison gas Prussians in the and German

British use of  85 army, Prussian Guard
German use of  69, 96, 130, 210, 270 War Ministry of  21
Poland  61, 111, 248, 265 Prussian army, the  17, 261 see also
politics  8, 77, 82, 101, 104, 112, 126, 134, German army, Sixth Army,
142, 163, 182, 188, 191, 203, Prussians in the
206–7, 222, 227, 244, 250–1, Przemyśl 68–9
257, 260, 266, 268–9, 273–5, psychological damage  142, 186
279, 283 public, the  272
instability in  5, 244, 247–8, 277 information withheld from  35
Polygon Wood  47, 173, 179 opinion of  184, 201, 276
Poperinghe 225
Portuguese forces  202, 220, 222 railway stations see trains/railways
Pour le Mérite 83 Rancourt 130–1
Pozières  38, 122 Rawlinson, General Sir Henry  63–5, 71,
practical concerns  33 100–2, 122, 128, 138, 156,
practice 21 234, 240
Prager, Major Karl von  138, 180, 215 reconnaissance see also intelligence
preparation 194 see also under artillery French 25
British  98, 162, 234 German  36, 47
French 98 aerial  36, 39, 62, 70, 130
German  20, 79, 161, 165, 196, 207–8, see also aeroplanes
211 Reichsarchiv (Potsdam)  7–8
Prior, Robin  72, 78, 132, 174, 188 Reichsland see Alsace-Lorraine
prisoners of war  248 Reims  106, 122, 152, 225, 228–9, 231,
captured by the British  64, 229, 237 236, 239
captured by the Entente  138, 242 reinforcements 194 see also roulement
captured by the Germans  60, 83, 87, British  218, 223
101, 158, 212, 219, 247, 267 lack of  101
Pritzelwitz, General Kurt von  87 Entente 152
progress see also goals/objectives French  46, 95, 212, 215, 218, 228
British  100, 127, 192 lack of  87–8, 241
lack of  87, 100, 128, 131, 141, German  28, 59, 64, 68, 75–6, 78, 82, 85,
162, 184 87, 91, 99, 103, 105, 118, 123–5,
Entente 240 128, 152, 160, 169, 175, 177,
lack of  141 179, 185, 190–1, 202, 211, 217,
French 227, 231–2, 234, 239, 261
lack of  128, 141 lack of  47, 102, 123, 185
German  68, 213, 218 religion  13, 261–2, 264
lack of  95, 141, 211, 213–16, 219 Catholicism  11, 13, 250, 264
propaganda  2, 21, 33–5 Protestantism  13, 264
Prussia  12, 252, 261–5, 274 see also Remarque, Erich Maria
East Prussia All Quiet on the Western Front 6
archives of  1, 6 Réméréville 28
army of see Prussian army, the reports/reviews  50, 53, 197 see also
and Bavaria  13, 18, 77, 150, 261–5, explanations and learning
272–4 British  191, 194, 197
culture of  13 French 50
people of  18, 91, 233, 262 see also German  26, 50–2, 62, 66, 90, 102,
German Armies, Sixth, 129–30, 145, 147, 160, 175,
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x367

177–8, 191, 193–7, 212, 243, Rhine, the  236–7

258–9, 313n.8 Richtofen, Captain Manfred von  140
reserves  98, 179 Riga  205, 265
British  49, 72, 86, 89, 95, 101, 152, 204, Roclincourt 85
207, 219 see also British Roeux 163
Armies, Reserve Röhm, Ernst  249, 251
Entente  49, 152, 228 Romania  106, 142, 149, 282
French  49, 86, 98, 152, 216, 225, 228 Rome 252–3
German  35, 40, 44, 63, 65, 71, 73, Rommel, Field Marshal Erwin  1
75, 79–80, 83, 85–6, 88, Rosenberg, Isaac  284
93, 98–9, 103, 116, 121, roulement
123, 126, 128, 135, 148, British 132
152–3, 157–8, 160–1, German  124–5, 136, 139, 171, 177
164, 166, 169, 171, 173, Roulers  43, 168, 186, 240
179–80, 183, 185, 190–1, Royal Air Force (RAF), the  231
205–6, 211, 215, 237, 240, Roye  36, 209, 214
243, 282 Roye-Noyon 209
freeing up of  149 Ruhr 12
shortage of  59, 70, 123, 131–2, 195, Rupprecht, Crown Prince of Bavaria  1,
240–1 5–6, 8, 20, 24, 28, 58, 60, 62,
tying down of  81 66, 71, 79, 83, 87, 91, 93–4,
use of  124, 146, 160, 192, 215, 225 123–4, 127, 133, 145, 148,
resources  98, 136, 145–6, 171–2, 189, 276, 150–1, 160, 169, 182, 195,
279 see also industrial 210, 212, 224, 231, 236,
production and matériel 244–6, 248–51, 253, 262, 273,
and supplies 275, 277, 280–4, 297n.6
British  122, 132, 162, 172, 212 actions of  17, 35–9, 52, 64–5, 70, 73, 77,
Entente  123, 132, 279 88, 117–18, 123–4, 147, 152,
French  60, 81, 132, 137 157, 173, 180, 185, 192–3, 196,
German  29, 105, 117, 121, 123, 131, 202–3, 216, 230, 233, 241,
144, 146, 175, 177, 208, 215, 243, 247, 250–3, 260, 266,
261, 276 270–4 see also orders,
lack/shortage of  132, 150, 161, 185, German, issued by
195, 208, 238, 271–2, 279 Rupprecht
responsibilities/duties  17, 23, 42, as commander of his own Army
71, 214 Group see under German
abdicating of  115 Army Groups, Crown
delegation of  21, 116, 140, 268 see also Prince Rupprecht
authority, delegation of as commander of the Sixth Army
responsibility 58 see also blame see under German Armies,
acceptance of  194 Sixth
moral  6, 31–2 activities/daily routine of  7, 108–13,
Rethondes 243 185, 230, 257, 281–2 see also
retreats see also disengagement and Rupprecht, Crown Prince
withdrawals of Bavaria, pastimes of
British  60, 205, 211–12 Army Group of see German Army
French  24, 38 Groups, Crown Prince
German  49, 129, 134, 149–52, 158, Rupprecht
180, 184, 186, 191, 236–7, bearing of  13, 108–9
241–3, 247 career of  2, 13–15, 108, 114, 260, 282
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

368 I n de x

Rupprecht (cont.) honouring of  24

character of  13–14, 27, 109–10, 117, and the Kaiser see Wilhelm II (Kaiser),
281–3 and Rupprecht
children of  15, 20, 27, 112, 235–6, 247, leave taken by  112–13, 141, 259–60
249, 252–3 see also Albrecht, life of  11, 13–15, 109, 282–4
Prince of Bavaria and marriages of  15, 247
Heinrich, Prince of Bavaria and the OHL see Oberste
and Luitpold, Prince of Heeresleitung (OHL),
Bavaria Rupprecht and the
as a commander  7–8, 14–15, 21, 25, 30, opinions/ideas of  20, 23–5, 29, 32–3,
33, 44–5, 52, 58, 73, 76–7, 85, 39, 42, 45–6, 51, 59, 66, 68–9,
88, 97, 108–10, 113–18, 76, 80, 88, 91, 93–4, 97,
121–4, 129–30, 140, 151–2, 102–4, 107, 123–5, 128, 140,
157, 171, 246, 248, 257, 158, 172, 175–6, 179, 185,
259–61, 263, 266, 268–9, 273, 190, 197, 204–6, 216, 221,
281–2 see also Germany 227, 234–5, 240, 262–3,
Armies, Sixth, Rupprecht’s 270–1, 275, 279, 281–4
command of the and impressions of the British  42–3,
German army Groups, 128, 143, 172, 175, 177, 212,
Crown Prince Rupprecht, 279–80
Rupprecht’s command of negative/critical  22, 25–6, 29, 41–2,
concerns of  78, 86 44, 66, 76–9, 94–5, 103, 150,
correspondence of  7, 27, 76–7, 106–7, 160, 167, 206–8, 223–4, 226,
142, 241 233, 241, 258–9, 265, 267
death of  253 observations  45, 60, 95, 128, 181–2,
decisions of  6, 275, 281–3 188, 192, 194, 229
as depressed  137–8, 259, 282 pessimistic  69, 91, 144, 183, 197,
diary of  7, 16, 19, 25, 51, 67, 73, 77–8, 209, 220, 228–9, 235, 270,
125, 141, 164, 167, 171–2, 175, 274, 276, 283
181–3, 186–7, 191, 205, 208, orders given to see orders, German,
217, 222, 229–31, 263, 279, given to Rupprecht
283–4 orders issued by see orders, German,
manuscript  32, 191 issued by Rupprecht
published  32, 69, 191, 283, 308n.13 pastimes of  13–14, 48, 109, 111–12
in disguise  1–2, 5, 247, 277 and politics  8, 269, 273–5, 281, 283
education of  13–14 promotions of  14–15, 106, 257
and Falkenhayn see Falkenhayn, reputation of  43, 83, 281
General Erich von, and as a royal  12–13, 15, 273–4, 282
Rupprecht staff/officers serving under  17–18,
family of  5, 11–12, 15, 27, 109, 112–13, 108–11, 113–17, 122–4,
208, 235–6, 246–9, 253 128–9, 138, 147, 180, 260,
flight of  246–7 281
headquarters of  46, 90, 96, 110–12, travels of  14–15
117–18, 236, 269 see also as a victim  5–6, 283–4
German Armies, Sixth, as a (possible) war criminal  5, 31–3, 83,
headquarters of the and 247–8, 297n.6
German Army Groups, Russia  2, 7, 15, 18, 57, 68–9, 92–3, 99,
Crown Prince Rupprecht, 104–5, 150, 187, 201, 206,
headquarters of 209, 221, 225, 227, 263, 279
health of  106, 235, 246, 253, 260, 282 attacks of see offensives/attacks, Russian
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x369

dropping out of the war  204, 265, 271 Siegfried Stellung see Hindenburg
revolution in  202, 246 see also Line, the
Bolsheviks Sixt von Armin, General Friedrich  42,
Russo-Japanese War, the  50 85, 168–9, 171–3, 177, 182,
184, 246
sacrifice 44 Smith-Dorrien, General Sir Horace  44
Sailly-Saillisel  130, 138, 141 socialists/socialism  2, 244, 249, 251, 276
St Avold  18, 20 Soissons  80, 152, 228, 231
St Eloi  59, 61–2 soldiers  6, 50, 53, 59, 61, 73, 284
St Julien  176 see also engineers/sappers
St Michael  203 and infantry
St Mihiel Salient  239 Austro-Hungarian 92
St Pierre Divion  140 Belgian  173, 186
St Quentin  35–6, 129, 203–5, 237, 240 British  42–3, 67, 69, 83, 86, 100, 127,
St Quentin Canal  191, 217, 239 165, 177, 186, 297n.6
Salzburg 245 actions of  40
Sambre River  243 as inexperienced  81, 89
Santerre Plateau  123, 129, 139 number of  69, 81, 123, 132, 143, 210,
Sarajevo 15 212 see also manpower, British
Sarrebourg 22–4 performance of  166
Sarrebourg, Battle of (1914)  31 quality of  49, 74, 132, 143, 175, 177,
Sassoon, Siegfried  6, 284 203, 212
Saxony  233, 264–5 training of  153, 166
Scarpe River  85, 157, 161–2, 206, 211, weariness of  162
215, 239 Canadian 169
Scheldt River  244 Entente  29, 49, 74, 151, 279
Schenk von Stauffenberg, Major French  18, 27, 42, 61, 69, 88, 100,
Berthold  31, 252 186, 228
Schlieffen, Field Marshal Alfred von  37, discipline of  42
41, 114, 182, 279 number of  123
Schlieffen Plan, the  18–19, 32, 41, 270, quality of  49, 74, 96, 100, 143, 203
276, 279 resilience of  37–8
Schmidt von Knobelsdorff, General weariness of  79, 87, 177
Konstantin 17–18 German  33, 43, 61, 68, 74, 93, 96, 98,
Schulenberg, Colonel Friedrich von  117, 148–9, 185–7, 213, 237–8,
202 242, 280, 303n.28
Schwaben Redoubt  131, 140 actions of  30–3, 222, 242–3, 297n.6
science  12, 96–8, 154–5, 191 see also Bavarian  16, 18, 128
technology and warfare, conditions for  59
scientific as inexperienced  43–4
Sclieffen Plan, the  7 motivation of  4, 237, 242
scorched earth policy  144, 150 number of  47, 57, 68, 95, 123,
Scottish forces  229 132, 210–11, 229, 238
Sedan  147, 243 see also manpower, German
Seille River  23 performance of  43
Selle River  241–2 poor  45, 49, 237–8
‘September Programme’, the  264 Prussian 18
Serbia  15, 265 quality of  125, 132, 143, 147, 177,
Serre  37, 100, 140 227, 237, 242, 279
Sheffield, Gary  73, 100, 139, 188 reserve see reserves, German
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

370 I n de x

soldiers (cont.) failure/failings of  3, 57, 223–4, 238,

training of  4, 147, 153, 197 266, 268, 278
weariness of  38–40, 92, 132, 142, lack of  143, 265
187, 214, 223–4, 226, 229, Stuarts, the  11
233, 243, 272 see also Stuff Redoubt, the  140
roulement suicide 20
movement of see movements/ supplies/supply lines  98
manoeuvres see also ammunition
suffering of  30, 61, 142, 184, 186–7, British  101, 137, 215, 219
282, 284 see also casualties Entente 240
Solesmes 147 French  137, 219
Solly-Flood, General Arthur  154–5 difficulty of delivering  75
Somme, the  1, 36, 93, 97–9, 101–4, 106, German  121, 140, 196, 213, 217, 242–3,
116–18, 121–5, 130, 132, 134, 281–2
136, 139, 141–3, 145, 147–9, shortage of  40, 195, 214, 229,
151–4, 156–8, 163–5, 172–4, 271–2
176, 183–5, 187, 189, 203, surprise  90, 96, 196
205, 213, 218, 222, 263, 270, British  62, 102, 180, 190
327n.1 French  74, 237
Somme, Battles of the  35–8, 103, 123 German  23, 193–4, 208, 211, 228
Somme River  34, 37, 98, 103, 121–2, Switzerland  21, 59, 205, 227, 253
129, 138–9, 212, 216, 224–5,
233–4, 236 tactics  3–4, 18, 48–51, 65, 72, 81–2, 90,
Souchez  39, 73–5, 78–9, 83, 85–7, 90, 152 95–6, 102, 127, 134, 183,
South African forces  229 188–9, 197, 202, 207, 217, 279
‘soviets’  2, 243–6 British  82, 98, 101–2, 127, 134, 143,
speed  80, 194, 204, 242 162, 162, 182, 188–9, 191,
British  131, 137 196, 236, 238
Entente 137 development of  154–5, 178, 189,
French 137 196–7, 238, 279
German  24, 32, 136, 175, 196, 204 good  49, 131, 182, 191, 229, 238
see also Schlieffen Plan, the poor  42, 89, 101, 128, 153, 166,
spies 98 see also intelligence 177, 188
Stalingrad  138, 182 Entente 129
Stapff, Major Max  172, 260, 261 French  52, 82, 89, 96, 98, 101–2,
Stegemann, Hermann  227 155, 238
Stein, General Hermann von  23, 37 good  49, 153, 238
Steinberg, Jonathan  327n.1 poor  22, 50
Stenger, General Karl  31, 248 German  3–4, 66, 82, 90, 97–8, 115–16,
Stetten, General Otto von  224, 266 134–6, 145–6, 166, 177,
Stevenson, David  189, 201, 214 179–80, 182–4, 193, 196–7,
Strasbourg 21 202–3, 206–7, 210–11, 214,
strategy  8, 48, 80, 117, 134, 188, 202, 216, 222, 224, 226, 236,
207, 279 258–9, 277, 279
British  188, 191 development of  52, 134–6, 140,
Entente 142 145–7, 167, 196–7
German  3, 18–19, 57, 93, 104–5, 113, failure/failings of  49, 51, 145, 159,
148, 202, 207, 221, 266, 269, 179–84, 214, 238, 268, 278
274, 279 see also Schlieffen Tamines 30
Plan, the tanks  97, 127, 132, 157, 190–1, 238
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x371

British  97, 140, 190–1, 196, 211, Entente 70

229, 234 French 70
D-17 (‘Dinnaken’)  127 German  62–3, 70, 73–4, 79, 82, 86,
Mark  IV 170 90, 102, 129, 134, 146, 169,
French  97, 231 180, 184
Tappen, Lieutenant-Colonel Trescault 194–5
Gerhard  23–4, 28–9, 49, Trônes Wood  102
99, 267 troops see soldiers
taxation 12 truces 61
technology  97–8, 154, 191, 196 Turkey 111
terrain  39, 146, 165, 213, 222
territory 221 U-boats  143–4, 149–50, 168, 227, 241,
under British control  52, 224 271 see also warfare,
under French control  20 submarine
under German control  18, 74, 80, 134, United States, the  7, 14, 83, 201, 208, 241
141, 146, 168, 186, 218, forces of  229, 236, 239–40, 242–3, 252–3
227–8, 241, 264, 272, 276 joining the War  165, 187, 202
ceding of  148, 176, 180, 185, provocation of  144
227, 241 d’Urbal,Victor  73–4, 78–80, 84, 87
Thaer, Lieutenant-Colonel Albrecht
von  185, 226 Valenciennes 242–3
Theatinerkirche St Kajetan  253 Valentini, Rudolf von  105–6
Thiepval  37, 100, 122–3, 127, 129, 131–2, Vatican, the  262
138, 140 Vaux 139
Thourout 186 Verdun  29, 58, 92, 125, 137, 139, 141–2,
time  148, 158, 165 145, 153, 177, 187–8, 202–4,
The Times (newspaper)  43 207, 239, 270
Tincourt 218 Verdun, Battle of  68, 92–7, 101, 103–4,
Tirol, the  77, 252, 273 124, 149
Toul 25 Versailles 235
Tournai  230–1, 267 Vesle River  232
training  4, 50, 53, 147, 197 see also victories/successes  80, 233, 282
practice see also exploitation of
British  127, 153–5, 166, 212 success
Entente 74 British  62, 85, 141, 145, 157–8, 160,
German  19, 147, 197, 212 see also Field 162, 171, 179–80, 185, 188,
War Schools 191–2, 229, 279–80
of officers see under officers French  73, 86–7, 100, 126, 141, 145,
of soldiers see under soldiers 149, 153, 175, 270, 279–80,
trains/railways  16, 20, 35–6, 41, 46, 51, 82, 282
98, 124, 137, 144, 150, 152, German  24, 40, 57, 60, 78, 90, 149,
168, 179, 186, 203, 211, 152, 183, 185, 187, 194,
214–15, 218, 221, 231, 210–12, 222, 228 see also
239–40, 263 under German Armies
traps  22–4, 41–2 victory  49, 144, 265
trenches  26, 51, 60–3, 75, 81, 90, 117, 146 conditions needed for  3, 92, 144
see also ‘digging in’ and elusiveness of  30, 38, 48–9, 57, 65, 87,
warfare, trench 91, 128, 144, 148, 162, 186,
British  70, 72, 195 189, 205, 270, 276, 279
‘Smith-Dorrien’ 63 Entente  143, 240–1
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

372 I n de x

victory (cont.) trench  26, 49–51, 61–2, 68, 116–18,

giving up hopes of  47, 57, 69, 91, 95, 146 see also trenches
206, 229, 236–7, 241, 270–1, Warneton line  169, 171
274–5, 283 Washington 201
hopes for  57, 90, 143–4, 202, 204–5, Waterloo 243
212, 229, 270–1, 276 Watson, Alexander  142, 187
Victory (Nelson’s ship)  16 weather  46, 59–60, 62, 88, 126, 131, 136,
Vienna 15 see also Austria-Hungary, 138, 175–6, 183–4, 206,
government of 209–11, 234, 238
Villalobar, Rodrigo, Marquis de  1–2 Der Weltkrieg see official histories, German
Villers-Bretonneux  218, 224, 229 Wenninger, General Karl von  27–8, 264
Villers-Cotterêts Forest  231 Western Front, the  3, 8, 26, 47, 57–8, 61,
Vimy Ridge  39, 60–1, 70, 73–5, 78–9, 68–9, 75, 91, 93, 99, 103, 124,
85–8, 94, 100, 152, 156, 158, 141–2, 148, 150, 182, 187–8,
162, 164, 168, 203, 206, 216, 195–6, 204, 216, 219, 228,
302n.13 240, 263, 269, 280, 282
violence  5–6, 30, 32–4, 59–60, 69, 247–8, establishment of  26, 30
270, 284 nature of  51, 53, 59
Voltaire 266 portrayal of  6
Vosges Mountains  18, 24 Wetzell, Major Georg  202, 204–5, 215,
235, 239, 260–1, 267
Wagner, Richard  11, 14 Wilhelm I (Kaiser)  13
war crimes  30–4, 83, 96, 247–8, 297n.6 Wilhelm II (Kaiser)  2, 34, 46, 58, 69, 71,
see also atrocities and Leipzig 76–7, 84, 93, 104–7, 111, 115,
War Crimes Trials 160, 172–3, 207–8, 227–8,
and Rupprecht see Rupprecht, Maria 240, 242–3, 257, 261–3,
Luitpold Ferdinand, as a 265–7, 269, 271, 273–5, 277,
(possible) war criminal 292n.15
war guilt  227 actions of  16, 24, 83, 107, 149, 224, 244,
warfare 275 246, 260, 267
aerial 173 see also aeroplanes opinions/ideas of  25, 271, 283
ideas about  6, 73, 80, 87, 96 and Rupprecht  44, 46, 83, 105, 109,
laws of  32–3, 144 see also Hague 267, 269
Conventions, the Wilhelm (Crown Prince of Prussia)  25,
mobile  22, 45, 61, 69, 116–17, 127, 141, 31, 58, 93, 124–5, 202, 244
172, 212 see also mobility as a commander  17–18, 29, 58,
nature of  8, 26, 51, 53, 63, 96, 117–18, 147, 207, 215, 224, 260, 282
184, 270, 278, 280, 282 see also German Armies,
new style of Fifth
adaptation to  50, 52–3, 61, 116–18, willpower 50
145–6, 155, 178, 188, 196, Wilson, Trevor  72, 78, 132, 174, 188
280–1 see also changes and Wilson, Woodrow  201, 240, 242, 275
learning Fourteen Points of  201, 275–6
dealing with  49–53, 96, 98, 282 withdrawals see also disengagement and
failure to adapt to  51, 63, 102, 179, retreats
181–3, 278, 280 British 211
shock of  49–50 French 27
scientific  96–7, 153–5 German  60, 78, 103, 125, 149, 151–2,
semi-open 166 169, 171, 191, 232, 234, 236
submarine  144, 149, 266, 270–1 Wittelsbachs, the  11, 236, 244–9,
see also U-boats 253, 265
 OUP CORRECTED PROOF – FINAL, 19/12/17, SPi

I n de x373

women  96, 111 Xylander, Major Rudolf von  25, 91,

World War I 111–12, 114, 160
blame for see war guilt
course of  265 Yeats, W.B.  284
experience of  2–3, 5–6, 8, 30, 59, 61, Oxford Book of Modern Verse 284
142, 177, 186, 282, 284 Ypres  1, 42–9, 61, 69, 76, 79, 168, 171,
ideas about  2, 4–6, 69, 277, 283–4 174, 177, 187–9, 202, 211,
stereotypical  5–6, 73, 92, 101 220, 222, 224–5, 239–40
impact of  283 Salient  58–9, 168, 171, 177, 204, 229
length of  69 Ypres, First Battle of (1914)  42–9, 57,
nature of  233, 257, 283 68, 223
start of  15–16, 269–70 Ypres, Second Battle of (1915)  69, 194
understanding of  3–6 Ypres, Third Battle of (1917)  187, 189, 192
World War II  1, 252–3, 284 Ypres–Comines Canal  171, 220
Württemberg  15, 46, 158, 264 Yser River  44–5, 173
Württemberg, Duke Albrecht of  41,
43–4, 47, 58, 124 Zabecki, David  204, 218
Wytschaete  46, 58, 168–70 Zonnebeke  70, 179