(Ebook) Holomorphic Operator Functions of One

Variable and Applications: Methods from Complex

Analysis in Several Variables by Israel Gohberg,
Jürgen Leiterer (auth.) ISBN 9783034601252,
9783034601269, 3034601255, 3034601263 download
https://ebooknice.com/product/holomorphic-operator-functions-of-
one-variable-and-applications-methods-from-complex-analysis-in-
several-variables-1146352

Explore and download more ebooks at ebooknice.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

(Ebook) Biota Grow 2C gather 2C cook by Loucas, Jason; Viles, James

ISBN 9781459699816, 9781743365571, 9781925268492, 1459699815,
1743365578, 1925268497

https://ebooknice.com/product/biota-grow-2c-gather-2c-cook-6661374

(Ebook) Matematik 5000+ Kurs 2c Lärobok by Lena Alfredsson, Hans

Heikne, Sanna Bodemyr ISBN 9789127456600, 9127456609

https://ebooknice.com/product/matematik-5000-kurs-2c-larobok-23848312

(Ebook) SAT II Success MATH 1C and 2C 2002 (Peterson's SAT II Success)

by Peterson's ISBN 9780768906677, 0768906679

https://ebooknice.com/product/sat-ii-success-
math-1c-and-2c-2002-peterson-s-sat-ii-success-1722018

(Ebook) Master SAT II Math 1c and 2c 4th ed (Arco Master the SAT
Subject Test: Math Levels 1 & 2) by Arco ISBN 9780768923049,
0768923042

https://ebooknice.com/product/master-sat-ii-math-1c-and-2c-4th-ed-
arco-master-the-sat-subject-test-math-levels-1-2-2326094
(Ebook) Cambridge IGCSE and O Level History Workbook 2C - Depth Study:
the United States, 1919-41 2nd Edition by Benjamin Harrison ISBN
9781398375147, 9781398375048, 1398375144, 1398375047

https://ebooknice.com/product/cambridge-igcse-and-o-level-history-
workbook-2c-depth-study-the-united-states-1919-41-2nd-edition-53538044

(Ebook) Operator Algebras, Operator Theory and Applications (Operator

Theory: Advances and Applications) by Maria Amélia Bastos, Israel
Gohberg, Amarino Brites Lebre, Frank-Olme Speck ISBN 9783764386832,
9783764386849, 3764386835, 3764386843
https://ebooknice.com/product/operator-algebras-operator-theory-and-
applications-operator-theory-advances-and-applications-2004600

(Ebook) Introduction to Complex Analysis in Several Variables, Second

Edition by Volker Scheidemann ISBN 9783031264276, 3031264274

https://ebooknice.com/product/introduction-to-complex-analysis-in-
several-variables-second-edition-49767850

(Ebook) Functions of One Complex Variable I by John B. Conway ISBN

9780387942346, 0387942343

https://ebooknice.com/product/functions-of-one-complex-
variable-i-48784272

(Ebook) Complex Analysis - Several Complex Variables and Connections

with PDE Theory, and Geometry by Peter Ebenfelt, Norbert Hungerbuhler,
Joseph J. Kohn, Ngaiming Mok, Emil J. Straube ISBN 9783034600088,
3034600089
https://ebooknice.com/product/complex-analysis-several-complex-
variables-and-connections-with-pde-theory-and-geometry-1875538
Operator Theory: Advances and Applications

Vol. 192

Editor:
I. Gohberg

Editorial Office:
School of Mathematical Sciences V. Olshevski (Storrs, CT, USA)
Tel Aviv University M. Putinar (Santa Barbara, CA, USA)
Ramat Aviv A.C.M. Ran (Amsterdam, The Netherlands)
Israel L. Rodman (Williamsburg, VA, USA)
J. Rovnyak (Charlottesville, VA, USA)
B.-W. Schulze (Potsdam, Germany)
F. Speck (Lisboa, Portugal)
Editorial Board: I.M. Spitkovsky (Williamsburg, VA, USA)
D. Alpay (Beer Sheva, Israel) S. Treil (Providence, RI, USA)
J. Arazy (Haifa, Israel) C. Tretter (Bern, Switzerland)
A. Atzmon (Tel Aviv, Israel) H. Upmeier (Marburg, Germany)
J.A. Ball (Blacksburg, VA, USA) N. Vasilevski (Mexico, D.F., Mexico)
H. Bart (Rotterdam, The Netherlands) S. Verduyn Lunel (Leiden, The Netherlands)
A. Ben-Artzi (Tel Aviv, Israel) D. Voiculescu (Berkeley, CA, USA)
H. Bercovici (Bloomington, IN, USA) D. Xia (Nashville, TN, USA)
A. Böttcher (Chemnitz, Germany) D. Yafaev (Rennes, France)
K. Clancey (Athens, GA, USA)
R. Curto (Iowa, IA, USA)
K. R. Davidson (Waterloo, ON, Canada) Honorary and Advisory Editorial Board:
M. Demuth (Clausthal-Zellerfeld, Germany) L.A. Coburn (Buffalo, NY, USA)
A. Dijksma (Groningen, The Netherlands) H. Dym (Rehovot, Israel)
R. G. Douglas (College Station, TX, USA) C. Foias (College Station, TX, USA)
R. Duduchava (Tbilisi, Georgia) J.W. Helton (San Diego, CA, USA)
A. Ferreira dos Santos (Lisboa, Portugal) T. Kailath (Stanford, CA, USA)
A.E. Frazho (West Lafayette, IN, USA) M.A. Kaashoek (Amsterdam, The Netherlands)
P.A. Fuhrmann (Beer Sheva, Israel) P. Lancaster (Calgary, AB, Canada)
B. Gramsch (Mainz, Germany) H. Langer (Vienna, Austria)
H.G. Kaper (Argonne, IL, USA) P.D. Lax (New York, NY, USA)
S.T. Kuroda (Tokyo, Japan) D. Sarason (Berkeley, CA, USA)
L.E. Lerer (Haifa, Israel) B. Silbermann (Chemnitz, Germany)
B. Mityagin (Columbus, OH, USA) H. Widom (Santa Cruz, CA, USA)
Holomorphic
Operator Functions
of One Variable
and Applications
Methods from Complex Analysis in Several Variables

Israel Gohberg
Jürgen Leiterer

Birkhäuser
Basel · Boston · Berlin
Authors:

Israel Gohberg Jürgen Leiterer

Department of Mathematical Sciences Institut für Mathematik
Raymond and Beverly Sackler Humboldt-Universität zu Berlin
Faculty of Exact Sciences Rudower Chaussee 25
Tel Aviv University 12489 Berlin
69978 Ramat Aviv Germany
Israel e-mail: [email protected]
e-mail: [email protected]

2000 Mathematical Subject Classification: 47-02, 47A56, 47A68, 47B35; 30-02, 30E05;
32L05, 32L10, 32L20

Library of Congress Control Number: 2009927556

Bibliographic information published by Die Deutsche Bibliothek.

Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data is available in the Internet at http://dnb.ddb.de

ISBN 978-3-0346-0125-2 Birkhäuser Verlag AG, Basel - Boston - Berlin

This work is subject to copyright. All rights are reserved, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, re-use of
illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage
in data banks. For any kind of use permission of the copyright owner must be obtained.

© 2009 Birkhäuser Verlag AG

Basel · Boston · Berlin
P.O. Box 133, CH-4010 Basel, Switzerland
Part of Springer Science+Business Media
Printed on acid-free paper produced from chlorine-free pulp. TCF∞
Printed in Germany

ISBN 978-3-0346-0125-2 e-ISBN 978-3-0346-0126-9

987654321 www.birkhauser.ch
 Dedicated to the memory
of our friend, student and colleague

Georg Heinig
Contents

Preface xi

Introduction xiii

Notation xix

1 Elementary properties of holomorphic functions 1

1.1 Definition and first properties . . . . . . . . . . . . . . . . . . . . . 1
1.2 The maximum principle . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Contour integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 The Cauchy integral theorem . . . . . . . . . . . . . . . . . . . . . 11
1.5 The Cauchy formula . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.6 The Hahn-Banach criterion . . . . . . . . . . . . . . . . . . . . . . 15
1.7 A criterion for the holomorphy of operator functions . . . . . . . . 18
1.8 Power series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.9 Laurent series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.10 Isolated singularities . . . . . . . . . . . . . . . . . . . . . . . . . . 25
1.11 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2 Solution of ∂u = f and applications 29

2.1 The Pompeiju formula for solutions of ∂u = f on compact sets . . 29
2.2 Runge approximation . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3 Solution of ∂u = f on open sets . . . . . . . . . . . . . . . . . . . . 41
2.4 OE -cocycles and the Mittag-Leffler theorem . . . . . . . . . . . . . 43
2.5 Runge approximation for invertible scalar functions and the
Weierstrass product theorem . . . . . . . . . . . . . . . . . . . . . 44
2.6 OE -cocycles with prescribed zeros and a stronger version of the
Mittag-Leffler theorem . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.7 Generalization of the Weierstrass product theorem . . . . . . . . . 54
2.8 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
viii Contents

3 Splitting and factorization with respect to a contour 59

3.1 Splitting with respect to a contour . . . . . . . . . . . . . . . . . . 59
3.2 Splitting and the Cauchy Integral . . . . . . . . . . . . . . . . . . . 61
3.3 Hölder continuous functions split . . . . . . . . . . . . . . . . . . . 65
3.4 The splitting behavior of differentiable functions . . . . . . . . . . 72
3.5 Approximation of Hölder continuous functions . . . . . . . . . . . . 75
3.6 Example: A non-splitting continuous function . . . . . . . . . . . . 77
3.7 The additive local principle . . . . . . . . . . . . . . . . . . . . . . 82
3.8 Factorization of scalar functions with respect to a contour.
First remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.9 Factorization of Hölder functions . . . . . . . . . . . . . . . . . . . 89
3.10 Factorization of Wiener functions . . . . . . . . . . . . . . . . . . . 91
3.11 The multiplicative local principle . . . . . . . . . . . . . . . . . . . 93
3.12 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

4 The Rouché theorem for operator functions 97

4.1 Finite meromorphic Fredholm functions . . . . . . . . . . . . . . . 97
4.2 Invertible finite meromorphic Fredholm functions . . . . . . . . . . 101
4.3 Smith factorization . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.4 The Rouché theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 111
4.5 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

5 Multiplicative cocycles (OG -cocycles) 113

5.1 Topological properties of GL(E) . . . . . . . . . . . . . . . . . . . 114
5.2 Two factorization lemmas . . . . . . . . . . . . . . . . . . . . . . . 120
5.3 OE -cocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.4 Runge approximation of G-valued functions. First steps . . . . . . 125
5.5 The Cartan lemma . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.6 OG -cocycles. Definitions and statement of the main result . . . . . 131
5.7 Refinement of the covering . . . . . . . . . . . . . . . . . . . . . . . 133
5.8 Exhausting by compact sets . . . . . . . . . . . . . . . . . . . . . . 137
5.9 Proof of the main theorem in the case of simply connected open
sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.10 Runge approximation of G-valued functions. General case . . . . . 143
5.11 Proof of the main theorem in the general case . . . . . . . . . . . . 147
∞
5.12 OG (E) -cocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.13 Weierstrass theorems . . . . . . . . . . . . . . . . . . . . . . . . . . 153
5.14 Weierstrass theorems for G ∞ (E) and G ω (E)-valued functions . . . 156
5.15 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

6 Families of subspaces 159

6.1 The gap metric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.2 Kernel and image of operator functions . . . . . . . . . . . . . . . 172
6.3 Holomorphic sections of continuous families of subspaces . . . . . . 184
Contents ix

6.4 Holomorphic families of subspaces . . . . . . . . . . . . . . . . . . 185

6.5 Example: A holomorphic family of subspaces with jumping
isomorphism type . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
6.6 Injective families . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
6.7 Shubin families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
6.8 Complemented families . . . . . . . . . . . . . . . . . . . . . . . . . 206
6.9 Finite dimensional and finite codimensional families . . . . . . . . 209
6.10 One-sided and generalized invertible holomorphic operator
functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
6.11 Example: A globally non-trivial complemented holomorphic family
of subspaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
6.12 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216

7 Plemelj-Muschelishvili factorization 219

7.1 Definitions and first remarks about factorization . . . . . . . . . . 220
7.2 The algebra of Wiener functions and other splitting R-algebras . . 222
7.3 Hölder continuous and differentiable functions . . . . . . . . . . . . 230
7.4 Reduction of the factorization problem to functions, holomorphic
and invertible on C∗ . . . . . . . . . . . . . . . . . . . . . . . . . . 237
7.5 Factorization of holomorphic functions close to the unit . . . . . . 240
7.6 Reduction of the factorization problem to polynomials in z and 1/z 240
7.7 The finite dimensional case . . . . . . . . . . . . . . . . . . . . . . 242
7.8 Factorization of G ∞ (E)-valued functions . . . . . . . . . . . . . . . 245
7.9 The filtration of an opeator function with respect to a contour . . 251
7.10 A general criterion for the existence of factorizations . . . . . . . . 259
7.11 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

8 Wiener-Hopf operators, Toeplitz operators and factorization 269

8.1 Holomorphic operator functions . . . . . . . . . . . . . . . . . . . . 269
8.2 Factorization of G ω (E)-valued functions . . . . . . . . . . . . . . . 273
8.3 The space L2 (Γ, H) . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
8.4 Operator functions with values acting in a Hilbert space . . . . . . 287
8.5 Functions close to the unit operator or with positive real part . . . 291
8.6 Block Töplitz operators . . . . . . . . . . . . . . . . . . . . . . . . 297
8.7 The Fourier transform of L1 (R, E) . . . . . . . . . . . . . . . . . . 305
8.8 The Fourier isometry U of L2 R, H . . . . . . . . . . . . . . . . . 313
8.9 The isometry V from L2 (T, H) onto L2 (R, H) . . . . . . . . . . . . 317
8.10 The algebra of operator functions L(H) ⊕ L1 R, L(H) . . . . . . 320
8.11 Factorization with respect to the realline . . . . . . . . . . . . . . 324
8.12 Wiener-Hopf integral operators in L2 [0, ∞[, H . . . . . . . . . . . 325
8.13 An example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
8.14 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
x Contents

9 Multiplicative cocycles with restrictions (F-cocycles) 343

9.1 F-cocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
9.2 The main results on cocycles with restrictions. Formulation and
reduction to OD,Z,m . . . . . . . . . . . . . . . . . . . . . . . . . . 346
9.3 The Cartan lemma with restrictions . . . . . . . . . . . . . . . . . 347
9.4 Splitting over simply connected open sets after shrinking . . . . . . 352
9.5 Runge approximation on simply connected open sets . . . . . . . . 354
9.6 Splitting over simply connected open sets without shrinking . . . . 358
9.7 Runge approximation. The general case . . . . . . . . . . . . . . . 362
9.8 The Oka-Grauert principle . . . . . . . . . . . . . . . . . . . . . . . 365
9.9 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367

10 Generalized interpolation problems 369

10.1 Weierstrass theorems . . . . . . . . . . . . . . . . . . . . . . . . . . 369
10.2 Right- and two-sided Weierstrass theorems . . . . . . . . . . . . . . 371
10.3 Weierstrass theorems for G ∞ (E)- and G ω (E)-valued functions . . . 374
10.4 Holomorphic G ∞ (E)-valued functions with given principal parts of
the inverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
10.5 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378

11 Holomorphic equivalence, linearization and diagonalization 379

11.1 Introductory remarks . . . . . . . . . . . . . . . . . . . . . . . . . . 379
11.2 Linearization by extension and equivalence . . . . . . . . . . . . . 380
11.3 Local equivalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
11.4 A theorem on local and global equivalence . . . . . . . . . . . . . . 392
11.5 The finite dimensional case . . . . . . . . . . . . . . . . . . . . . . 394
11.6 Local and global equivalence for finite meromorphic Fredholm
functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
11.7 Global diagonalization of finite meromorphic Fredholm functions . 406
11.8 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411

Bibliography 413

Index 419
Preface

This is a book on holomorphic operator functions of a single variable and their ap-
plications, which is focussed on the relations between local and global theories. It is
based on methods and technics of Complex analysis of scalar and matrix functions
of several variables. The applications concern: interpolation, holomorphic families
of subspaces and frames, spectral theory of polynomials with operator coefficients,
holomorphic equivalence and diagonalization, and Plemelj-Muschelishvili factor-
ization. The book also contains a theory of Wiener-Hopf integral equations with
operator-valued kernels and a theory of infinite Töplitz matrices with operator
entries.
We started to work on these topics long ago when one of us was a Ph.D. stu-
dent of the other in Kishinev (now Cisinau) University. Then our main interests
were in problems of factorization of operator-valued functions and singular inte-
gral operators. Working in this area, we realized from the beginning that different
methods and tools from Complex analysis of several variables and their modifica-
tions are very useful in obtaining results on factorization for matrix and operator
functions. We have in mind different methods and results concerning connections
between local and global properties of holomorphic functions. The first period was
very fruitful and during it we obtained the basic results presented in this book.
Then World Politics started to interfere in our joint work in the new area. For
a long time the authors became separated. One emigrated to Israel, the other was
a citizen of East Germany, and the authorities of the second country prevented
further meetings and communications of the authors. During that time one of
us became more and more involved in Complex analysis of several variables and
finally started to work mainly in this area of mathematics. Our initial aims were
for a while frozen. Later the political situation in the world changed and after the
reunification of Germany the authors with pleasure continued the old projects.
During the time when our projects were frozen, the scientific situation
changed considerably. There appeared in the literature new methods, results and
applications. In order to cover the old and new material entirely in a modern form
and terminology we decided to write this book. As always happens in such cases,
during the writing new problems and gaps appear, and the material requires in-
clusion of additional material with new chapters containing new approaches, new
results and plenty of unification and polishing. This work was done by the authors.
xii Preface

We hope the book will be of interest to a number of large groups of experts in

pure and applied mathematics as well as for electrical engineers and physicists.
During the work on the book we obtained support of different kinds for
our joint activities from the Tel-Aviv University and its School of Mathemati-
cal Sciences, the Family of Nathan and Lilly Silver Foundation, the Humboldt
Foundation, the Deutsche Forschungsgemeinschaft and the Humboldt University
in Berlin and its Institute of Mathematics. We would like to express our sincere
gratitude to all these institutions for support and understanding. We would also
like to thank the Faculty of Mathematics and Computer Sciences of the Kishinev
University and the Institute of Mathematics and Computer Center of the Academy
of Sciences of Moldova, where the work on this book was started.

Berlin, Tel-Aviv, November 2008 The authors

Introduction

The book. This book contains a theory and applications of operator-valued holo-
morphic functions of a single variable. (By operators we always mean bounded
linear operators between complex Banach spaces.) The applications concern some
important problems on factorization, interpolation, diagonalization and others.
The book also contains a theory of Wiener-Hopf integral equations with operator-
valued kernels and a theory of infinite Töplitz matrices with operator entries.
Our main attention is focussed on the connection between local and global
properties of holomorphic operator functions. For this aim, methods from Complex
analysis of several variables are used. The exposition of the material appears in
style and terms of the latter field.

Multiplicative cocycles. Grauert’s theory. The theory of multiplicative cocycles

plays a central role in this book. It is a special case of the very deep and powerful
theory of cocycles (fiber bundles) on Stein manifolds (any domain in C is a Stein
manifold), which was developed in the 1950s by H. Grauert for cocycles with values
in a (finite dimensional) complex Lie group. This theory then was generalized into
different interesting directions. In 1968, L. Bungart obtained it for cocycles with
values in a Banach Lie group, for example, the group of invertible operators in a
Banach space.
One of the main statements of Grauert’s theory is a principle which is now
called the Oka-Grauert principle. Non-rigorously, this principle can be stated as fol-
lows: If a holomorphic problem on a Stein manifold has no topological obstructions,
then it has a holomorphic solution. This important principle was first discovered
in 1939 by K. Oka in the case of scalar functions.
For domains in the complex plane C, Grauert’s theory is much easier but still
not simple. It is even not simple for the case of cocycles with values in the group
of invertible complex n × n-matrices when no topological obstructions appear.
For operators in infinite dimensional Banach spaces, we meet essential diffi-
culties, which are due to the fact that the group of invertible operators in a Banach
space need not be connected. This becomes a topological obstruction if the domain
in C is not simply connected. So, for operator functions, the Oka-Grauert principle
is meaningful also for domains in C.
xiv Introduction

For the problem of Runge approximation, the Oka-Grauert principle claims

the following: Runge approximation of a holomorphic invertible operator function
by holomorphic invertible functions is possible if this is possible by continuous
invertible functions. From this it follows that such a Runge approximation always
holds when the domain is simply connected or the group of invertible operators is
connected. The latter is the case for the group of invertible operators in a Hilbert
space, and in particular, for the group of invertible complex n × n-matrices.
For simply connected domains, the proof of the Runge approximation theo-
rem for invertible operator functions is not difficult and can be obtained without
the theory of cocycles. We show this at the end of Chapter 2. For general domains
however, this proof is much more difficult (even in the case of matrix-valued func-
tions) and will be given only in Chapter 5 in the framework of the theory of
multiplicative cocycles.
A special type of multiplicative cocycles is given by two open sets D1 and D2
in C and an invertible holomorphic operator function on D1 ∩ D2 . For this type,
the following is proved:

0.0.1 Theorem. Let E be a Banach space, let GL(E) be the group of invertible
operators in E, let D1 , D2 ⊆ C be two open sets, and let A : D1 ∩ D2 → GL(E) be
holomorphic. Assume that at least one of the following two conditions is satisfied:

(i) The union D1 ∪ D2 is simply connected.

(ii) All values of A belong to the same connected component of GL(E).

Then there exist holomorphic operator functions Aj : Dj → GL(E), j = 1, 2, such

that
A = A1 A−1
2 on D1 ∩ D2 . (0.0.1)

If both topological conditions (i) and (ii) in Theorem 0.0.1 are violated, then
the assertion of Theorem 0.0.1 is not true. A simple counterexample will be given
in Section 5.6.2 for the case when D1 ∪ D2 is an annulus.
The following operator version of the Weierstrass product theorem (on the
existence of holomorphic functions with given zeros) is a straightforward conse-
quence of Theorem 0.0.1.

0.0.2 Theorem. Let E be a Banach space, let GL(E) be the group of invertible
operators in E, and let GLI (E) be the connected component in GL(E) which
contains the unit operator I. Let D ⊆ C be an open set and let Z be a discrete and
closed subset of D. Suppose, for each w ∈ Z, a neighborhood Uw ⊆ D of w with
Uw ∩ Z = {w} and a holomorphic operator function Aw : Uw \ {w} → GL(E) are
given. Further assume that at least one of the following two conditions is fulfilled:

(i) The set D is simply connected.

(ii) The values of each Aw , w ∈ Z, belong to GLI (E).
Introduction xv

Then there exist a holomorphic operator function B : D\Z → GL(E) and a family
of holomorphic operator functions Hw : Uw → GL(E) such that

Hw Aw = B on Uw \ {w}, , w∈Z.

The classical Weierstrass product theorem we get for E = C and Hw (z) =

(z − w)κw , κw ∈ N∗ .
There are also a “right-sided” and a “two-sided” version of Theorem 0.0.2.
Contents. The book consists of an introduction and eleven chapters. Let us now
describe in more detail the content of each chapter separately.
The first chapter contains the generalization to functions with values in Ba-
nach spaces of the traditional material from Complex analysis of one variable
which is usually contained in the beginning of a basic course.
Chapter 2 starts with Pompeiju’s integral formula for solutions of the in-
homogeneous Cauchy-Riemann equation, the Runge approximation theorem, the
Mittag-Leffler theorem, and the Weierstrass product theorem. Then, in Sections
2.6 and 2.7, we present the (less well known) “Anschmiegungsatz” of Mittag-Leffler
and a strengthening of the Weierstrass product theorem. In the case of the Weier-
strass product theorem and its generalization, in this chapter, we still restrict
ourselves to scalar functions. It is one of the main goals of this book, to generalize
these results to the case of operator functions, using Grauert’s theory of cocycles.
Chapter 3 is dedicated to the splitting problem with respect to a contour for
functions with values in a Banach space, as well as to the factorization problem
for scalar functions with respect to a contour.
In Chapter 4 we generalize to finite meromorphic Fredholm operator func-
tions the classical Rouché theorem from Complex analysis and the Smith factor-
ization form. The proof is based on the local Smith form.
Chapter 5 is entirely dedicated to the theory of multiplicative cocycles, which
were discussed in large before.
Chapter 6 contains a theory of families of subspaces of a Banach space E.
First we introduce a complete metric on the set G(E) of closed subspaces of E, the
so-called gap metric. A continuous family of subspaces of E then will be defined as
a continuous function with values in G(E), and a holomorphic family of subspaces
of E will be defined as a continuous family of subspaces which is locally the image
of a holomorphic operator function. Vector functions with values in such a family
are called sections of the family. Note that we do not require that the members of
a holomorphic family be complemented in the ambient space. It may even happen
they are not pairwise isomorphic. An example is given in Section 6.5.
First we prove the following results: any additive cocycle of holomorphic
sections in a holomorphic family of subspaces splits; for any holomorphic operator
function A whose image is a holomorphic family of subspaces, and any holomorphic
section f of this family, there exists a global holomorphic vector function u that
solves the equation Au = f ; for any holomorphic family of subspaces there exists
a global holomorphic operator function with this family as image. Proving this,
xvi Introduction

the main difficulty is the solution of certain local problems (in this generality,
published for the first time in this book). In terms of Complex analysis of several
variables, the solution of these local problems means that any holomorphic family
of subspaces is a so-called Banach coherent sheaf (a generalization of the notion
of coherent sheaves). After solving this we proceed by standard methods that are
well-known in Complex analysis of several variables.
Then we consider holomorphic families of subspaces, which we call injective
and which have the additional property that, locally, the family can be represented
as the image of a holomorphic operator function with zero kernel. We study the
problem of a corresponding global representation. Here we need the theory of
multiplicative cocycles from Chapter 5. It turns out that this is not always possible,
but we have again an Oka-Grauert principle.
Then we study holomorphic families of complemented subspaces (which are
injective), where we can prove more precise results than for arbitrary injective
families. Again there is an Oka-Grauert principle.
At the end we consider the special case of families of subspaces which are finite
dimensional or of finite codimension. Here there are no topological restrictions.
Chapters 7 and 8 are dedicated to factorization of operator functions with
respect to a contour and the connection with Wiener-Hopf and Töplitz operators.
This type of factorization was in fact considered for the first time in the pioneer-
ing works of Plemelj and of Muschelishvili. Because of that we call it Plemelj-
Muschelishvili factorization. We start with the local principle, which quickly fol-
lows from the theory of multiplicative cocycles and which allows us to prove theo-
rems on factorization for different classes of operator functions. The local principle
reduces the problem to functions which are already holomorphic in a neighborhood
of the contour.
For further applications we need a generalization of the theory of multiplica-
tive cocycles. This is the topic of Chapter 9, where we introduce cocycles with
restrictions. Let us offer an example (which is basic for all cocycles with restric-
tions). Suppose that in Theorem 0.0.1 an additional set Z ⊆ D1 ∪ D2 , discrete and
closed in D, and positive integers mw , w ∈ Z, are given. Assume that the function
A − I has a zero of order mw at each w ∈ D1 ∩ D2 ∩ Z. Then the theory of cocycles
with restrictions gives the additional information that the functions A1 and A2 in
Theorem 0.0.1 can be chosen so that, for all w ∈ Dj ∩ Z, j = 1, 2, the function
Aj − I has a zero of order mw at w.
In Chapter 10, by means of the theory of cocycles with restrictions, we es-
sentially improve the Weierstrass product Theorem 0.0.2: The functions Hw in
this theorem now can be chosen so that, additionally, for each w ∈ Z, the func-
tion Hw − I has a zero of an arbitrarily given order mw at w. This has different
consequences that are discussed in this short chapter.
Chapter 11 is dedicated to holomorphic equivalence and its applications to
linearization and diagonalization. Let E be a Banach space, let L(E) be the space
of bounded linear operators in E, let GL(E) be the group of invertible operators
from L(E), let D ⊆ C be an open set, and let Z be a discrete and closed subset
Introduction xvii

of D. Then two holomorphic operator functions A, B : D \ Z → L(E) are called

(globally) holomorphically equivalent over D if there exist holomorphic operator
functions S, T : D → GL(E) such that A = SBT on D.
In the first section, results are presented that explain the importance of
the notion of holomorphic equivalence in spectral theory of linear operators and
holomorphic operator functions. It contains the following two results: 1) For each
relatively compact open subset Ω of D, each holomorphic operator function A :
D → L(E), after an appropriate extension, becomes holomorphically equivalent
to a function of the form zI − T , z ∈ Ω, where T is a constant operator and I is
the identical operator (Theorem 11.2.1). 2) Two operators T, S ∈ L(E) with the
spectra σ(A) and σ(B) are similar if and only if some extensions of the functions
zI − T and zI − S are holomorphically equivalent over some neighborhood of
σ(A) ∪ σ(B) (Corollary 11.2.3).
The remainder of this section is devoted to the relation between global and lo-
cal holomorphic equivalence where two holomorphic operator functions are called
locally holomorphically equivalent if, for each point, they are holomorphically
equivalent over some neighborhood of this point. We prove that two meromorphic
operator functions with meromorphic inverse are locally holomorphically equiva-
lent if and only if they are globally holomorphically equivalent (Theorem 11.4.2),
and we prove that any finite meromorphic Fredholm operator function is globally
holomorphically equivalent to a diagonal function (Theorem 11.7.6). The local fact
behind this is the Smith representation of matrices of germs of scalar holomorphic
functions.
Acknowledgement. In the beginning of the 1970s, on an invitation of one of us,
M.A. Shubin visited Kishinev and gave two talks about applications of Grauert’s
theory and the theory of coherent analytic sheaves to different results for linear
operators. One of the talks was on the local principle for Plemelj-Muschelishvili
factorization of matrix functions and the second was about the analysis of holo-
morphic families of subspaces. These talks had on us an important influence. Very
soon after this we came up with a series of papers on operator-valued cocycles
in the case of one variable with new direct proofs and also with new results and
applications to operator functions. At the end this development led to this book.
It is our pleasure to thank M.A. Shubin providing us with the initial input.
Notation

Here we give a list of standard symbols and some remarks concerning the termi-
nology used in this book without further explanation:
– C is the complex plane, R is the real axis, C∗ := C \ {0}, R∗ := R \ {0}.
– N is the set of natural numbers (including 0), N∗ := N \ {0}.
– Z is the set of entire numbers.

– Banach spaces and Banach algebras are always complex.

– If E, F are Banach spaces, then we denote by L(E, F ) the Banach space of
bounded linear operators operators from E to F , endowed with the operator
norm. We set L(E) = L(E, E), and we denote by GL(E) the group of all
invertible operators from L(E). By a projector in E we always mean an
operator P ∈ L(E) with P 2 = P .
– By an operator we always mean a bounded linear operator between two Ba-
nach spaces.
– Let E, F be Banach spaces, and let A ∈ L(E, F ). Then we denote by Im A
the image, and by Ker A the kernel of A. The operator A is called injective,
if Ker A = {0}, and it is called surjective if Im A = F .
– The unit operator of a Banach space E will be denoted by I or IE .

– For n ∈ N∗ we denote by L(n, C) the algebra of complex n × n matrices, and

by GL(n, C) we denote the group of invertible elements of L(n, C).

– By a neighborhood we always mean an open neighborhood, if not explicitly

stated to be anything else.
– If U is a set in a topological space X, then U always denotes the topological
closure of U in X (and not the complement).
xx Notation

– By C 0 -functions or functions of class C 0 we mean continuous functions. If Γ

is a subset of C and M is a subset of a Banach space, then we denote by
C M (Γ) or by (C 0 )M (Γ) the set of all continuous functions f : Γ → M .

– If U ⊆ C is an open set, U = ∅, and E is a Banach space, then a function

f : U → E is called C k or of class C k on U , k ∈ N∗ ∪ {∞}, if it is k times
continuously differentiable with respect to the canonical real coordinates of
C.

– If U ⊆ C is an open set, U = ∅, and M is a subset of a Banach space E,

then we denote by (C k )M (U ) the set of all Ck -functions f : U → E such that
f (z) ∈ M for all z ∈ U , and by OM (U ) we denote the set of all holomorphic
(Def. 1.1.1) functions f : U → E such that f (z) ∈ M for all z ∈ U .

∗
– We set O(U ) = OC (U ), C k (U ) = (C k )C (U ) and O∗ (U ) = OC (U ) for each
open U ⊆ C and k ∈ N.

– If K ⊆ C is a (not necessarily open) set of uniqueness for holomorphic func-

tions (for example, the closure of an open set, or an interval) and E is a
Banach space, then we speak also about a holomorphic function f : K → E
to say that f is the restriction of an E-valued holomorphic function defined
in a neighborhood of K.

– If D ⊆ C is an open set with piecewise C 1 -boundary (Def. 1.4.1), then we

denote by ∂D the boundary of D endowed with the orientation defined by
D (Sect. 1.4.1), i.e., D lies on the left side of ∂D.
Chapter 1

Elementary properties of
holomorphic functions

This chapter is devoted to the basic facts usually contained in a basic course on
Complex analysis of one variable. The difference is that we do this for functions
with values in a Banach space. Many (not all) of these results will be deduced by
the Hahn-Banach theorem from the corresponding scalar fact.
Some care is necessary with respect to the maximum principle. The strong
version, that the norm of a non-constant holomorphic function does not admit
local maxima, is not true in general. For example, it fails for l∞ and it is true for
Hilbert spaces.

1.1 Definition and first properties

The notion of a holomorphic function with values in a Banach space can be defined
as in the scalar case by complex differentiability:
1.1.1 Definition. Let E be Banach space, and let U ⊆ C be an open set. A function
f : U → E is called complexly differentiable or holomorphic if, for each w ∈ U ,
f (z) − f (w)
f  (w) := lim
z→w z−w
exists. Clearly, then also the partial derivatives of f with respect to the canonical
real coordinates x, y exist, and the Cauchy-Riemann equation holds:
∂f ∂f
f  (w) = (w) = i (w) , w ∈ D. (1.1.1)
∂x ∂y
The function f  : U → E, which is then defined, will be called the complex
derivative or simply the derivative of f .
Exploring the Variety of Random
Documents with Different Content
comfort, which proved in the end to be a trench two feet deep,
which had been started by the Germans as a cable trench.

It had been reported that the 6th Battalion was holding the Flers
Line, and the Civil Service Rifles were accordingly going to occupy
Drop Alley, a communication trench leading from the strong point
called the Cough Drop to the Flers Line. But on arrival at the Cough
Drop, Lieutenant Ind found that the report was untrue, and he had
perforce to squeeze his small body of men in the western half of the
Cough Drop and the afore-mentioned cable trench which ran out of
it. It had been a long and weary journey, but the men set to work
like niggers to dig a decent trench. There were now only two officers
and about 100 other ranks, but these included a good sprinkling of
seasoned warrant officers and N.C.O.’s, and the force made up in
quality what it lacked in quantity. C.S.M. Callingham and Sergeant
Irving of “A” Company had done their comrades a very great service
by struggling along with a jar of rum, which was practically all that
turned up that night in the shape of rations, unless mention is to be
made of rations sent up by a thoughtful Quartermaster for the two
officers—a bag of candles!

The process of digging in was no sooner finished than the exhausted

troops had to stand to for about an hour and a half, on information
from the 6th that the enemy was “coming over in large numbers.”

So the day went on with a constant succession of alarms, intense

bombardments and standing to. It was indeed a trying time for all
that was left of the four Companies and Lewis gun teams who, forty-
eight hours ago, had been so full of hope. But they all “stuck it” very
valiantly, and the excellent spirits of the men—prominent among
whom were Lance-Corporal F. A. Coward and his Lewis gun team,
Privates Hundleby, Lynch, and E. H. Lyons—together with the
splendid example set by Paddy Brett and Bob Harris, served to
sustain the excellent morale of the Civil Service Rifles. Special
mention should be made of the excellent patrol work done by
Sergeant D. Gooding, of “D” Company, who went out in broad
daylight “to find touch on the left.” The left flank of the position was
exposed, and it was not known whether friend or foe occupied the
country beyond. Sergeant Gooding, with two men, started off,
therefore, without any information and, although under rifle fire
from shell holes and isolated posts, they carried out a complete
reconnaissance of the country which separated the Civil Service
Rifles from the nearest friendly troops, four or five hundred yards
away. Many others of that little band distinguished themselves by
their devotion to duty during a day when there was no
communication of any kind from Battalion Headquarters, and the
party in the line became attached to the 6th Battalion in the Cough
Drop.

On the 18th, a small reinforcement arrived in the shape of three

officers—Lieutenants W. L. C. Rathbone, G. M. Hoste, and B. K.
Ware, and fifty other ranks, from the “Non Starters” camp in Bottom
Wood, where a few officers and other ranks had been kept out of
the fight in order to form a nucleus for reorganisation in case of
heavy losses.

Before being relieved on the night of the 19th September by the 1st
Battalion The Black Watch, the Civil Service Rifles undertook two
more operations. The first was an advance into the Flers Line on the
18th, but as the enemy had by now evacuated this trench, the
advance passed off without loss.

But the enemy still held the junction of the Flers Line and Drop Alley,
and that portion of the Flers Line west of the junction. The New
Zealanders had a party in the Flers Line between the Bosches and
the Civil Service Rifles, but the Civil Service Rifles had a small force
under Lieutenant B. K. Ware in Drop Alley. These two forces
attempted, by joint bombing attacks, to dislodge the Bosches, but
the attempt failed. The men were now thoroughly exhausted, for in
addition to the enormous amount of work of the past few days and
the excitement of the fray, the last twenty-four hours had been
endured in a pitiless rain, which caused huge chunks of the trenches
to give way. There was mud and rain everywhere and, as there was
no shelter, rifles and Lewis guns eventually became choked with
mud. It was while in this state that the enemy attempted to drive
Lieutenant Ware’s party out of Drop Alley. He partially succeeded at
first, but was afterwards driven back. But at 7.0 p.m. on the 19th, he
came again with renewed vigour, and got down Drop Alley, where
the defending troops, with rifles and Lewis guns out of action, and
themselves quite worn out, were unable to dislodge him. They did
not give up without a struggle, however, and Lieutenant Ware died
that night in a plucky attempt to achieve the impossible. Thus ended
the operations of the Civil Service Rifles at High Wood, but it was
indeed a skeleton of a battalion that Lieut.-Colonel Warrender led
down the New German Road to Bottom Wood on the morning of the
20th September. Round a huge bonfire these remnants threw
themselves down to get their first rest since leaving Becourt Wood,
and here a pause was made to count the cost of what was so far the
greatest trial of the Civil Service Rifles and, at the same time, surely
their greatest achievement.

To this day, High Wood is regarded by many as the finest

performance of the Battalion during the war. But whether this is true
or no, it is certain that this battle was the most distinctive landmark
in the history of the Civil Service Rifles, for it was at High Wood
where the first great changes took place in the personnel of the
Battalion. Some say it was the last of the original Battalion, but such
a statement is open to question.

Many old faces had gone, but the old spirit still remained, and there
were enough old hands left to train drafts in the way they should go,
and to tell them what manner of men they had been whose places
these freshmen had the honour to fill.

There fell during the fighting at High Wood, so many of the real
flower of the Battalion that it is impossible to do justice to them by
any eulogy here, and it would be invidious to single out any in
particular among so many illustrious dead. Their names will all be
found recorded elsewhere. Suffice it to say that they died like the
true Englishmen of tradition, every one gallantly and gamely carrying
on against odds. In the four days the casualties amounted to 15
officers, 365 other ranks.
 CHAPTER XI
EAUCOURT L’ABBE AND THE BUTTE DE WARLENCOURT

The so-called camp in Bottom Wood was the essence of discomfort,

but after a meal and a few hours rest, a welcome move was made in
the evening of the 20th September to Albert, where one night was
spent in deserted houses.

The march was continued the following day, and the Battalion
arrived at a tented camp in a wood just outside the village of
Henencourt, where Corps Headquarters was situated in a
magnificent château, the grounds of which were a replica of those of
the Palace of Versailles. There was little to suggest the luxury of
Versailles, however, in the camp occupied by the Civil Service Rifles,
for although the Battalion was depleted, the accommodation was
scarce and every one was crowded.

The process of refitting and reorganisation was begun, and to a draft

of one officer and 375 other ranks, who joined at Henencourt was
added a fair sprinkling of officers, N.C.O.’s and men who had missed
High Wood through leave, courses, or other causes. Thus the
strength of the Battalion was restored on paper, but in actual fact it
was still but a shadow of its former self.

The officers spent most of their time at Henencourt Wood in writing

letters of condolence to bereaved parents, and the Company
Quartermaster Sergeants and senior N.C.O.’s were busy packing up
and sending off the personal effects of the killed and wounded, so
that on the whole, the ten days’ sojourn in this camp was not a
joyous one. The Divisional Follies tried to cheer things up by giving a
show one evening, but the proceedings fell flat, and those who
wanted a little diversion while the Battalion was at Henencourt
sought it in Amiens.

The last day of the month of September, 1916, found the Civil
Service Rifles once more on the way back to war, for after spending
one night in Albert, they occupied some disused trenches, entirely
devoid of dug-outs, in what was known as the Quadrangle, near
Mametz Wood. Here they waited eagerly for news of the attack by
the 141st Brigade on the village of Eaucourt L’Abbé, for which the
140th Brigade was in reserve.

Numerous contradictory reports reached Colonel Warrender during

the few days spent in the Quadrangle, but at last it became known
definitely that Eaucourt L’Abbé had been captured, and that the
140th Brigade would go there to relieve the 141st, but would only
hold the line—there would not be any further attack!

The relief which took place on the night of the 4th October, when
the Civil Service Rifles relieved the Poplar and Stepney Rifles in the
Flers Line at Eaucourt L’Abbé, was an ordeal almost as trying as a
battle.

The march from the Quadrangle began at 4.0 p.m. on the 4th, and
the tail of the Battalion reached the Flers Line at dawn the next day.
The event was so unique that no apology is offered for a somewhat
lengthy description:—

All was going well until the corner of High Wood was reached,
where, according to plan, guides would be picked up. There were,
however, several corners to High Wood, and the Lewis gun limbers,
mess cart and medical cart were not taken to the same corner as the
one to which the Battalion went.

After a very long delay, while Lewis guns, etc., were carried through
the wood from the limbers to the Battalion over many awkward
obstacles such as wide trenches and barbed wire, a start was made
by half the Battalion, and about two hours afterwards the remainder
of the Battalion was ready. The way was along a track of sticky mud
of the typical Somme variety. The night was pitch black and the men
slipped about and frequently their feet stuck in the mud. It was
often necessary for two men to pull at another man to get him out
of the mud, and as they got their man out they found themselves
stuck in in turn. At one time Colonel Warrender was heard to tell the
M.O. that a man had fallen down, but he feared it was no use going
back to him “as he must have been trampled in by now.” The
progress along the track, slow as it was, became slower still when
one after another the guides announced that they were lost and had
not the slightest idea in which direction to go. A touch of humour
crept into the adventure when Colonel Warrender, addressing a
guide who said he hadn’t the remotest idea where he was, told him
to go back to his Commanding Officer and report that he was of no
use!

After many hours the Cough Drop was reached by the party bringing
up the rear, which included Battalion Headquarters, and here one of
the other Companies was met coming in the opposite direction.
They, too, had a guide who was lost. The Adjutant now took up the
running alone and plunged into the darkness on an exploring tour.
He soon came back, and then led the whole party, now consisting of
a good many more than half the Battalion, through Drop Alley to the
Flers Line. The going now began to tell on the exhausted troops and
several there were who collapsed unconscious in Drop Alley, weighed
down by the heavy loads they were carrying, and did not finish their
journey until the following day.

The Flers Line is chiefly remembered for the number of dead, both
English and German, who were still lying about on the floor of the
trench and on all the firesteps. There were a few hurriedly-made
dug-outs, but these were in such a filthy state as to be unfit to
occupy, and although much hard work was done for the next two
days, the cleaning of the trench was still unfinished when the troops
learnt to their astonishment, on the 7th of October, that they were to
attack the Butte de Warlencourt and the Warlencourt Line—an
objective some 2,800 yards distant.

Zero was at 2.0 p.m., and the Companies occupied the same relative
positions as at High Wood, “A” Company again being on the right.
The three Companies on the left were unfortunate once more, for
they had to file through the village of Eaucourt l’Abbé soon after
leaving their assembly trenches and extend into waves again after
negotiating the village. They were caught by the full fury of the
German artillery barrage, and those who got through the village
were swept down by a most intense machine-gun fire. “A” Company
on the right made some little progress, and after crossing the
Eaucourt l’Abbé-Le Barque road dug a new line alongside the
remnants of other units of the Division, all of whom had met a
similar fate. Another attempt was made at night by the 142nd
Brigade, but as these troops had not even seen the country in
daylight, their attempt failed so completely that they were all
withdrawn shortly after zero.

The attack on the Butte de Warlencourt failed, like many attacks

subsequently delivered by other Divisions, and the famous Butte did
not fall into English hands until the German retreat from the Somme
battlefield during the winter of 1916-17.

The attack of the 7th of October differed in many respects from that
of the 15th of September. On this occasion there had been no
training, no rehearsal over a marked-out course, and in fact some of
the troops did not even know there was to be an attack until an hour
or so before zero. Even then there were many who were not sure
what was the objective. To this day there are some in the Civil
Service Rifles who talk of it as the attack on Eaucourt l’Abbé. There
was an artillery creeping barrage on this occasion, it is true, but as it
moved at the rate of 100 yards per minute and there were 2,800
yards between the jumping-off trenches and the objective, the
advancing waves of infantry soon got badly left behind. Tanks were
said to be co-operating, but nothing was seen of them.
There were only two officers per Company present on this occasion,
and the C.S.M. and one or two senior N.C.O.’s of each Company
were kept out of the fight, so the experienced soldier was in a
distinct minority. More than half of the Battalion had never been
under fire before, and, as these had only joined a few days
previously, a good many of them were not known even by name to
the older members of their Platoons. Thus it came about that many
men were reported missing on this occasion, and, as none of the
survivors knew them, it was impossible to say with any certainty
where they had last been seen.

The losses on the 7th of October amounted to five officers, 344

other ranks, and although numerically they are not quite so great as
at High Wood, it should be remembered that on this occasion the
Battalion was not more than 500 strong at the outset.

During the operations around Eaucourt l’Abbé there was one

member of the regiment who added to his already brilliant
reputation as a soldier. The work of Lieutenant W. E. Ind on this
occasion was more than wonderful. Quickly grasping the situation
when the attack failed, by his hard work and resourcefulness he
succeeded in restoring something like order out of chaos, not only in
his own unit but also in several neighbouring units.

The relief by the 7th Seaforth Highlanders on the 9th of October was
a welcome contrast to the previous relief in this sector. The troops
quickly found their way out and before midnight had reached the
transport lines in Bottom Wood.

Three nights were spent in Albert before the Division entrained on

the 13th of October for Longpré, near Abbeville, en route for the
Ypres Salient.

Before leaving Albert, the Quartermaster aroused the wrathful

indignation of the C.Q.M.S.’s by the issue of a quantity of clothing
and equipment which had been applied for at Henencourt. Many of
the men for whom it was intended had now become casualties, but
that made no difference to the Quartermaster’s stores of the Civil
Service Rifles. The most important article of clothing was the clean
shirt which was issued just before leaving Albert. The troops had not
had a clean shirt for many weeks, and the one they discarded was
naturally somewhat the worse for wear. One C.Q.M.S. on inquiring at
the Orderly Room what should be done with the old shirts was told
by an Orderly Room clerk to burn them. The clerk was trying to be
funny, but the Q.M.S. missed the point of his humour, and all
Companies thereupon threw their old shirts on the dust heap. When
he was demobilised some years later the C.Q.M.S. was still
explaining to the authorities why he had destroyed his ultra lousy
shirts.

The train journey from Albert to Longpré is surely a record even for
the R.O.D. A distance of just over thirty miles was covered in the
astonishingly short time of twenty-six hours, during which time many
men had left the train, dined in Amiens, visited the local cinema, and
still caught the train up again without being recorded as absent.
Indeed, during one part of the journey there seemed to be more
men walking than were riding. At the same time every one seemed
conscious of the fact that he had said good-bye to the dreaded
Somme battlefield, so few felt disposed to complain of the
shortcomings of the R.O.D.

After detraining at Longpré, two happy days were spent in the

village of Villers-sous-Ailly. The men received a hearty welcome from
the natives and M. le Maire, who seemed to be the greatest French
authority on the organisation of an English infantry battalion. This
worthy was very popular with the billeting party, for he had his
village completely mapped out, and could tell them whether a
particular barn was big enough to hold a platoon, a section, or a
Lewis gun team.

The Battalion returned to Longpré on the 16th of October and

entrained for Caestre, which was reached in the early hours of the
following morning, whence a long and uninteresting march brought
the Battalion to scattered billets outside the village of Boeschepe,
and after another long march on the 19th, the Civil Service Rifles
relieved the 16th Battalion Australian Infantry in support to what
was called the Bluff Sub-sector (or Canal Sub-sector) south of Ypres,
and close to what had once been the Ypres-Comines canal.
 CHAPTER XII
A REST CURE IN THE YPRES SALIENT

To those whose memories of Ypres are only associated with thoughts

of mud and slaughter, and who at the mention of the word “Salient”
instinctively think of the horrors of Passchendaele, the Menin Road
and Hooge, it will seem incredible that there was a time during the
war when the Ypres Salient was peaceful and quiet, a place where
Divisions, shattered on the Somme, came for recuperation.

It was in such a state that the 47th Division found the Ypres Salient
in October, 1916, and after what had been endured in the previous
month, it was particularly welcome.

In the Civil Service Rifles reconstruction had only just begun. No

drafts had reached the Battalion, which was very much below
strength. Some Companies had only one officer, the Company
Commander, and practically no N.C.O.’s above the rank of Lance-
Corporal. It was well, therefore, that there was no fighting and the
sound of a shell was the exception rather than the rule.

The Division had had no experience of trench warfare for some

months, and when the Civil Service Rifles on the 24th of October,
1916, relieved the Post Office Rifles in the front line in what was
called the Ravine, a section of the Bluff Sector, they found several
features of trench warfare which were quite new to them.

In the first place, each Company had a cookhouse in the trenches,

and the Company cooks came in with their Companies and cooked
all the meals on the spot. Rations were pushed up almost to
Company Headquarters in trucks along a light railway, and there was
a dump of R.E. material actually in the Battalion area. These were all
amenities of trench life hitherto unknown, and all helped to convey
the idea that the Civil Service Rifles were making a new start in life.
These conditions helped materially to restore the confidence and
fighting spirit of troops who were rapidly approaching the “fed-up”
state.

The trenches were mostly sandbagged barricades such as had been

seen as Festubert, but here they were neatly revetted with expanded
metal, and although there was a good deal of water, all the trench
floors were boarded. There were recesses labelled for bombs and
S.A.A., and although the trench shelters were not by any means
shell-proof, the majority at least were weatherproof.

There were many ambitious schemes for winter comforts. A Brigade

gum boot store with lots of thigh gum boots was already
established, and large shelters were being erected in each Battalion
area as drying rooms, where men would be able to dry their clothes.
These shelters, however, never got into working order. Another
novelty was the precaution taken to prevent trench feet or frost bite.
It was arranged that every man should change his socks and rub his
feet with whale oil every day, the old socks being sent down every
night and exchanged at the Divisional Laundry for alleged clean ones
which were brought up with rations the following night.

Thus it was hoped to combat some of the evils which beset the
Army during the previous winter, and there is no doubt that these
measures bore good fruit, for the losses through sickness during the
winter of 1916-17 were less than half those of the previous winter.

But although the prospect at the beginning of winter was very

bright, and the troops were looking forward confidently to a spell of
quiet life, it was not long before things began to liven up, as though
the Ypres Salient had begun to look to its reputation.
The change was first noticeable when, on returning to the front line
after a few days in Divisional Reserve at Ottawa Camp, near the
village of Ouderdom, the Battalion took over a section of the front
line in what was called the Hill 60 Sector, on the 13th of November.

These trenches differed in many respects from those in the Ravine,

in spite of the fact that they were practically adjacent. Mining
activities were carried on here on a very elaborate scale, and there
were several deep tunnels, some, it was said, running as far forward
as the German front line on Hill 60 itself. These tunnels were all
being worked by one of the Australian Tunnelling Companies, to
whom working parties were sent day and night. The main line of the
Ypres-Roulers railway ran through the sector, and the old railway
cutting formed the right boundary of the Battalion front. The
trenches had been in existence for many months, and owing to the
continual bombardments, the ravages of weather, and the quaint
ideas of sanitation of former occupants of the sector, the area could
hardly claim to be a health resort. Large fat rats abounded in and
around every trench, and so fat were they that they had lost their
turn of speed, and fell easy victims to any who could find time to
hunt them. Bully beef and Maconochie’s famous meat and vegetable
rations were to be found everywhere. Some men say that these
were often used in place of trench boards.

Those working in the tunnels with the Australians were impressed,

almost awestruck, by the magnitude of the mining operations, which
they felt sure would end one day in a miniature earthquake, and
they fervently hoped they would be at a safe distance when that
should happen. The tunnels were lighted by electricity, the power for
which was produced by a gas engine installed underground. A
privileged few were allowed to explore the wonderful Berlin Sap, a
long tunnel which stretched from some distance in rear of Battalion
Headquarters to the German lines.

But it must not be imagined that the troops enjoyed home comforts
in this area. The Companies holding the left of the Battalion front
had practically no protection from either shell fire or weather, and
those who have occupied the curiously-named Metropolitan Left and
Metropolitan Right will be ready to swear that there was no more
miserable place on the western front. They were, however, little
better off than those who were stowed away in the tunnels of
Marshall Walk, where the atmosphere reduced the occupants,
packed in tight, to a state of coma.

The enemy had now begun to bombard the area fiercely with
various kinds of shells and minenwerfer bombs, but fortunately he
was kind enough, at first at all events, to limit his bombardments
strictly to certain hours of the day. His special effort was always
served up during the two hours after lunch, and strangely enough it
was mostly bestowed on the Marshall Walk area, where the troops
were able to squeeze into the tunnels. The men in other parts of the
line had to sit under a ground sheet or a bit or corrugated iron and
hope that nothing would come their way. In this way five somewhat
anxious days were endured with comparatively few casualties before
the Battalion moved into support in another of the wonders of the
Ypres Salient—the Railway Dug-outs. These were dug-outs formed
by tunnelling into the railway embankment between the village of
Zillebeke and Ypres itself. Half the Battalion was accommodated
here, the men occupying wire beds which were erected in tiers. The
atmosphere was thick, to say the least, and fatigue parties were
frequently told off to try to fan the foul air out with gas fans. The
other two Companies were at Battersea Farm and Château Belge.

The Railway Dug-outs area had its advantages, however, for there
was little shelling and there were opportunities during the day to
wander out into the fresh air, to visit the Brigade canteen, and
sometimes to visit the ruins of the historic city of Ypres. The working
parties at night were employed in pushing trucks of R.E. material
along what remained of the railway line to the ration dump of the
front line Battalion, in the Hill 60 sector. On the whole the five days
at Railway Dug-outs were written down as not too bad, and after
five more days in the front line in Hill 60 sector, the end of
November saw the Battalion in Divisional Reserve in the huts at
Ottawa Camp.

The Division had now settled down to a very stereotyped form of

warfare, and as there seemed every likelihood that no move would
take place for some months, an elaborate programme of work for
improving the accommodation both in trenches and camps was
embarked upon.

There were two Brigades holding the line and one in reserve
occupying four hutted camps in the neighbourhood of the villages of
Ouderdom and Busseboom. It was arranged that whenever a
Brigade was in reserve, the various Battalions should always go to
the same camp. So it came about that the Civil Service Rifles always
went to Ottawa Camp. This arrangement, it was hoped, would
encourage Battalions to work hard at camp improvements. Works
Officers were appointed and pioneer platoons were detailed in each
Company for this purpose, but every time the Civil Service Rifles
returned to Ottawa Camp they swore no work had been done since
they were last there.

Somebody did work in Ottawa Camp, however, for in course of time

it became transformed from the sea of mud, with a collection of
broken-down, draughty huts, into a tolerably comfortable camp—if
any camp in Belgium could be comfortable. The Battalion Mess for
sergeants was revived, and under the stewardship of Sergeant R. F.
M. Bigby, a fairly successful attempt was made to restore the former
glory of the Civil Service Rifles Sergeants’ Mess.

In looking back on the year spent in the Ypres Salient, the average
member of the Civil Service Rifles, full of the bitter memories of the
Menin Road, Hooge and Château Wood, is apt to forget that there
was a time when life was quite enjoyable in Ottawa Camp, with the
trips to Poperinghe, where there was much gaiety.
It is quite true, however, that at first Ottawa Camp was better known
for its discomfort than for anything else, and it was a curious fact
that Halifax Camp, which was the home of one of the Support
Battalions of a Brigade holding the front line, was much more
comfortable.

The month of December 1916 is chiefly notable for the formation of

what was known as the “football team”—two officers (Second
Lieutenants H. S. Gosney and C. E. Groves) and fifty other ranks.
This team began to train for a raid on the German trenches, when
they hoped to atone for the ill-luck which had attended previous Civil
Service Rifles’ raids.

Sketch Map to illustrate the twelve months in the Ypres Salient October 1916-
September 1917.

The party was housed in reserve dug-outs in the trenches about

Château Ségard, one of the support positions for the Ravine sector,
and in addition to training in the surrounding trenches after dark,
parties went up to the front line nightly to patrol No Man’s Land and
inspect the enemy wire.
The scheme was entered into with enthusiasm by the whole party,
which was split up into six groups, each with a definite job to do,
and they were all brim full of confidence when, at 5.45 p.m. on the
23rd of December, they set out from the front line on their
adventure—to the strains of music from a violin in the German lines!

The troops entered the German front line safely enough and worked
their way round the appointed area. The opposition, not very strong,
was quickly overcome, but no prisoners could be brought back. Two
Bosches did get as far as the parapet but there they decided to stay,
and as nothing would induce them to come over, “they had to be
disposed of,” as one of the N.C.O.’s afterwards said in his report.

The return home at the appointed time was carried out successfully
and as numerous identifications were brought back, the object of the
raid had been achieved. The casualties were very few, but
unfortunately they included two killed—Lance-Corporal A. T. C. Geary
and Private A. F. Pearson.

The success of the raid put the whole party in good spirits for
Christmas, which, as in 1915, was spent in the trenches. The
Christmas celebrations duly took place, however, early in January at
Ottawa Camp, when each Company had a Christmas dinner and
concert. The festival lacked nothing on account of the
postponement, and in many sections of the Battalion it was kept up
for several days. To celebrate their success the survivors of the raid
were given a dinner at which the heroes of the evening were Lance-
Sergeant H. J. Steele and Corporal J. H. Swain, who had both been
awarded the M.M.
 CHAPTER XIII
THE SALIENT IN WINTER

Life in the Ypres Salient could now no longer be described as a rest

cure, for in addition to increased activity on both sides in the line,
the weather was of the real wintry type.

The trenches, where there had been “water, water, everywhere,” had
become ice-bound, and remained so for many weeks. Trench stores
were often taken over by a C.S.M. who could scarcely see them
through the ice, but who was told that he would “find they were all
right when the thaw came.”

To complete the wintry scene, snow had fallen and cast a mantle of
white over the ugly sights of war. The Ravine certainly looked pretty
now, with the feathery snowflakes glistening on the trees, and here
and there an icicle giving the genuine Christmas-card impression. No
Man’s Land, too, has rarely looked more picturesque with the
festoons of barbed wire daintily picked out in white. Yes, it was a
beautiful scene on a moonlight night in January 1917, but the sentry
on the firestep in the front line, with feet frozen, nose, ears and
hands feeling as though they were about to drop off, had no eye for
such beauty. His idea of beauty at the time was a little so-called dug-
out, with a ground sheet or an old post bag (contrary to G.R.O.
“XYZ”) hanging over the entrance, and inside a glorious warm “fug”
with three or four of his pals stewing in the fumes given off by a tiny
brazier. It was so cold on the firestep and the front line trenches
were so near to each other, that he daren’t stamp his feet, for fear
of being heard in the German line. Not that it mattered a great deal
about being heard, thought the sentry, for with his hands in such a
frozen state that he did not know whether he was holding his rifle or
not, he couldn’t do much to stop the Bosche if he did come over to-
night. Things would not be much better when the Battalion was
relieved. He supposed it would be Ottawa Camp again, where it was
bitterly cold and the Quartermaster would never give you any fuel.
Why couldn’t he have a job like “Posh Harry” at the Brigade School,
he wondered? At any rate, there would be a comfortable billet there,
and a fire. He must get a stripe, that’s what it came to, and then if
he could not get a job at the Brigade School or at the Divisional
School in Poperinghe, at any rate he might get sent there on a
course. He was fed up with this life, he was sure, and when his turn
of sentry duty was finished and he was promptly put on an ice-
breaking fatigue, he began to think that there was some sense after
all in the peace talk that one read of in the newspapers at this time.
If only they would talk about it seriously! But what could one expect
when the newspaper folk described the peace talk as “an insult to
Tommy in the trenches”?

Meditations such as the foregoing were not uncommon in the early

part of 1917, when the wintry weather was almost the sole topic of
conversation. But what was dreaded more than the frost and snow
was the thaw which would follow, and how every man prayed that
his Battalion would move into Divisional Reserve the day the thaw
came!

Before the thaw came, however, the Civil Service Rifles were to have
a little excitement in the front line, for one night in the middle of
January, the Bosche, who had evidently been reading the story of
the escape of Mary Queen of Scots from Lochleven Castle, had
attired his patrols in white raiment and sent them out across the
snow. The trick came off, and the Bosches entered the trench known
as Berry Post, inflicted casualties on the garrison, and got back to
their own lines unhurt.

This feat so impressed the authorities that by the time the thaw had
fairly set in, white patrol jackets were awaiting collection from
ordnance.

Many Transport men have unpleasant recollections of nights on this

sector, where rations were taken up to the front line in trucks drawn
by mules on the light railways from Woodcote Farm. In theory,
loaded trucks were picked up at the Farm and hauled to the Ravine
or elsewhere, there unloaded, and taken back to be ready for use
next day. In practice, the trucks were usually at the wrong end of
the railway to start with; and when obtained they invariably came off
the rails at intervals on the up-journey—to a chorus of curses from
the accompanying fatigue party. Drivers have bitter memories of
nights when shelling occurred while trucks were off the rails. They
admit that it was only human for the fatigue party to go to ground;
but they still cannot see how one man could be expected to manage
a distracted mule and also unload, re-rail and re-load a truck of
trenchboards.

It was here that Onions, chief Bolshevik of the Battalion mules was
lost. After the line had been broken by shell-fire, she was sent up at
dawn to bring back stranded trucks. Enemy observers traced her
back to Brisbane Dump and sent out an S.O.S. reporting her
presence there. Onions left hastily with a dislike of barrage fire and a
wound in the head, and was sent forthwith to the Base to be seen
no more by the Civil Service Rifles.

Small drafts were continually arriving during the winter of 1916-17,

and they often included several old N.C.O.’s and men of the original
Battalion, who were coming out for their second trip to France. Such
men were generally given a rousing send off by their companions at
Hazeley Down Camp, Winchester, where the Reserve Battalion was
stationed. The war cry of the returned warriors at the time was
“Everybody once, before anybody twice” and the unofficial flag,
known as the “Twicers” Flag, which was usually carried aloft on the
march to the station, has since been framed, and now hangs in the
Civil Service Rifles’ Club to commemorate the once famous “F”
Company of the Civil Service Rifles Reserve Battalion.

For some unknown reason the officers joining the Battalion belonged
to regiments outside London. There were representatives of the
various Battalions of the Manchester Regiment, the Northern Cyclists
and the Hampshire Regiment. In fact, so many changes had taken
place among the officers of the Battalion, that by February, 1917,
not a single one remained of those who had embarked as officers in
March 1915.

An interesting innovation during the early weeks of 1917, was the

starting of a Regimental Drum and Fife Band. The R. S. M. called it a
Corps of Drums, the troops knew it at first as those —— tin whistles,
but under the leadership of Sergeant Drummer Harmon, the
Regimental Band became an accomplished fact, and the Civil Service
Rifles had music on the march for the first time since coming to
France—except for the early spring of 1916, when “Mattie” Hull
conducted a mouth organ and tin whistle band among the Lewis
gunners.

A change from the eternal round of trench life came about towards
the end of February, when it fell to the lot of the Civil Service Rifles
to act for one month as Works Battalion in the Tenth Corps area.
The Companies were scattered over a very wide area, “A” being at
Château Belge, near Kruistraathoeke, “B” at Coppernollehoek and
Poperinghe, “C” at Pacific Sidings between Busseboom and
Poperinghe, and “D” at Vancouver Camp, close to Vlamertinghe.
Battalion Headquarters remained at Ottawa Camp. The Companies
were employed daily on working parties, chiefly under the Canadian
Railway Construction Company, on work connected with the laying of
a light railway from Poperinghe to the forward area. After the
discomforts of trench life, the change was very welcome, particularly
with “B” Company, who had rather wonderful billets, and “C”
Company, who were all under one roof within easy distance of
Poperinghe.

The Companies were still scattered on the 17th of March, so it was

not possible for the Battalion to celebrate the day with a united
gathering, but “C” Company had a very successful show at Pacific
Siding which was attended by representatives from all the other
Companies. The Sergeants held a belated but very hearty
celebration on another day, and a party from “B” Company held a
dinner in Poperinghe.

Throughout the first six months of the time spent in the Ypres
Salient, the Transport Lines were established at a typical Belgian
farm, and consequently no gathering of Transport men is now
complete without a few tales of Delanotte Farm.

In addition to the Transport, the Quartermaster and his staff spent a

good deal of their time at Delanotte Farm, where also the Civil
Service Rifles’ dentist, Corporal E. Pitt, was installed with all his stock
in trade. The Civil Service Rifles claim to be unique among Infantry
Battalions in the B. E. F. in that they alone possessed their own
dentist, who, although fully qualified, was primarily an infantryman
and was not one of the R. A. M. C. attached.

During his service in France, Corporal Pitt has attended to a

distinguished clientele, including at least one Brigadier General, but
he showed no class distinctions in his dental chair, and the humble
private was always sure of just as careful treatment as was given to
his Brigadier.

Pitt often had to work amid strange surroundings, but his surgery at
Delanotte Farm was perhaps his best known home, and one of his
patients has recorded his impressions of it:
“In the last great European War the ambitious Emperor who
may be regarded as the Kaiser’s prototype stated that an army
marched on its stomach. But what is a stomach without its
teeth? (vide any advertisements of ‘a complete set from one
guinea upwards’). At any rate the British Army has come to
regard the teeth of its lads as anxiously as the fond mother
regards her little one’s chewing organs. These few remarks will
serve to introduce our hero.

“Imagine a brick farm-house in a part of Belgium where the

mud is too muddy for words. The house nestles in a swamp of
green viscous slime. This was for many months the locale of the
C. S. R. Transport while the boys were disporting themselves in
the ditches (misnamed ‘trenches’) in the Ypres district. The
room of the farmhouse was of fairly decent size, with a low
ceiling supported by stout smoke-grimed beams. It was always
well patronised by the lads of the Transport, who would discuss
the inevitable eggs and chips and sip their coffee or beer all day
long. At a large sink by the window the good lady of the house,
assisted variously by a submissive husband, a daughter (who
could by no stretch of imagination be called a coquette!) and a
son, seemed eternally engaged either in preparing a salad of
chickweed and groundsel (or so it seemed), or in counting the
stock of dried haricots. In the other corner by the window, there
was a complete dental establishment installed. This was the
scene of the labours of the indefatigable Pitt, and there was
nothing of the horse-doctor’s methods about him. I can assure
you that he wielded the cocaine-injecting needle as expertly as
the one and only Sherlock himself. Did you want a tooth scraped
or stopped, or filled, or coddled in any way whatever, our
dentist would say ‘right,’ and place a fresh cigarette in his holder
(would that I received a royalty on the cigarettes he smoked!)
Then he would select some fearsome-looking—but really
harmless—instruments from his plenteous stock, and carry on.
His sideline (something like the ‘making bricks in spare time’
stunt) was treating cases that would ordinarily fall to the care of
 the M.O. if he were available. It was surprising how the fame of
Pitt spread near and far; and many and various were his clients.
The amusing part was the complete nonchalance of the people
of the house. They would carry on their weird and wonderful
culinary processes at the sink while the amateur doctor
extracted teeth and poured away bloody water and rinsed foul
dressings under their very noses. In this atmosphere of eggs
and chips, steaming coffee, stale beer, tobacco smoke, flies, and
sometimes washing, Pitt carried on his labours day by day,
month by month, until the Battalion was sent away to happier
(?) hunting grounds....

“Jolly old Pitt! How many of us had cause to bless the fact that
we could go to him for healing balm in our time of bitter
sorrow!”

The Battalion was reassembled after being employed for a month as

Works Battalion, and on the 21st of March renewed its acquaintance
with the front line trenches in the Ravine. The trenches were no
longer ice-bound, but the official first day of Spring did not live up to
its name, for snow and sleet fell throughout the night. Eight
uneventful days in the Ravine were followed by a similar period in
support in the neighbourhood of Swan Château and Château Segard.
Hitherto when the Battalion had been at Swan Château the chief
recreation had been sliding on the pond in the château grounds.
Captain Ind, in fact, made a nightly practice of leading Headquarter
Company in sliding on the ice by moonlight. On this occasion,
however, boating and fishing were freely indulged in. There was an
odd looking craft on the pond, which was in great demand, and the
most popular bait for the fish seemed to be a Mills bomb—though
this bait was not sanctioned by the B. E. F. Angling Society. It was in
fact forbidden by G.R.O., so it naturally follows that no angler was
ever known to use the bait.
Any man who thought the Battalion had come to the Ypres Salient
for boating and fishing, however, was rudely disillusioned when a
return was made to the front line in the familiar Ravine. The London
Irish had just finished a big raid on the German lines when the Civil
Service Rifles relieved them on the night of the 7th of April, and
henceforth things livened up very considerably in this sector, where
life had previously been tolerably quiet. The Bosche now bombarded
furiously, and on the 9th of April (Easter Monday) he raided the
Battalion on the immediate left of the Civil Service Rifles, causing
pretty heavy casualties. The raid took place at 6.30 p.m. and the
accompanying fireworks were kept up throughout the night. As a
sample of the frightfulness that could be served up in the Salient, it
was fairly complete, and the Civil Service Rifles, although not in the
raided trenches, lost thirteen killed and eighteen wounded during
the night.

The Division very soon afterwards had to take over a little more of
the line immediately south of the Ypres-Comines Canal, known as
the Spoil Bank Sector, and as it meant giving up the hated Hill 60
Sector, the change was a very popular one.

Ottawa Camp now came within the area of another Division, and the
Civil Service Rifles, on being relieved on the 12th of April by the First
Surrey Rifles, moved into Divisional Reserve in Devonshire Camp,
near Busseboom.

Early in April rumours of a big Spring offensive began to relieve the

monotony, and the story was passed from one to another in strict
confidence that the Civil Service Rifles would soon attack a château
in a wood just south of the canal, where the Adjutant was of opinion
that “we shall have some interesting wood fighting.”

The next trip to the front line was to the Spoil Bank section itself,
whence rumour had it that the attack would some day be launched,
and amateur tacticians were thus able to study the scheme on the
spot. A preliminary reconnaissance generally ended in the observer
hoping he would be away on leave or on a course when the attack
should eventually take place.

An unsuccessful attempt at a raid by the enemy at 4.30 a.m. on the

25th April was the only incident of note in a somewhat uninteresting
stay in the front line, where nearly every trench appeared to be
open to direct observation from the Bosche. The mystery about the
Spoil Bank sector, with its trenches so open and devoid of shelter,
was that in five days there were only eight casualties, all of which
occurred during the enemy’s attempted raid. The bombardment
during the raid was such as to make every man look forward with
more than usual keenness to the relief on the following night by the
6th Battalion, when the Civil Service Rifles moved to Dominion
Camp, adjoining Devonshire Camp.
 CHAPTER XIV
THE MORINGHEM TREK

A long stay of eleven days in Dominion Lines ended with a return to

the Support positions around Swan Château on the 8th of May, but
only three days were spent here (during which time there were
thirteen casualties, an unusually high number for the comparative
safety of support positions) before the Battalion was relieved by the
Poplar and Stepneys, and moved back to the village of Dickebusch
for two days, before starting on the “Moringhem Trek,” the first affair
of its kind since the memorable trek to the Somme.

The trek began on the 13th of May with a march to the village of
Watou, and the old soldier now knew that he would soon be taking
part in an attack. “They’re not taking us all this way for exercise, or
simply for our amusement” he told the latest joined recruit as they
marched along, “but it’s worth it to get away from the Salient for a
few days and to see the civvies once more without any fear of
shelling. And remember, when we go over the odds are generally
about four to one on a blighty, so don’t worry.”

The billets were good in Watou, but they were even better in the
village of Sercus, where the Battalion, after marching through
Hazebrouck, spent the second night. General regret was felt that
only one night was spent in this village, and on the third day, after
passing through Arques and St. Omer, the training area was reached
and the Companies were billeted in the village of Moringhem and the
neighbouring villages. On the whole the billets were poor and
uncomfortable, besides being scattered. “A” and Headquarters
Company were in Moringhem itself, “B” and “C” were about a mile
and a half away in Petit Difques, and “D” were in the little hamlet of
Cuslinghem.

It was announced that about three weeks would be spent in this

district, and a somewhat ambitious programme of sports and
recreation was drawn up. The training was often finished soon after
mid-day. An inter-platoon football competition was started, and
preliminary heats for the Battalion Sports Meeting, to be held on the
20th of May, were run off during the early days of the “holiday.”

The Brigadier having expressed a wish that officers and men should
be given facilities to visit St. Omer, parties were made up each day
for that purpose. The men generally went in G.S. wagons and
limbers, and the officers usually returned a voiture.

A novelty in Regimental Sports was introduced on the 19th of May at

what was reported in the Financial Times as the Moringhem May
Meeting, when the only event was a horse race for officers’ chargers.

The race took place after church parade, and was over a five furlong
course behind the village church. Unfortunately, one or two of the
starters were not quite sure where the course was, and
consequently several horses had to wait at “the tapes” while
Battalion orderlies scoured the village for missing runners and
jockeys. More unfortunate still was the experience of the “bookie”
who, at great trouble, had secured costumes for himself and his
clerk, but who spent the morning wandering about the neighbouring
hills, vainly searching for the course, and cursing his clerk for a fool.

Lieutenant Craig, the Transport Officer, acted as starter to a field of

seven. The Adjutant’s horse, Bunty, was first away, but his jockey
lost a stirrup and could not keep the lead. Entering the straight, the
Medical Officer (Captain C. M. Gozney) on the Boy, was in front, and,
shaking off the challenge by Bunty and Sunshine, he rode a good
race and won by a short head from the former, with Sunshine
(ridden by Captain F. D. Balfour) close up third. The winner started
at 5-2, Bunty at 3-1, and Sunshine at 100-3. The favourite Polly (5-
4), ridden by Captain Bowers Taylor, a heavyweight jockey, was quite
unable to give the weight away to the leaders. The bookie, it should
be mentioned, arrived on the course just in time to “pay out,” as
there had been considerable ante-post betting, and all the placed
horses had been well backed.

The Battalion Sports Meeting on the following day was marred by

orders to send off a digging party of 200 other ranks to dig trenches
on the training ground. The whole Battalion had turned up on the
Transport Field, and cookers were in attendance so that tea could be
had on the ground, but after the departure of the working party,
interest in the Sports died down, and the crowd, somewhat
disappointed, filtered away.

Much keener interest was taken in the Inter-platoon football

competition, in the final of which Numbers 5 and 15 Platoons met
three times at Moringhem without being able to come to a decision.
Two hours exciting play on each of the first two occasions failed to
produce even a goal, but at the third meeting, play actually
continued until Lights Out, when the score was two goals all. It
should, perhaps, be mentioned that in these days of summer time it
was not uncommon for Lights Out to be sounded while it was still
daylight. The football final was not decided until after the Battalion
had been in the front line. The score on that occasion, when Number
15 Platoon won by no less than nine goals to nothing, was an
eloquent but sad testimony to what had happened since the three
drawn games at Moringhem.

The stay at Moringhem differed in many respects from the

preparation for the Somme battle in 1916. On this occasion the
Division marched straight to its training ground, which was reached
in three days. The marked out course quickly made its appearance,
and it was made known at the outset that the 140th Brigade was
training for the attack on the German positions around the White
Château, just south of the Canal, near Hollebeke. Moreover, the
“non-starters” were selected at the beginning of the training, and
the rehearsals were carried out with the officers, N.C.O.’s and men
who were going to take part in the actual attack.

The training was nothing like so strenuous as that for the High Wood
battle, and the men thus finished their day’s work with sufficient
energy left for football and other sports in the evening. In one
respect the experience of the Somme preparation was repeated. The
Adjutant, Captain W. E. Ind, M.C., threw himself whole heartedly and
enthusiastically into the work of training the Battalion and had very
quickly mastered every detail of the scheme. It was very largely due
to his efforts that the Civil Service Rifles left Moringhem thoroughly
prepared for their share in the battle and so full of confidence in
their success.

On the whole a very happy time had been spent at Moringhem, and
there was no great anxiety to leave the place on the 31st of May,
when the Battalion returned by train from St. Omer to Poperinghe,
where, as they marched out of the station, the troops were greeted
at once with a few shells—just as a reminder that they were back in
the Salient.

The discomforts of the Ypres Salient were rarely more forcibly

illustrated than on the afternoon of the arrival at the so-called tented
camp in Dominion Lines. The march from Poperinghe Station took
place in the afternoon, and the Battalion marched into Dominion
Lines with visions of a comfortable camp and a welcome cup of tea!
The men were doomed to disappointment, for the “camp” proved to
be a strip of waste ground, very dusty, and without even a blade of
grass on it. There were a few bivouac sheets to be issued to each
Company, and there were about three bell tents for officers. Beyond
this there was not a stick of camp equipment of any kind. The
disappointment was all the keener because the Area Commandant
was a Civil Service Rifles officer, and it was thought he might have
treated his own Regiment a little better. The other three Battalions of
the Brigade were comfortably housed in huts.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebooknice.com