Stochastic Analysis and Diffusion Processes 1st

Edition Gopinath Kallianpur download pdf

https://ebookfinal.com/download/stochastic-analysis-and-diffusion-
processes-1st-edition-gopinath-kallianpur/

Visit ebookfinal.com today to download the complete set of

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

Stochastic Processes and Models David Stirzaker

https://ebookfinal.com/download/stochastic-processes-and-models-david-
stirzaker/

Ergodic Control of Diffusion Processes 1st Edition Ari

Arapostathis

https://ebookfinal.com/download/ergodic-control-of-diffusion-
processes-1st-edition-ari-arapostathis/

Levy processes and stochastic calculus 1st Edition David

Applebaum

https://ebookfinal.com/download/levy-processes-and-stochastic-
calculus-1st-edition-david-applebaum/

Applied Stochastic Processes 1st Edition Ming Liao

(Author)

https://ebookfinal.com/download/applied-stochastic-processes-1st-
edition-ming-liao-author/
Applied Probability and Stochastic Processes 2nd Edition
Frank Beichelt

https://ebookfinal.com/download/applied-probability-and-stochastic-
processes-2nd-edition-frank-beichelt/

Probability and Stochastic Processes 2ed. Edition Roy D.

Yates

https://ebookfinal.com/download/probability-and-stochastic-
processes-2ed-edition-roy-d-yates/

Stationary Stochastic Processes for Scientists and

Engineers 1st Edition Georg Lindgren

https://ebookfinal.com/download/stationary-stochastic-processes-for-
scientists-and-engineers-1st-edition-georg-lindgren/

Stochastic Processes An Introduction 2nd Edition Peter

Watts Jones

https://ebookfinal.com/download/stochastic-processes-an-
introduction-2nd-edition-peter-watts-jones/

Stochastic Calculus of Variations for Jump Processes 1st

Edition Yasushi Ishikawa

https://ebookfinal.com/download/stochastic-calculus-of-variations-for-
jump-processes-1st-edition-yasushi-ishikawa/
Stochastic Analysis and Diffusion Processes 1st Edition
Gopinath Kallianpur Digital Instant Download
Author(s): Gopinath Kallianpur, P Sundar
ISBN(s): 9780199657063, 0199657068
Edition: 1
File Details: PDF, 1.88 MB
Year: 2014
Language: english
OXFORD GRADUATE TEXTS IN MATHEMATICS

Series Editors
R. Cohen | S. K. Donaldson | S. Hildebrandt
T. J. Lyons | M. J. Taylor
OXFORD GRADUATE TEXTS IN MATHEMATICS

Books in the series

1. Keith Hannabuss: An Introduction to Quantum Theory

2. Reinhold Meise and Dietmar Vogt: Introduction to Functional Analysis
3. James G. Oxley: Matroid Theory
4. N. J. Hitchin, G. B. Segal, and R. S. Ward: Integrable Systems:
Twistors, Loop Groups, and Riemann Surfaces
5. Wulf Rossmann: Lie Groups: An Introduction through Linear Groups
6. Qing Liu: Algebraic Geometry and Arithmetic Curves
7. Martin R. Bridson and Simon M. Salamon (eds): Invitations to Geometry
and Topology
8. Shmuel Kantorovitz: Introduction to Modern Analysis
9. Terry Lawson: Topology: A Geometric Approach
10. Meinolf Geck: An Introduction to Algebraic Geometry and Algebraic Groups
11. Alastair Fletcher and Vladimir Markovic: Quasiconformal Maps
and Teichmüller Theory
12. Dominic Joyce: Riemannian Holonomy Groups and Calibrated Geometry
13. Fernando Villegas: Experimental Number Theory
14. Péter Medvegyev: Stochastic Integration Theory
15. Martin A. Guest: From Quantum Cohomology to Integrable Systems
16. Alan D. Rendall: Partial Differential Equations in General Relativity
17. Yves Félix, John Oprea, and Daniel Tanré: Algebraic Models in Geometry
18. Jie Xiong: Introduction to Stochastic Filtering Theory
19. Maciej Dunajski: Solitons, Instantons, and Twistors
20. Graham R. Allan: Introduction to Banach Spaces and Algebras
21. James Oxley: Matroid Theory, Second Edition
22. Simon Donaldson: Riemann Surfaces
23. Clifford Henry Taubes: Differential Geometry: Bundles, Connections,
Metrics and Curvature
24. Gopinath Kallianpur and P. Sundar: Stochastic Analysis and Diffusion
Processes
Stochastic Analysis
and Diffusion Processes

gopinath kallianpur a nd p. sundar

3
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Gopinath Kallianpur and P. Sundar 2014
The moral rights of the authors have been asserted
First Edition published in 2014
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2013943837
ISBN 978–0–19–965706–3 (hbk.)
ISBN 978–0–19–965707–0 (pbk.)
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Dedicated to
Krishna Kallianpur
and
Mrs. Sundara Padmanabhan
Preface

T he idea of writing a book on stochastic analysis arose from a suggestion that we write
an enjoyable book on Brownian motion and stochastic analysis. After starting our
work, we felt that the word “enjoyable” should not be construed as a synonym for “heur-
istic and non-rigorous”; rather, it should mean “readable and concise with full details”.
Writing such a book is a tall order, and at best, only partial success is all that one can
hope for.
The origin of the book owes itself to our lecture notes formed over a number of years
and drawn from many inspiring sources. In the present book, we build the basic theory
of stochastic calculus in a self-contained manner in the first six chapters. Starting with
Kolmogorov’s construction of stochastic processes, Brownian motion and martingales
are presented with a view to build stochastic integration theory and study stochastic
differential equations. This would constitute the first part of the book.
The next six chapters deal with the probabilistic behavior of diffusion processes and
certain finer aspects, applications, and extensions of the theory. One can view it as the
second part of this book. The selection of material for the second part of the book
reflects our own tastes and provides only a glimpse of some of the active areas of research.
Stochastic analysis being so vast, important topics such as Malliavin calculus, stochastic
control, and filtering theory, though of interest to us, had to be left out. Instead, we start
with martingale problems, a method unique to stochastic analysis, and proceed to dis-
cuss the connection between stochastic analysis and partial differential equations and
then study Gaussian solutions of stochastic equations. Jump Markov processes, invari-
ant measures, and large deviations principle for diffusions are presented in successive
chapters, though each of these can easily form the subject matter for a whole book.
The book is written for graduate students, applied mathematicians, and anyone inter-
ested in learning stochastic calculus. The reader is assumed to be knowledgeable in
probability theory at a graduate level. A course on stochastic analysis can be designed
using the first part of this book along with parts of Chapter 8. A selection of the last six
chapters can be used for a second course on stochastic analysis.
Regarding interdependence of chapters, the first part of the book is connected natur-
ally as a sequence with one exception: Chapter 4 is not needed to read Chapters 5 and
6. A good knowledge of the first part is required to read any of the following chapters.
However, the second part affords more flexibility in reading. For instance, each of the
Chapters 9, 10, and 12 can be read independently of the others.
viii | Preface

We are quite indebted to several of our friends and colleagues who helped and inspired
us to write this book. We thank several of our colleagues, especially G. Ferreyra, H.-H.
Kuo, U. Manna, P. E. Protter, B. Rüdiger, A. Sengupta, S. S. Sritharan, W. Woyczynski,
and H. Yin. Several of our graduate students read parts of the book and spotted numerous
typos. We appreciate their efforts and thank them for their careful reading. We thank our
families, especially Krishna, Kathy, and Vijay for being patient with us and helping us
cheerfully while the book was written. Thanks to Ms. Elizabeth Farrell for a thorough
proofreading of the entire manuscript.
Contents

1 Introduction to Stochastic Processes 1

1.1 The Kolmogorov Consistency Theorem 1
1.2 The Language of Stochastic Processes 11
1.3 Sigma Fields, Measurability, and Stopping Times 14
Exercises 17
2 Brownian Motion 19
2.1 Definition and Construction of Brownian Motion 20
2.2 Essential Features of a Brownian Motion 27
2.3 The Reflection Principle 34
Exercises 39
3 Elements of Martingale Theory 41
3.1 Definition and Examples of Martingales 41
3.2 Wiener Martingales and the Markov Property 44
3.3 Essential Results on Martingales 49
3.4 The Doob-Meyer Decomposition 54
3.5 The Meyer Process for L2 -martingales 67
3.6 Local Martingales 71
Exercises 73
4 Analytical Tools for Brownian Motion 75
4.1 Introduction 75
4.2 The Brownian Semigroup 76
4.3 Resolvents and Generators 79
4.4 Pregenerators and Martingales 87
Exercises 89
5 Stochastic Integration 90
5.1 The Itô Integral 90
5.2 Properties of the Integral 98
5.3 Vector-valued Processes 105
5.4 The Itô Formula 106
x | Contents

5.5 An Extension of the Itô Formula 111

5.6 Applications of the Itô Formula 113
5.7 The Girsanov Theorem 124
Exercises 132
6 Stochastic Differential Equations 134
6.1 Introduction 134
6.2 Existence and Uniqueness of Solutions 137
6.3 Linear Stochastic Differential Equations 144
6.4 Weak Solutions 146
6.5 Markov Property 153
6.6 Generators and Diffusion Processes 161
Exercises 164
7 The Martingale Problem 166
7.1 Introduction 166
7.2 Existence of Solutions 174
7.3 Analytical Tools 183
7.4 Uniqueness of Solutions 189
7.5 Markov Property of Solutions 193
7.6 Further Results on Uniqueness 196
8 Probability Theory and Partial Differential Equations 202
8.1 The Dirichlet Problem 202
8.2 Boundary Regularity 212
8.3 Kolmogorov Equations: The Heuristics 218
8.4 Feynman-Kac Formula 221
8.5 An Application to Finance Theory 223
8.6 Kolmogorov Equations 224
Exercises 239
9 Gaussian Solutions 240
9.1 Introduction 241
9.2 Hilbert-Schmidt Operators 245
9.3 The Gohberg-Krein Factorization 248
9.4 Nonanticipative Representations 252
9.5 Gaussian Solutions of Stochastic Equations 257
Exercises 265
10 Jump Markov Processes 266
10.1 Definitions and Basic Results 266
10.2 Stochastic Calculus for Processes with Jumps 271
10.3 Jump Markov Processes 275
10.4 Diffusion Approximation 283
Exercises 290
 Contents | xi

11 Invariant Measures and Ergodicity 292

11.1 Introduction 293
11.2 Ergodicity for One-dimensional Diffusions 295
11.3 Invariant Measures for d-dimensional Diffusions 301
11.4 Existence and Uniqueness of Invariant Measures 304
11.5 Ergodic Measures 310
Exercises 314
12 Large Deviations Principle for Diffusions 315
12.1 Definitions and Basic Results 316
12.2 Large Deviations and Laplace-Varadhan Principle 318
12.3 A Variational Representation Theorem 329
12.4 Sufficient Conditions for LDP 338
Exercises 341

Notes on Chapters 343

References 347
Index 351
1 Introduction to Stochastic
Processes

T he main result that guarantees the existence of a wide class of stochastic processes
is the Kolmogorov consistency theorem. Though the Kolmogorov construction of
stochastic processes is set on a very large space equipped with a small σ -field, it is canon-
ical and its applicability is wide. First, we give a proof of this result, followed by important
examples that illustrate its usefulness. Next, we introduce basic terminology and notation
that are useful in stochastic analysis. The chapter ends with a brief overview of stopping
times, associated σ -fields, and progressive measurability.

1.1 The Kolmogorov Consistency Theorem

Throughout the book,  will denote an abstract space which, in probability theory, is
called the sample space or the space of all outcomes. Let F denote a σ -field of subsets of
, known as the class of events or measurable sets in . A measure P on (, F) is said
to be a probability measure if it is a nonnegative, countably additive set function with
P () = 1. The triplet (, F, P) is called a probability space.
In several applications, it is more natural to encounter a finitely additive probability
measure, P0 , on a field G of sets rather than a measure on a σ -field F. The first question
that arises is to find conditions under which P0 can be extended to a probability meas-
ure on the σ -field generated by G. The answer is provided by a well-known theorem on
extension of measures.
The following proposition is quite useful and easy to prove.
Proposition 1.1.1 Suppose P0 is a finitely additive probability measure defined on a field G
of subsets of a space . Let P0 be continuous at ∅, that is, if En ∈ G for all n and En ↓ ∅,
then P0 (En ) ↓ 0. Then P0 is a probability measure on G.
Proof Take any sequence {En } of pairwise disjoint sets from G such that ∪∞
j=1 Ej ∈ G.
Then for any finite n, ∪∞ ∞
j=n+1 Ej ∈ G. Also, one can easily check that ∪j=n+1 Ej ↓ ∅ as
2 | Stochastic Processes
 
n → ∞. Therefore P0 ∪∞
j=n+1 Ej ↓ 0 as n → ∞. By finite additivity,

     
P0 ∪ ∞
j=1 Ej = P 0 ∪n
j=1 Ej + P 0 ∪ ∞
j=n+1 Ej


n
   
= P0 Ej + P0 ∪∞ E
j=n+1 j .
j=1

n  
The partial sums P0 Ej form a bounded increasing sequence indexed by n, and
j=1
hence as n → ∞, we have
  ∞
 
P0 ∪∞ E
j=1 j = P0 Ej . 䊏
j=1

The theorem stated below is a standard result in measure theory. A proof of it can be
found in many texts (for e.g., [56], pp. 19-23).
Theorem 1.1.2 (The Kolmogorov extension theorem) If P0 is a probability measure on a
field G of subsets of , and if F denotes the σ -field generated by G, then P0 can be uniquely
extended to a probability measure on F.
A function X :  → R1 is called a random variable provided X is an F -measurable
function. In probability theory, what we are about to observe is a random variable. The
probabilistic information on X is fully contained in the distribution of X. Recall that given
a random variable X on (, F, P), the distribution of X is the measure μ defined on
(R1 , B) by μ (B) = P {X ∈ B} for all Borel sets B ∈ B.
A natural question that arises is the following converse: Given a probability meas-
ure μ on (R1 , B), can one construct a random variable X on some probability space
(, F , P) such that the distribution of X coincides with the given measure μ? The
answer to this question is quite simple. One can take  = R1 , F = B, and P = μ. Define
X (ω) = ω for all ω ∈ . Thus one can go back and forth between a random variable and
its distribution.
If the range of X is Rn , then X is said to be an n-dimensional random vector. The
discussion in the above paragraph carries over for probability measures on Rn .
In applications, one assumes the existence of a family of random variables defined
on a probability space. When a finite number, n, of random variables out of the given
family is considered, it is a random vector and hence gives rise to a finite-dimensional
distribution which is a probability measure on Rn . As we vary the selections of the ran-
dom variables from the given family, we obtain a family of finite-dimensional measures.
The assumption mentioned above can be removed, had one started with a family of
finite-dimensional distributions. In this situation, it is unclear if there exists a common
probability space on which a corresponding family of random variables can be defined.
To get an affirmative answer, one needs a certain consistency property in the above fam-
ily of finite-dimensional measures. This is the content of a famous result known as the
 Kolmogorov Consistency Theorem | 3

Kolmogorov consistency theorem. We start with a lemma on regularity of probability

measures that holds in general for measures defined on the Borel σ -field of complete,
separable metric spaces (see [33]). However, we prove the lemma for measures on
(Rn , Bn ).
Lemma 1.1.3 Let μ be any given probability measure on (Rn , Bn ). Given any B ∈ B n ,
and  > 0, there exist a compact set K and an open set G such that K ⊆ B ⊆ G with
μ (G \ K) < .
Proof Let A denote the class of all Borel sets B that satisfy the stated property. It suffices
to show that A contains all closed sets and is a σ -field.
Step 1 Clearly, the empty set ∅ ∈ A. Let BN = [–N, N]×n , the n-dimensional closed
box with origin as the center and side-length 2N. Since μ (BN ) ↑ 1 as N increases
to ∞, one can choose N large enough so that μ (Rn \BN ) < . Thus, the full space
Rn ∈ A.
Step 2 We will show that A is closed under complements and countable unions. If
A ∈ A, then by the definition of A, it is possible to choose K ⊆ A ⊆ G such that
μ (G \ K) < /2. As in Step 1, one can choose N large so that μ (BcN ) < /2.
Clearly, Gc ⊆ Ac ⊆ K c .

μ (K c \ (Gc ∩ BN )) = μ (K c ∩ (G ∪ BcN ))
≤ μ (G \ K) + μ (BcN )
< .

 A ∈ A.
c
Thus,
If Aj is a sequence of sets in A, one can choose Kj ⊆ Aj ⊆ Gj such that
μ Gj \ Kj < /2j+1 . Therefore,
      
μ ∪Gj \ ∪Kj ≤ μ Gj \ Kj < /2.
j

Let G denote ∪j Gj . Then, limN→∞ μ(G \ (∪Nj=1 Kj )) < /2. Therefore, for large
enough N, one gets μ(G \ (∪Nj=1 Kj )) < . Thus ∪∞
j=1 Aj ∈ A. We have thus shown
that A is a σ -field.
Step 3 The class A contains all closed sets. For, if F is a closed set, then let Fδ denote
the delta neighborhood of F. That is,

Fδ = x : |x – a| < δ for some a ∈ F .
The set Fδ is an open set and decreases to F as δ ↓ 0. Therefore, one can choose δ
small enough so that μ (Fδ \ F) < /2.
As in Step 1, choose N large so that μ (BcN ) < /2. Then μ (Fδ \ (F ∩ BN )) < 
which completes the proof. 䊏
4 | Stochastic Processes

Let T denote an index set such as [0, ∞).

Definition 1.1.1 For any n ∈ N, let t1 < · · · < tn be any selection (that is, finite
sequence) of distinct elements in T. Let μt1 ,...,tn be a probability measure on the Borel
σ -field of Rn that corresponds to the selection. The family of probability measures
{μt1 ,...,tn : t1 < · · · < tn , ti ∈ T ∀ i, n ∈ N} is said to be a consistent family of probab-
ility measures if the following condition holds:
Let n ∈ N, and t1 < · · · < tn+1 be any choice of n + 1 distinct elements from T. Let
B1 , . . . , Bn be one-dimensional Borel sets, arbitrarily chosen. Then, for any k, we have

μt1 ,...,tn+1 (B1 × · · · × Bk × R × Bk+1 × · · · × Bn ) = μt1 ,...,tk ,tk+2 ,...,tn+1 (B1 × · · · × Bn ).

Definition 1.1.2 Let RT denote the space of all real-valued functions defined on T. A sub-
set of RT is said to be a finite-dimensional cylinder set if it is of the form {x ∈ RT :
(xt1 , . . . , xtn ) ∈ B} for any n ∈ N and distinct t1 < · · · < tn ∈ T and B, Borel in Rn .
The class of all cylinder sets in RT will be denoted C. The σ -field generated by C is
denoted by F. For x ∈ RT , the projection x → (xt1 , . . . , xtn ) is denoted by πt1 ,...,tn .
Theorem 1.1.4 (Kolmogorov consistency theorem) Suppose the family of probability
measures {μt1 ,...,tn : t1 < · · · < tn , n ∈ N} is consistent.
(i) Then there exists a probability measure μ on (RT , F) such that μπt–1 1 ,...,tn
= μt1 ,...,tn
for any n ∈ N and distinct t1 < · · · < tn .
(ii) If  := RT and P := μ, then on (, F , P), there exist random variables {Xt : t ∈
T} such that for any distinct t1 < · · · < tn , the joint distribution of Xt1 , . . . , Xtn
coincides with μt1 ,...,tn .

Proof

Step 1 The class C is a field of subsets of RT . For each C ∈ C, define a set function

P0 (C) = μt1 ,··· ,tn (B)

if C = {x ∈ RT : (xt1 , . . . , xtn ) ∈ B}. By using the consistency condition, it follows

that P0 is well defined. It is easy to verify that P0 is a finitely additive probability meas-
ure on C. If P0 were countably additive on C, we can use the extension theorem to
obtain a unique extension of P0 to a measure on F .
Step 2 By the proposition, it suffices to show that P0 is continuous at ∅. We will use the
method of contradiction. Let An ∈ C ∀ n, and An ↓ ∅. Suppose there exists δ > 0
such that limn→∞ P0 (An ) ≥ δ.
Let ({t1 , . . . , tkn }, Bn ) be a representation in finite dimensions for An for each n.
Then,

P0 (An ) = μt1 ,...,tkn (Bn ) .

 Kolmogorov Consistency Theorem | 5

By inserting as many sets as needed between each Aj and Aj+1 , we can assume without
loss of generality that kn = n for all n.
Given any 0 <  < δ, by Lemma 1.1.3, there exists a compact set Fn such that Fn ⊆
Bn and
μt1 ,...,tn (Bn \ Fn ) < /2n .
Let us denote by En the cylinder set represented by ({t1 , . . . , tn }, Fn ).
Then, P0 (An \ En ) < /2n . Let n denote ∩nj=1 Ej . Then,
   
n
 
P0 (An – n ) = P0 ∪nj=1 An – Ej ≤ P0 Aj \ Ej < 
j=1

since An ⊆ Aj for j ≤ n. Using n ⊆ En ⊆ An , we get

P0 (n ) ≥ P0 (An ) –  ≥ δ –  > 0.
Each n is nonempty, since P0 (n ) > 0. Therefore, for each n, there exists a
(n) (n)
x(n) ∈ n . By using the definition of n , it follows that for each k, (xt1 , . . . , xtk ) ∈
Fk for all n ≥ k.
(n)
Therefore, as a sequence in n, {xt1 } is in a closed bounded set F1 of the real line so
(1,n)
that there exists a convergent subsequence {xt1 } with limit denoted by α1 ∈ F1 .
(1,n) (1,n)
The sequence {xt1 , xt2 } is a sequence in F2 and has a convergent subsequence
whose limit is denoted by (α1 , α2 ) ∈ F2 . This procedure can be continued to obtain
an element (α1 , . . . , αk ) ∈ Fk for any finite k.
It is clear that there exist elements y of RT such that ytk = αk for all k. Any
such element y has the property that (yt1 , . . . , ytk ) ∈ Fk for any finite k. Therefore,
y ∈ Ek ⊆ Ak for all k, which implies that ∩∞ n=1 An is non-empty. This contradicts
the assumption that ∩∞ n=1 An is empty. We have thus proved that P0 on C is count-
ably additive. Invoking the Kolmogorov extension theorem, there exists a unique
probability measure μ on (RT , F ) that extends P0 . The proof of part (i) is over.

Step 3 Set  = RT and P = μ on (, F). Let ω denote a generic element of . Define
the functions Xt (ω) = ωt for all t ∈ T and ω ∈ .
Clearly, {Xt ≤ x} = {ω : ωt ≤ x} ∈ C ⊆ F , so that Xt is a random variable.
Further,
   
P ∩nj=1 Xtj ≤ xj = μ ω : ∩nj=1 ωtj ≤ xj = μt1 ,...,tn ×nj=1 (–∞, xj ] ,

so that the finite-dimensional distribution of the vector random variable Xt1 , . . . , Xtn
is given by μt1 ,...,tn . 䊏

It is possible to reformulate the Kolmogorov consistency theorem in terms of charac-

teristic functions. Toward building such a statement, consider a probability measure P
6 | Stochastic Processes
 
on RT , F . Let F T denote the collection of all points λ in RT such that all but a finite
number of coordinates of λ are zero. Define for any λ ∈ F T ,

φ (λ) = ei(λ,x) P (dx) (1.1.1)

RT

where (λ, x) = j λtj xtj is a finite sum.
Let (t1 < · · · < tn ) be a fixed finite number of distinct elements of T, and let λ be any
element of F T with λt = 0 if t  = tj for any j = 1, . . . , n. The function φ(λ) restricted to
such λ is known as the section of φ determined by (t1 , . . . , tn ).
Theorem 1.1.5 (A reformulation of the Kolmogorov consistency theorem) Let φ be a
given complex-valued function on FT . If any arbitrary section of a function
 φ is a charac-
teristic function, then there exists a probability measure P on RT , F such that (1.1.1)
holds for all λ ∈ F T .
We call the family of random variables {Xt : t ∈ T} obtained in the consistency
theorem the canonical process. It should be observed that the space  in part (ii)
of the above theorem is too large while the σ -field F is too small. In fact,  is the
space of all real-valued functions defined on T. The smallness of F can be inferred
from the following result and examples.
  We will assume that T is nondenumerable.
In what follows, the symbols ω tj and ωtj are synonymous. Let A be a subset in .
 
If there exist a countable set {t1 , t2 , . . .} in T and a Borel set B ∈ B RT such that
A = {ω : (ω (t1 ) , ω (t2 ) , . . .) ∈ B}, then A will be called as a set with a countable base.
The following theorem is due to Dynkin, who extracted a particularly useful part of an
otherwise general, result due to Sierpinski. The theorem is given without a proof. Before
stating it, let us recall that a collection A of subsets of a space E is known as a Dynkin
class or a λ-system if E ∈ A and if A is closed under proper differences and increasing
unions.
Theorem 1.1.6 (The Dynkin class theorem) Let S be a π system, that is, a collection of
subsets of E that is closed under finite intersections. If A is a λ-system with S ⊆ A, then
σ (S) ⊆ A.
Lemma 1.1.7 Let D denote the class of all sets with a countable base. Then F := σ (C) ⊆
D.
Proof It is clear that C ⊆ D . The class C, being a field, is a π -system.
By taking any countable set {t1 , t2 , . . .} in T and B = RT , we get the full space
 ∈ D. Let A1 , A2 ∈ D and A1 ⊆ A2 . If A1 is represented as A1 = {ω : (ω(t1 ),
ω(t2 ), . . .) ∈ B1 }, then there exists a Borel set B2 such that A2 = {ω : (ω(t1 ),
ω(t2 ), . . .) ∈ B2 } with B1 ⊆ B2 . Therefore, A2 \A1 = {ω : (ω(t1 ), ω(t2 ), . . .)
∈ B2 \B1 }, so that D is closed under proper differences.
If {Aj } is an increasing sequence in D and if A1 = {ω : (ωt1 , ωt2 , . . .) ∈ B1 },
then each Aj can be represented by using the same set {t1 , t2 , . . .}. In fact,
Aj = {ω : (ωt1 , ωt2 , . . .) ∈ Bj } with Bi ⊆ Bj for all i ≤ j. Thus, ∪Aj = {ω :
(ωt1 , ωt2 , . . .) ∈ ∪Bj }, which shows that D is closed under increasing unions.
The class D is thus a Dynkin class, and hence, F = σ (C) ⊆ D. 䊏
 Kolmogorov Consistency Theorem | 7

Example 1.1.8 Let c be the set of all continuous functions in . Then c is not in F σ (C).

Proof If c were in F , then by the above lemma, c is a set with a countable base so that

c = {ω : (ωt1 , ωt2 , . . .) ∈ B}.

However, this is not possible since discontinuous functions are also included in the
set on the right side. 䊏

Next, we show that for any probability measure P on (, F ), the inner measure P of
c is zero. Recall that if A is any subset of , then the outer measure of A is defined by

P̄ (A) = inf P (E) : A ⊆ E = ∪j Ej where Ej ∈ C .

The inner measure P (A) is defined by P (A) = 1 – P̄ (Ac ).

Proposition 1.1.9 P (c ) = 0.
Proof We will show that P̄(cc ) = 1. Let (cc ) ⊆ E = ∪j Ej where Ej ∈ C for all j. Let
S = {s1 , s2 , . . .} ⊆ T be a countable base for E so that there exists a Borel B ∈ B(R∞ )
such that

E = {ω : (ωs1 , ωs2 , . . .) ∈ B}. (1.1.2)

If ω∗ ∈  is any continuous, real-valued function on T, then let ω0 be a discon-

tinuous function on T whose values coincide with the values of ω∗ on S. It follows
that ω∗ ∈ E, since the right side of (1.1.2) depends only on the function values on
the countable set S. Thus, E = , and hence P (E) = 1, which yields that, P̄ (cc ) = 1.
䊏

Another example of a set with zero inner measure is given by

m = {ω ∈  : ω (t) is a Lebesgue measurable function of t}.

The proof is similar to the one given above and consists in showing that P̄ (cm ) = 1.
We will now construct a rather important measure P on (, F) when the index set
T = [0, ∞) by specifying the family of finite-dimensional measures as follows:

 
μt1 ,t2 ,...,tn (B1 × B2 · · · × Bn ) = ··· n
j=1 p tj – tj–1 ; xj – xj–1 dxn · · · dx1
B1 Bn
(1.1.3)

where 0 < t1 < t2 · · · < tn , Bj are any Borel sets in R, t0 = 0, x0 = 0, and


1 x2
p (t; x) = √ exp –
2π t 2t

for all t > 0 and any real number x. When t = 0, μ0 (B) = δ0 (B) for any Borel set B.
Random documents with unrelated
content Scribd suggests to you:
grave defect. He was conscious—more, he was proud—of being an
amateur soldier, and knowing himself to be modest, he did not fear
any comparison between the actual results obtained by English
amateurs like himself, and the far more largely professional armies
of other countries. And now these over-brainy ones had gone and
done it. He knew as well as anyone the hardships and dangers of
soldiering, had experienced them, shared them with the ranks, in the
trenches. Why even in this beastly Vanderlynden affair, it would have
puzzled him to say if he were more sorry than glad that the private
soldier had never been brought to Justice. But English—and even
Frenchmen—as he had seen with his own eyes, if they mutinied, got
over it, and went on. It was only people like the Russians that went
and pushed things to their logical conclusion.
He had a hatred of that, being subconsciously aware that the logical
conclusion of Life is Death. Naturally, from his upbringing and mental
outlook, he had no sympathy with the alleged objects and
achievements of the Russian Revolution. He could not see what
anyone wanted with a new social order, and as for the domination of
Europe by the Proletariat, if he understood it, he was all against it in
principle. He was against it because it was Domination. That was
precisely the thing that had made him feel increasingly antagonistic
to Germany and German ideas. It had begun long ago, during brief
continental holidays. He had met Germans on trains and steamers,
in hotels and on excursions. He had grudged them their efficient way
of sight-seeing, feeding and everything else. But he had grudged
them most their size and their way of getting there first. If it had not
been for that, he had a good deal more sympathy with them, in most
ways, than with the French. Subsequently he had found Germans
infringing on the business of his native town, selling cheaper, better-
tanned hides than its tanners, more scientifically compounded
manures than its merchants. Then they invaded politics and became
a scare at election times. And after the false start of 1911, in 1914
they had finally kicked over the tea-table of the old quiet comfortable
life. He did not argue about this. He had felt it simply, truly, directly.
Under all the hot-air patriotism and real self-sacrifice of August,
1914, it had been this basic instinct which had made him and all his
sort enlist. The Germans had asked for it, and they should darn-well
have it. If they didn’t they would go on asking. They were after
Domination.
That craze had started something that would be difficult now to stop.
Dormer saw very well that other people besides Russians might find
grievances and the same wrong-headed way of venting them. The
Russians would probably go on with their propaganda, all over the
world. The Germans, on the other hand, had probably set the
Japanese off. And so we should go on, all the aristocratic classes
calling for Domination by their sort, all the ultra-brainy democracies
calling for their particular brand.
So when he was passed as fit and told to rejoin the depôt of his
regiment, at a seaport town, he went without any panic fear of the
future, German or otherwise. He went with a deep conviction that
whatever happened, life had been cheapened and vulgarized. It was
not by any means mere theory. He had seen what sort of a home he
might hope to make after the Peace, with his mother or sisters, or if,
conceivably, he married. Not a bad home, his job would always be
there, and certain remnants of that bourgeois comfort that had grown
up in all the old quiet streets of the provincial towns of England
during the nineteenth century, privileged, aloof from the troubles of
the “continent,” self-contained. But remnants only, not nearly
enough. He and all his sort had been let down several pegs in the
social scale. Without any narrow spite, or personal grievance, he felt
that the Germans had caused this upset and the Russians had put
the finishing stroke to it, made it permanent, as it were. He happened
to be opposite the Germans in the particular encounter that was not
yet ended, and he was able to draw upon an almost inexhaustible
supply of obstinate ill-will.
He went to the depôt in its huts on a sandy estuary. It was
commanded by a Major of the usual type, and no one knew better
than Dormer how to keep on the right side of such a one. He was, of
course, a Godsend to the Major. He had all the practical experience
and none of the fussiness. He merely wanted the job finished. That
suited the Major exactly, who didn’t want it to finish in a hurry, but
wanted even less to have to find ideas for training troops. Dormer,
with his two and a half years in France, was the very man. He looked
trustworthy. He was set to instructing the raw material, of which the
camp was full. He disliked it intensely, but, as always, took what was
given him in his sober fashion and did his limited best with it. He was
amazed to find such reserves of men still untouched. His own
recollections of early 1915 were of camps filled with an eager
volunteer crowd of all ages and conditions, who were astounded
when it was suggested to them that certain of them ought to take a
commission. Now he found that his sort went a different way, direct
to O.T.C. or Cadet Corps. There was a permanence about the camp
staff that he had never seen in the old days. But most of all he was
impressed with the worn appearance of the camp. Thousand after
thousand had passed through it, been drafted overseas, and
disappeared. Thousand after thousand had followed. In the town and
at the railway, there were no longer smiles and encouragement.
People had got painfully used to soldiers, and from treating them as
heroes, and then as an unavoidable, and profitable incident, had
come to regard them chiefly as a nuisance. He forgot how he had
wondered if the men would stand it, he forgot how often he had
heard the possibility of an early Peace discussed. He began to
wonder now if people at home would stand it—the lightless winter
nights, the summer full of bombing, the growing scarcity of comforts,
the queues for this, that, and the other, the pinch that every gradually
depleted family was beginning to feel, as one after another of its
members had to go. He had been so long out of all this, up against
the actual warfare, glad enough of small privileges and of the
experience that enabled him to avoid the more onerous duties, the
worst sorts of want, that he only now began to realize what he had
never grasped, in his few short leaves, that there was still quite a
considerable, probably the greater portion of the nation, who did not
share his view of the necessity of going on. Another avenue of
speculation was opened to him. What if all the people at home made
Peace behind the backs of the Armies. Yet, being Dormer, he did not
submit to this home-grown philosophy. He just went on and did the
next thing that his hand found to do.
Of one thing he became pretty certain. All these people at home had
“got the wind up.” He didn’t know which were the worst, the lower
middle class, who were beginning to fear invasion, as a form of
damage to their shops and houses. He thought of those ten
departments of France that were either occupied by, or shot over, by
the Germans. Or again the newspapers, with their scare-lines, their
everlasting attempt to bring off this or that political coup. Or again the
people in power, who were keeping this enormous number of troops
in England, presumably to defend the beaches of the island from an
armed landing. He had become during the three years that had
contained for him an education that he could not otherwise have got
in thirty, a more instructed person.
An offensive was an offensive, could be nothing more or less. Every
offensive had been a failure except for some local or temporary
object, and in his opinion, always must be a failure. The idea of an
offensive conducted across a hundred leagues of sea made him
smile. It was hard enough to get a mile forward on dry land, but
fancy the job of maintaining communications across the water! He
attended enough drills to fill in the time, organized the football of the
Brigade to his liking and let it go at that. At moments he was tempted
to apply to be sent to France, at others to try and join one of these
Eastern expeditions, Salonika, Palestine or Mespot. But the certainty
of being more bored and of being farther than ever from the only life
he cared for, made him hesitate. He hesitated for two long months.
Then on the 21st March he was ordered by telegram to proceed to
France. He felt, if anything, a not unpleasant thrill. With all his care,
he had not been able to dodge boredom altogether. The depôt camp
had also been much too near the scenes of his pre-War life. He had
gone home, as a matter of duty, for several week-ends and had
always returned finely exasperated, it was so near to and yet so far
from home as he had pictured it, in his dreams. Now, here was an
end to this Peace-time soldiering. The news, according to the
papers, seemed pretty bad, but he remembered so well the awful
scurry there was for reinforcements on the morning that the nature of
the Second Battle of Ypres became known. This could not be so
desperate as that was. Practically the whole of the rank and file in
the depôt were under orders. He took jolly good care not to get
saddled with a draft, and spent the night in London. People were in a
rare stew there. He had a bath and a good dinner and left it all
behind. He took a little more note of the traffic at the port of
embarkation. On the other side, he found lorries waiting and went
jolting and jamming away up to Frecourt, forty miles. He rather
approved. It looked as though our people were waking up.
At Corps reinforcement camp—a new dodge evidently—he got
posted to a North Country battalion; and proceeded to try and find
their whereabouts. He was told that they were going to Bray, but it
took him some time to understand that they were falling back on that
place. When, by chance, he hit upon the Division to which they
belonged, they were on the road, looking very small, but intact and
singing. He soon found plenty to do, for he grasped that practically
the whole battalion was composed of reinforcements, and had only
been together two or three days. They set to work at once to
strengthen some half-completed entrenchments, but after two days
were moved back again.
It was during those two days that he saw what he had never to that
moment beheld, an army in retreat. The stream of infantry, artillery
and transport was continuous—here in good formation, there a mere
mass of walking wounded mixed up with civilians, as the big
hospitals and the small villages of the district turned out before the
oncoming enemy. He thought it rotten luck on those people, many of
whom had been in German hands until February, 1917, and had only
had a twelvemonth in their small farms, living in huts, and had now to
turn out before a further invasion. The bombardment was distinctly
nasty, he never remembered a nastier, but as usual, the pace of the
advance soon outdistanced the slow-moving heavy artillery, whose
fire was already lessening. He had no feelings of sharp despair, for
as he had foreseen, a modern army could not be crumpled up and
disposed of. What he did now anticipate, was any amount of
inconvenience.
Amiens, he gathered, was uninhabitable, that meant many good
restaurants out of reach. New lines of rail, new lateral
communications would be necessary, that meant marching. Just
when they had begun to get the trenches fairly reliable, they were
entrained and sent wandering all round the coast. The wonderful
spring weather broke with the end of March, as the weather always
did, when it had ceased to be of any use to the Bosche, and had he
been superstitious, he might have thought a good deal of that. It was
in a cold and rainy April that he found himself landed on the edge of
the coal-fields, behind a canal, with a slag heap on one side of him,
and a little wood on the other, amid an ominous quiet.
The company of which he had been given command was now about
a hundred and fifty strong and he had done what little he could to
equalize the four platoons. He had one officer with him, a middle-
aged Lieutenant called Merfin, of no distinguishable social status, or
local characteristics. The day when a battalion came from one town
or corner of a county, under officers that were local personages in
the civil life of its district, was long past. Dormer placed his second-
in-command socially as music-hall, or pawnbroking, but the chap
had been out before and had been wounded, and probably knew
something of the job. The men were satisfactory enough, short,
stumpy fellows with poor teeth, but exactly that sort of plainness of
mind that Dormer appreciated. They would do all right. Perhaps a
quarter of them had been out before, and the remainder seemed
fairly efficient in their musketry and bombing, and talked pigeons and
dogs in their spare time, when not gambling.
The bit of line they held was Reserve, a bridge-head over the canal,
a strong point round a half-demolished château in the wood, and
some wet trenches to the right, where the next battalion joined on.
Battalion Head-quarters was in a farm half a mile back. Dormer and
Merfin improvised a Mess in the cellar of the Château, saw that the
cooker in the stables was distributing tea, and let all except the
necessary guards turn in. He had some machine gunners at the
strong point, and across the canal were two guns, whose wagons
had just been up with rations and ammunition. His own lot of rations
came soon after and he told Merfin to take the first half of the night,
and rolled himself in his coat to sleep.
As he lay there, listening to the scatter of machine-gun fire, and the
mutter of officers’ servants in the adjoining coal-hole, watching the
candle shadows flicker on the walls that had been whitewashed, as
the draught stirred the sacking over the doorway, his main thought
was how little anything changed. Two and a half years ago he had
been doing exactly the same thing, a few miles away, in the same
sort of cellar, in front of an enemy with the same sort of advantage in
ground and initiative, machine guns and heavy artillery. He was as
far from beating the Germans as ever he had been. He supposed
that practically all the gains of 1916 and 1917 south of Arras had
been lost. On the other hand, the Germans, so far as he could see,
were equally far from winning. What he now feared was, either by
prolonged War or premature Peace, a continuance of this sort of
thing. And slowly, for he was as mild and quiet-mannered a man as
one could find, his gorge began to rise. He began to want to get at
these Germans. It was no longer a matter of principle, a feeling that
it was his duty as it had been in the days when he enlisted, took a
commission, and had come to France. He was no longer worrying
about the injustice of the attack on Belgium or the danger of a
Germany paramount in Europe. He had now a perfectly plain and
personal feeling. But being Dormer, this did not make him cry out for
a sortie en masse like a Frenchman, nor evolve a complicated and
highly scientific theory as to how his desire was to be realized. The
French and Portuguese who fought beside him would have found
him quite incomprehensible. The Germans actually invented a logical
Dormer whom they had to beat, who was completely unlike him. If he
had any ideas as to what he was going to do, they amounted to a
quiet certainty that once the enemy came away from his heavy and
machine guns, he, Dormer, could do him in.
So he went on with the next thing, which was to turn over and sleep.
He woke, sitting bolt upright, to the sound of two terrific crashes. One
was right over his head. The candle had been blown out, and as he
struggled out of the cellar, barking his shins and elbows, he was
aware that the faint light of the sky was obscured by a dense cloud
all round him. Instinctively he pulled up his gas mask, but the sound
of falling masonry and the grit he could taste between his lips,
reassured him. It was a cloud of brick dust. Across the canal, the
barrage was falling on the front lines with the thunder of a waterfall.
The Bosche had hit the Château, and if he were not mistaken, had
put in another salvo, somewhere near by. At the gate of the little
park-like garden he ran into a figure he recognized for Merfin, by the
red light of the battle, just across the canal.
“What is it?”
“Aw—they’ve knocked in the bridge!”
“Every one standing-to?”
“Can’t help ’emselves.”
They went to look at the damage. The bridge was a small, one
vehicle affair, with steel lattice sides, and an asphalt roadway. The
bridge piers at the near end had been blown away, and the whole
had settled down some four or five feet, on to the mud of the tow-
path.
“Can you get across?”
“Aw—yes—easy!”
“Better get across and wait a bit!”
He himself went back to find up his stretcher bearers, who, he had
always noticed, wanted an order to get them in motion. The guard on
the bridge was dead so far as he could see, but some one was
shouting, behind, at the Château.
He found the C.S.M with two men digging out the servants whose
coal-cellar had been blocked. One of them was badly crushed, but
his own man only shaken. Then there were horses on the road.
Gunners, trying to get their teams up to the advanced guns.
Hopeless, of course. Then came a runner from battalion. Send
Merfin with two platoons. He saw to that, and rearranged his
depleted company. It took some time. The barrage appeared to be
creeping nearer. The ground shook with the continuous concussion
and whiffs of gas were more and more noticeable, but the heavier
stuff was already falling farther to the rear. Then came a runner from
across the bridge. There was a crowd on the road. Dormer went and
found just what he expected. Walking wounded and those who
wanted to be treated as such. He sorted them out, directing the
former down the road to the dressing station, and setting the others
to dig. If he had got to hang on to this place, and he supposed he
had, he meant to have some cover. The stream of people across the
broken bridge increased. Trench mortars and machine gunners,
platoons of his own regiment. The Bosche was “through” on the left,
and they were to come back behind the canal. The barrage died out,
to confirm this. The machine-gun fire came nearer and nearer.
In the cold grey light of a wet April dawn, a tin-helmeted figure
dashed up on a borrowed motor-cycle. It was the Brigade Major.
What had Dormer got? He heard and saw, and took a platoon and all
the sundries. His last words were: “Hang on here, whatever you do!”
Dormer heard the words without emotion. He realized that it meant
that he was expected to gain time. He got hold of his Sergeant, and
overhauled the rations and ammunition. They were not too badly off,
and the cooker lay stranded in the stable yard. That meant hot water,
at least. He took a turn round the place. The Château grounds had
once been wired as part of some forgotten scheme of defence of
1915 or early 1916. That was all right. On the other hand, the
“bridge-head”—a precious half-boiled concoction—was full of gas
and the barrier on the road blown away.
He got his few men out of it, with their several casualties, and started
them carting brick rubble from the dilapidations of the Château to
make an emplacement for a machine-gun on the near side of the
bridge. He stood looking at the road by which the Bosche must come
—a mere lane that led from one of the neighbouring coal-pits, and
was used, he imagined, for transport of coal that was required
locally. It meandered out of sight, among low fenceless fields, until
the shallow undulations of the ground hid it. In the distance was the
steamy reek of last night’s battle, but nothing that moved, amid the
silence broken only by long-distance shots, and fusillade somewhere
on the left. Then, down that road he saw a party advancing, led by
an officer. There was no doubt that they wore khaki. He waited by
the bridge for them, and shouted directions to them how to cross. He
got an answer:
“Hallo, you old devil, what are you doing?”
It was that Kavanagh. There had been an advanced signal
exchange, and he had gone to bring his men in. They were tired,
hungry and disgusted, but Kavanagh had the jauntiness of old. He
wasn’t going back to Division, he was going to stay with dear old
Dormer, and see this through. Dormer thought a moment, then said:
“All right.”
“All right. I should think so. I don’t suppose I could catch Division,
even on a motor-byke. They must be nearly at Calais. It’s all rot. The
Bosche are done!”
“Are they?”
“Sure. What are they waiting for now?”
“Bringing up their artillery?”
“That won’t blow the water out of the canal.”
“Possibly not. But we may as well have some food while it’s
possible.”
“You old guts. Always eating!”
“Yes, when I can. Aren’t you?”
“Now, Dormer. You know me better than that. Glory is my manna.”
“Will you take cold bully and tea with it?” asked Dormer as they
dropped into the cellar.
Kavanagh made no objection, and they ate in silence, fast, for ten
minutes. Then they saw the men were being fed, and relapsed, in
their hiding-place, into pipes, and whisky out of Kavanagh’s flask.
“How did you get into this show?” Dormer asked.
“The Division—your old Division, my boy, left me here to hand over!
They might have spared themselves the trouble. But I’d got a most
lovely scheme of lateral communication. Corps gave me a lot of
sweet words about it. I suppose I shall get the M.C. Now the silly old
Hun has gone and blown it all to bits. What about you?”
“You know I got wrong and was sent home sick.”
“I heard all that. It was about that Vanderlynden affair, wasn’t it?”
“It was!”
“Well, you’ve no idea what a sensation you created. Vinyolles got
simply wet behind the ears with it. Some French Deputy said, after
the Somme show, that English troops did more damage to France
than to Germany. Of course every one on Divisional H.Q. has
changed in the last few months. They all established an alibi or
Habeas Corpus or something. It was one of the things that made the
French Press go for unity of command! You were a boon to them!”
“I wish them joy of the business. I don’t know why you mix me up
with it.”
“Why, it was your pet show, wasn’t it?”
“It got fathered on to me because I could understand what it was
about.”
“Yes, you told Vinyolles, didn’t you?”
“The ignorant brute asked me.”
“I know. He’s all fresh. I find him trying also. Well, he knows all about
it now.”
“Tell you the truth, I’ve no idea what I said, Kavanagh! I was feeling
queer!”
“Vinyolles thought you’d gone potty.”
“He wasn’t far wrong.”
“He said you told him the whole British Army was guilty of the
Kerrime at Vanderlynden’s!”
It was the first time Dormer had heard it called that.
“Well, in a sense, so they are.”
“In a sense, War is a foolish business!”
“I thought you liked it?”
“I was trying to talk like you——”
Before Dormer could reply, the sacking over the door was lifted, by
Dormer’s Sergeant.
“Cop’l Arbone is back, sir!”
“Very good. Did he get in touch with the Major?”
“He only found a Lewis-gun section, sir. The Major moved most of
the men along the canal, where there’s more trouble!”
“All right!”
“Well, I suppose I may as well go and have a look at my lot,”
Kavanagh stretched himself.
“I told ’em to hunt round and see if they could get this place wired
up!”
“Umpteenth Corps ought to have thought of that, long ago!”
“Did you ever know Corps think of anything.”
While Kavanagh was so engaged, Dormer took a turn round the
various guards and posts he had established. There appeared to be
fair cover from view, and even from small-arm and field-gun fire. Of
course when the Bosche really wanted to get the place, nothing
Dormer and Kavanagh and some forty men could do would stop it. In
coming round to the stables behind the Château he found his
Sergeant with two men, laboriously trundling on a hand cart what he
soon verified to be slabs of marble. What would they think of next?
The explanation was, “There was a champion bathroom, sir, an’ I
thought we could set up our Lewis better with these!”
When Kavanagh saw what was going on, he laughed.
“More damage in billets, Dormer!”
“Well, the stuff will be smashed up anyhow, won’t it?”
“Two blacks don’t make a white. I understand why you told Vinyolles
the whole army was guilty. You’re doing just what your friend did
about his mules.”
“Why will you drag in that beastly business? This has nothing in
common with it.”
“To the common all things are common. You tell the owner of the
Château that when he finds out.”
Dormer was going to say “He won’t find out!” but refrained. He
disliked arguing. This seemed a particularly bad argument. Also, at
that moment, a Lewis gun began, just below. Then another. He went
to the garden wall, and peered out. Nothing visible, as usual. He
thought of all the battle pictures he had ever seen. The prancing
horses, the gay uniforms, the engrossing action of figures that
pointed muzzle or bayonet at each other, that wielded sword or
lance. Here he was, an incident in one of the biggest battles in the
world. All he could see was neglected arable, smashed buildings, a
broken bridge and a blocked by-road, all shrouded in steamy vapour.
He made out that it was the Lewis opposite the end of the bridge that
was firing. He crawled along the gully that had been dug from the
Château gate to the roadway, and so to the emplacement by the
step-off of the bridge. The Corporal in charge of the section turned to
him.
“Got ’im, sir!”
“What is it?”
“Bosche in the ditch, under them bushes!”
Dormer waited a moment, but nothing happened. He crawled back,
and sent his Sergeant round to see that every one was under cover.
Back in the cellar he found Kavanagh, and told him.
“I know. Once more into the breach!”
“It’s not poetry, Kavanagh. This is the start. Once they find we’re
stopping them here, they’ll shift us, you may bet!”
“I shouldn’t wonder. My lot are trying to get into touch with Brigade.
They’re running a line back behind the wood. There’s no one on our
left, as far as can be found.”
“Must be some one.”
“Why should there be? Brigade have probably moved by this time.”
“Ah, well, can’t be helped.”
No use telling the chap that it was all useless. He just sat down and
lit his pipe. He perceived clearly enough that they were being
sacrificed—just left there to hold the Bosche up for a few hours,
while the Division went back.
During the day there was sporadic machine gunning. The Bosche
was feeling his way for crossing the canal, but had found it far less
easy than in the sectors farther north. Tolerably certain that the main
attack would come at dawn, Dormer and Kavanagh got what rest
they could, though proper sleep was out of the question. Their
servants had found a well-upholstered sofa, and a superior brass
bedstead, which now adorned the cellar, causing Kavanagh to gibe
about damage in billets. Their vigil was lightened by the sounds of
song from the stables where such men as they had set apart as
reserves were lodged.
“Old soldiers never die,
They only fade away.”
to a well-known hymn tune, made Dormer home-sick, but delighted
Kavanagh.
“Listen to that!”
“I can’t help it, unless I send and stop them.”
“Never, man, never stop men who can sing at such a moment. It
means philosophy and courage!”
“It means foolishness and rum!”
“Dormer, I fear you are no born leader!”
“No, of course I wasn’t.”
“But you’ve got to lead men now, and lead ’em to victory.”
“I don’t mind much so long as I lead ’em to Peace!”
“Yes, but don’t you see, mere Peace will mean Revolution!”
“I don’t believe it. I saw that affair at Étaples. I saw the trouble
among the French troops in May. Those chaps prefer to take orders
from you and me rather than from their own sort.”
“How do you account for Russia, then?”
“I can’t. But it’s an object lesson rather than an example, I should
say.”
“You used not to talk like that. You used to say that the men wouldn’t
stand it.”
“I’ve lived and learned!”
“Both, I am sure.”
“You needn’t be so superior. No one knew what any of this would be
like until it was tried. We’ve something to go by, now! This War
depends on turning a crank. The side that goes on turning it
efficiently the longer will win. Our chaps look like lasting!”
“So do the Bosche. No, Dormer, you’re all wrong——”
At that moment a fresh burst of song came from the stables. A
Cockney voice to a waltz tune:
“Orl that I wawnt is larve,
Orl that I need is yew——”
“There,” cried Kavanagh, his voice rising into his excited croak.
“That’s what we want!”
Dormer did not reply. With dusk came a few long-range shots,
gradually broadening and deepening into a bombardment towards
dawn. Both of them had to be out and about all night. They had
several casualties, and the whole place reeked with gas. As the grey
light of another day began to change the texture of the shadows,
movement was discernible about the road. It was their chance and
with a higher heart and the feeling of relief, they were able to let
loose the Lewis guns, which they had managed to save intact. For
more than an hour, Dormer crawled from one to the other, seeing
that they did not overheat or jam, for the fact that they were killing
Germans pleased him. Then there was a slackening of fire on both
sides.
They waited and the suspense from being irksome, became
tolerable. There was a good deal of noise each side of them, and
Dormer began to wonder if his detachment were surrounded,
especially as the servants whom he had sent back to get into touch
with Brigade, had not returned. It was a dull rainy afternoon
prematurely dark, and the rain as it increased, seemed to beat down
the gunning, as water quenches a fire. He must have been in that
half-waking state that often superimposed on sleeplessness and the
awful din, when he was thoroughly roused by trampling in the trees
round the Château. He called to Kavanagh but got no reply. Then
there was a pushing and scrambling at the wall behind the stable,
and English cavalrymen came swinging over it. Dormer and
Kavanagh were relieved, and were shortly able to hand over and
prepare to march their command back to rejoin their Division, which,
depleted by four weeks of continual mauling, was being taken out of
the line.
The battle was by no means over. They next went in farther north,
and Dormer had the queer experience of going into trenches where
Corps H.Q. had been, of billeting in rooms where Major-Generals
had slept. Gradually he became aware of lessening tension, reduced
shelling, and slackened machine-gun fire, but it was the end of May
before he found, when sent to raid an enemy post, that there was no
one there. He had been right after all. The German offensive also
had failed. Anticlimax was the rule of the War. He was glad that he
had parted from Kavanagh, who had gone back to his proper job
with his Division, goodness knew where. He felt that the fellow would
remind him that for several hours while they lay together in those
scratched-out trenches round that little Château by the canal, he had
given up hope. He need not have bothered. If the Bosche could not
win on that day, he never would. Slowly now the British lines were
creeping forward. Then he found American troops behind him.
It was during this phase of things that he found himself upon familiar
ground. Except on Kavanagh’s lips, he had not heard of the crime at
Vanderlynden’s since before Christmas. It was now September. Here
he was, detrained and told to march to Hondebecq. He passed what
had been Divisional Head-quarters in 1916 and noticed the shell-
holes, the open, looted, evacuated houses. He passed along the
road which he and Major Stevenage had traversed all those years
ago. The Brigade were in Divisional Reserve, and were quartered in
a string of farms just outside the village. He looked at the map
squares attentively, but on the larger scale map he found it actually
marked Ferme l’Espagnole. Being Dormer, he just saw to the
billeting of his company and then learned that the Battalion Head-
quarters were located at the Vanderlyndens’, and had no difficulty in
finding good reason to walk over there, after tea.
The place was not much changed. It was soiled, impoverished,
battered by War, but the German advance, which had stopped dead
a few miles short of it, had been spent by the time it reached its limits
in this sector, and had early been pushed back. Trenches had been
dug and camouflage erected all round the place, but it had not
suffered damage except by a few long-distance shots, the routine of
trench warfare had never reached it. In the kitchen, darkened by the
fact that the glass was gone from the windows, which were blinded
with aeroplane fabric, stood the familiar figure of Mademoiselle
Vanderlynden. He asked for the Colonel, and was civilly directed to
the parlour on the other side of the door. Not a word of recognition,
hardly a second glance. He did not know if he were sorry or glad. He
would have felt some relief to hear that the claim that had caused all
that trouble had been settled. But he did not know what he might
bring down upon his head by inquiry and held his tongue. His
business with the Colonel was the usual regimental routine, nominal
and numerical rolls, reinforcements and indents, training and
movements. It did not take long. On his way out he passed the
kitchen door and said just:
“Good night, Mademoiselle!”
“Good night, M’sieu!” And then calmly: “They are going to pay us for
the damage to La Vierge!”
“I am glad to hear it.”
“I thought you would like to know. It has been a long time.”
“Yes, a long time. I hope it will soon be settled.”
“Ah, not yet. I know these offices at Boulogne! They have a good
deal to pay for, no doubt.”
“No doubt. Good night, Mademoiselle!”
“Good night, mon capitaine.”
Walking back to his billet, he had once more that sensation of
escape. Was he really going to get away from that business, this
time, for ever? True, Mademoiselle Vanderlynden seemed little
enough inclined to be vindictive. He could not help feeling that her
view of the affair was after all reasonable and just. She bore no
malice, she wanted things put right. Money would do it. She was
going to get the money, or so she seemed to think. She had no
animus against the man who had broken a piece of her property.
She had neither animus against nor consideration for himself, the
representative of the British Army, who had so signally failed to
hasten the question of compensation. She took it all as part of the
War, and she was seeing it correctly. It was the British Army that had
done it. Her home, where she was working so peacefully in 1914,
had become first a billet, then all but a battlefield. The Crime at
Vanderlynden’s was the War, nothing more nor less. That was
exactly what he felt about it. No damage had been done to any
furniture or valuables that he owned, but he had still to get out of it
with his body intact, and resume the broken thread of existence,
where it had been snapped off, all those four years ago. True he had
not been badly paid, but he had taken a considerable risk—it was
much more dangerous to be an officer than a private, more
dangerous to be a private than a civilian. She had gauged the whole
thing correctly, right down to the necessarily slow and complicated
process of getting it adjudicated by some set of fellows down by the
coast, who ran these things off by the hundred and had a whole set
of rules that had to be complied with. He turned at the end of the
farm road and took a look back at the old place. There were worse
billets than the Spanish Farm and people more awkward to deal with
than the Vanderlyndens. In the Somme he had come across farms
where they charged you for the water and people who removed
everything right down to the bedsteads. Vanderlynden had only
wanted to be paid for what was wantonly damaged. They were
French, you couldn’t expect them to be sympathetic about other
people’s mules. What a queer world it was, he would never have
suspected all the crotchets that human nature could present, had he
not been thrust nose-foremost into this infernal show.
All his philosophy forsook him, however, on entering the billet where
his company was lodged. The woman had been selling not merely
beer, which was connived at, but spirits, to the men. Two of them
had got “tight” and had been arrested, and he would have them up
before him in the morning. Then there would be the question as to
where she got the spirits from, whether some Quartermaster-
sergeant had been making away with the rum, or whether she had
induced some one to buy it for her at the Expeditionary Force
Canteen. It all came back to the same thing. Men kept under these
conditions too long.
No one had been more surprised than Dormer, when the Allied
Armies took up the initiative again in July, and appeared to keep it.
With a lugubrious satisfaction he found himself retracing the
advances in the Somme district of 1916. It was an ironical comment
on his hard-earned War-wisdom, two years devoted to doing
precisely the same thing at precisely the same place. Of course, he
had learned some lessons, but his estimate of one hundred and
eighty years was still too small. But when the movement became
perpetual and he found himself on ground he no longer recognized,
among villages that showed all the signs of methodical German
occupation, he began to wonder. A slight wound in the forearm threw
him out of touch for a week or two, and when he went back, he found
himself in a more northern sector again, and for the first time found
cavalry in front of him. It suited him all right, he didn’t want to have
the job of bombing out little nests of machine gunners, that marked
each step in the line of advance. His feelings were pretty generally
shared. Men began to ask themselves whether there was any glory
in being knocked out at the moment of victory. When his battalion
was again obliged to move in advance of the cavalry, against
obstacles which, although always evacuated, were out of the sphere
of cavalry tactics, he found for the first time a definite unwillingness
among his command to obey orders in any but the most perfunctory
manner.
He had sufficient sense to see that it was very natural. In the early
days the job had been to keep men under cover, to avoid useless
and wasteful casualties. The lesson had been learned at length with
a thoroughness that he could never have instilled. The old, old boast
of the Territorial Colonel who had first enlisted him, and whose
tradition was actually of pre-Territorial days, from the period of the
Volunteers of before the Boer War, was far better founded than he
had ever supposed. He had been inclined to scoff when he had
heard the old boy talk: “Our motto was Defence not Defiance!” He
did not scoff now. It was deeply, psychologically true. The army that
had survived was an army that had been made to fight without much
difficulty, while its back was to the sea, with the knowledge that
trenches lost meant worse, if possible, conditions of existence, and it
was moved by some rags of sentiment, as to holding what one had
got; an army which displayed all the slowly aroused, almost passive
pugnacity of the English working class, so docile, yet so difficult to
drive out of a habit of mind, or an acquired way of living. They had
no real imperialism in them, none of the high-faultin’ Deutschland
über Alles, none of the French or Italian bitter revengefulness, nor
peasant passion for acquisition. The Rhine had never figured in their
primary school education. They had no relatives groaning under
Austrian or German domination—no rancorous feelings bred from
the attempt to force alien language or unassimilated religious forms
down their throats.
He had always regarded the boast about an Englishman’s House
being his Castle as so much claptrap. He knew by daily experience
of business, that any Englishman was governed by economic
conditions. Religious and racial tyranny were so far removed from
the calculations of all his sort, and all above and below it, that the
very terms had ceased to have any meaning. This War had no effect
on the lightly borne if real tyranny of England, the inexorable need to
get a permanent job if possible and keep it, with constant anxiety as
to the tenure of one’s lodging, and the prospect of old age. These
fellows who fell in with blank unmeaning faces, in which there was
no emotion, and who marched with the same old morose jokes, and
shyly imitated the class standards which he and those like him
handed down to them from the fount of English culture and fashion in
the Public Schools, had done what they had promised to do, or had
(the late comers) been conscripted to do. They had engaged or been
called up for duration. That was a typically English slogan for a
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!

ebookfinal.com