Multivariate Approximation V Temlyakov download

https://ebookbell.com/product/multivariate-approximation-v-
temlyakov-57410534

Explore and download more ebooks at ebookbell.com

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

Multivariate Approximation For Solving Ode And Pde Clemente Cesarano

Editor

https://ebookbell.com/product/multivariate-approximation-for-solving-
ode-and-pde-clemente-cesarano-editor-51685468

Multivariate Approximation For Solving Ode And Pde Clemente Cesarano

https://ebookbell.com/product/multivariate-approximation-for-solving-
ode-and-pde-clemente-cesarano-50655224

Multivariate Approximation And Applications 1st N Dyn D Leviatan

https://ebookbell.com/product/multivariate-approximation-and-
applications-1st-n-dyn-d-leviatan-1644914

Multivariate Approximation Theory Iii Proceedings Of The Conference At

The Mathematical Research Institute At Oberwolfach Black Forest 1st
Walter Schempp Karl Zeller

https://ebookbell.com/product/multivariate-approximation-theory-iii-
proceedings-of-the-conference-at-the-mathematical-research-institute-
at-oberwolfach-black-forest-1st-walter-schempp-karl-zeller-12060418
Recent Progress In Multivariate Approximation 4th International
Conference Wittenbommerholz Germany September 2000 1st Edition Manfred
Reimer

https://ebookbell.com/product/recent-progress-in-multivariate-
approximation-4th-international-conference-wittenbommerholz-germany-
september-2000-1st-edition-manfred-reimer-4210154

Topics In Multivariate Approximation And Interpolation Kurt Jetter

https://ebookbell.com/product/topics-in-multivariate-approximation-
and-interpolation-kurt-jetter-4426518

Modern Developments In Multivariate Approximation 5th International

Conference Wittenbommerholz Germany September 2002 1st Ed Werner
Haussmann

https://ebookbell.com/product/modern-developments-in-multivariate-
approximation-5th-international-conference-wittenbommerholz-germany-
september-2002-1st-ed-werner-haussmann-10489326

Multivariate Polynomial Approximation 1st Edition Manfred Reimer Auth

https://ebookbell.com/product/multivariate-polynomial-
approximation-1st-edition-manfred-reimer-auth-4210670

Nonasymptotic Analysis Of Approximations For Multivariate Statistics

1st Ed Yasunori Fujikoshi

https://ebookbell.com/product/nonasymptotic-analysis-of-
approximations-for-multivariate-statistics-1st-ed-yasunori-
fujikoshi-22477300
 CAMBRIDGE MONOGRAPHS ON
A P P L I E D A N D C O M P U TAT I O NA L
M AT H E M AT I C S

Series Editors
M . A B L O W I T Z , S . D AV I S , J . H I N C H ,
A . I S E R L E S , J . O C K E N D O N , P. O LV E R

32 Multivariate Approximation
The Cambridge Monographs on Applied and Computational Mathematics series reflects the crucial
role of mathematical and computational techniques in contemporary science. The series publishes
expositions on all aspects of applicable and numerical mathematics, with an emphasis on new devel-
opments in this fast-moving area of research.
State-of-the-art methods and algorithms as well as modern mathematical descriptions of physical
and mechanical ideas are presented in a manner suited to graduate research students and professionals
alike. Sound pedagogical presentation is a prerequisite. It is intended that books in the series will serve
to inform a new generation of researchers.

A complete list of books in the series can be found at

www.cambridge.org/mathematics.
Recent titles include the following:

17. Scattered data approximation, Holger Wendland

18. Modern computer arithmetic, Richard Brent & Paul Zimmermann
19. Matrix preconditioning techniques and applications, Ke Chen
20. Greedy approximation, Vladimir Temlyakov
21. Spectral methods for time-dependent problems, Jan Hesthaven, Sigal Gottlieb & David Gottlieb
22. The mathematical foundations of mixing, Rob Sturman, Julio M. Ottino & Stephen Wiggins
23. Curve and surface reconstruction, Tamal K. Dey
24. Learning theory, Felipe Cucker & Ding Xuan Zhou
25. Algebraic geometry and statistical learning theory, Sumio Watanabe
26. A practical guide to the invariant calculus, Elizabeth Louise Mansfield
27. Difference equations by differential equation methods, Peter E. Hydon
28. Multiscale methods for Fredholm integral equations, Zhongying Chen, Charles A. Micchelli &
Yuesheng Xu
29. Partial differential equation methods for image inpainting, Carola-Bibiane Schönlieb
30. Volterra integral equations, Hermann Brunner
31. Symmetry, phase modulation and nonlinear waves, Thomas J. Bridges
32. Multivariate approximation, V. Temlyakov
Multivariate Approximation

V. T E M LYA K O V
University of South Carolina,
Steklov Institute of Mathematics, Moscow
and
Lomonosov Moscow State University
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781108428750
DOI: 10.1017/9781108689687
© V. Temlyakov 2018
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2018
Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A.
A catalogue record for this publication is available from the British Library.
ISBN 978-1-108-42875-0 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy
of URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
 Contents

Preface page ix

1 Approximation of Univariate Functions 1

1.1 Introduction 1
1.2 Trigonometric Polynomials 6
1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz
Theorem 17
1.4 Approximation of Functions in the Classes Wq,r α and Hqr 25
1.5 Historical Remarks 34

2 Optimality and Other Properties of the Trigonometric System 36

2.1 The Widths of the Classes Wq,r α and Hqr 36
2.2 Further Properties of the Trigonometric System 54
2.3 Approximation of Functions with Infinite Smoothness 62
2.4 Sampling and Numerical Integration 73
2.5 Historical Remarks 79

3 Approximation of Functions from Anisotropic Sobolev

and Nikol’skii Classes 81
3.1 Introduction 81
3.2 Trigonometric Polynomials 82
3.3 The Bernstein–Nikol’skii Inequalities and Their
Applications. A Generalization of the
Marcinkiewicz Theorem 89
3.4 Approximation of Functions in the Classes Wq,r α and Hqr 104
3.5 Estimates of the Widths of the Sobolev and Nikol’skii Classes 113
3.6 Sampling and Numerical Integration 121
3.7 Historical Remarks 126
vi Contents

4 Hyperbolic Cross Approximation 129

4.1 Introduction 129
4.2 Some Special Polynomials with Harmonics in
Hyperbolic Crosses 138
4.3 The Bernstein–Nikol’skii Inequalities 151
4.4 Approximation of Functions in the Classes Wrq,α and Hrq 162
4.5 Some Further Remarks 186
4.6 Historical Comments 189
4.7 Open Problems 190
5 The Widths of Classes of Functions with Mixed Smoothness 191
5.1 Introduction 191
5.2 The Orthowidths of the Classes Wrq,α and Hrq 193
5.3 The Kolmogorov Widths of the Classes Wrq,α and Hrq 216
5.4 Universality of Approximation by Trigonometric
Polynomials from the Hyperbolic Crosses 231
5.5 Historical Remarks 241
5.6 Open Problems 242
6 Numerical Integration and Approximate Recovery 244
6.1 Introduction 244
6.2 Cubature Formulas and Discrepancy 246
6.3 Optimal Cubature Formulas and Nonlinear Approximation 253
6.4 Lower Estimates 262
6.5 The Fibonacci Cubature Formulas 272
6.6 The Korobov Cubature Formulas 284
6.7 The Frolov Cubature Formulas 289
6.8 Universal Cubature Formulas 302
6.9 Recovery of Functions 305
6.10 Historical Notes, Comments, and Some Open Problems 315
6.11 Open Problems 320
7 Entropy 321
7.1 Introduction. Definitions and Some Simple Properties 321
7.2 Finite-Dimensional Spaces. Volume Estimates 323
7.3 Some Simple General Inequalities 325
7.4 An Inequality Between Entropy Numbers and Best m-Term
Approximations 328
7.5 Volume Estimates for Balls of Trigonometric Polynomials 333
7.6 Entropy Numbers of the Balls of Trigonometric Polynomials 345
7.7 Entropy Numbers for the W-Type Function Classes 363
 Contents vii

7.8 Entropy Numbers for the H-Type Function Classes 373

7.9 Discussion and Open Problems 380
7.10 Some Historical Comments 383
8 Greedy Approximation 387
8.1 Introduction 387
8.2 The Trigonometric System 394
8.3 Wavelet Bases 398
8.4 Some Inequalities for the Tensor Product of Greedy Bases 404
8.5 Weight-Greedy Bases 412
8.6 The Weak Chebyshev Greedy Algorithm 415
8.7 Sparse Approximation With Respect to General Dictionaries 422
8.8 Open Problems 447
9 Sparse Approximation 449
9.1 Introduction 449
9.2 Constructive Sparse Trigonometric Approximation 454
9.3 Constructive Sparse Trigonometric Approximation for Small
Smoothness 472
9.4 Open Problems 495
9.5 Concluding Remarks 496
Appendix Classical Inequalities 500
References 520
Index 532
 Preface

The twentieth century was a period of transition from univariate problems

(i.e., single-variable problems) to multivariate problems in a number of areas
of mathematics. In many cases this step brought not only new phenomena
but also required new techniques. In some cases even the formulation of a
multivariate problem requires a nontrivial modification of a univariate problem. For
instance, the problem of the convergence of the multivariate trigonometric series
immediately encounters the question of which partial sums we should consider;
there is no natural ordering in the multivariate case. In other words: what is a natural
multivariate analog of the trigonometric polynomials? In answering this question
mathematicians have studied different generalizations of the univariate trigono-
metric polynomials: those with frequencies from a ball, a cube, a hyperbolic cross.
Multivariate problems turn out to be much more difficult than their univariate
counterparts. The main goal of this book is to demonstrate the evolution of theoret-
ical techniques from the univariate case to the multivariate case. We do this using
the example of the approximation of periodic functions. It is justified historically
and also it allows us to present the ideas in a concise and clear way. In many cases
these ideas can be successfully used in the nonperiodic case as well.
We concentrate on a discussion of some theoretical problems which are impor-
tant in numerical computation. The fundamental problem of approximation theory
is to resolve a possibly complicated function, called the target function, into sim-
pler, easier to compute, functions called approximants. Generally, increasing the
resolution of the target function can be achieved only by increasing the complexity
of the approximants. The understanding of this tradeoff between resolution and
complexity is the main goal of approximation theory. Thus the goals of approxima-
tion theory and numerical computation are similar, even though approximation the-
ory is less concerned with computational issues. Approximation and computation
are intertwined and it is impossible to understand fully the possibilities in numer-
ical computation without a good understanding of the elements of approximation
x Preface

theory. In particular, good approximation methods (algorithms) from approxima-

tion theory find applications in image processing, statistical estimation, regularity
for PDEs, and other areas of computational mathematics.
We now give a brief historical overview of the challenges and open problems
in approximation theory, with emphasis on multivariate approximation. It was
understood at the beginning of the twentieth century that the smoothness properties
of a univariate function determine the rate of approximation of this function by
polynomials (trigonometric in the periodic case and algebraic in the nonperiodic
case). A fundamental question is: what is a natural multivariate analog of the uni-
variate smoothness classes? Different function classes have been considered in the
multivariate case: isotropic and anisotropic Sobolev and Besov classes, classes of
functions with bounded mixed derivative, and others. The next fundamental ques-
tion is: how do we approximate functions from these classes? Kolmogorov intro-
duced the concept of the n-width of a function class. This concept is very useful
in answering the above question. The Kolmogorov n-width is a solution to an opti-
mization problem where we optimize over n-dimensional linear subspaces. This
concept allows us to understand which n-dimensional linear subspace is the best for
approximating a given class of functions. The rates of decay of the Kolmogorov
n-width are known for the univariate smoothness classes; in some cases, exact
values are known. The problem of the rates of decay of the Kolmogorov n-width for
the classes of multivariate functions with bounded mixed derivatives is still an open
problem. We note that the function classes with bounded mixed derivatives are not
only an interesting and challenging object for approximation theory but are also
important in numerical computations. M. Griebel and his group have used approx-
imation methods designed for these classes in elliptic variational problems. Recent
work of H. Yserentant on new regularity models for the Schrödinger equation
shows that the eigenfunctions of the electronic Schrödinger operator have a
certain mixed smoothness similar to that of the bounded mixed derivative. This
makes approximation techniques developed for classes of functions with bounded
mixed derivatives a proper choice for a numerical treatment of the Schrödinger
equation.
Following the Kolmogorov idea of the n-width the optimization approach to
approximation problems has been accepted as a fundamental principle. We follow
this fundamental principle in this book. We will demonstrate the development of
ideas and techniques when we pass from the univariate case to the multivariate
case, and we will solve some optimization problems for the multivariate approxi-
mation. We find the correct order of decay for a number of asymptotic character-
istics of smoothness function classes. We provide the rate of decay, in the sense of
order, as a function of the complexity of the approximation method. For example,
in the case of approximation by elements of a linear subspace the complexity
parameter is the dimension of the subspace. In the case of numerical integration the
 Preface xi

complexity parameter is the number of points (knots) of the cubature formula.

Solving the problem of the correct (in the sense of order) decay of an asymptotic
characteristic, we need to solve two problems: (U) prove the upper bounds
and (L) prove the lower bounds. Very often the technique used to solve the
problem (U) is very different from the one for solving (L). We present both the
classical, well-known, techniques and comparatively recent modern techniques.
Usually, when we solve an approximation problem, we use a combination of
fundamental results (theories) and special tricks. In this book the fundamental
results from harmonic analysis, for instance, the Littlewood–Paley theorem, the
Marcinkiewicz multiplier theorem, the Hardy–Littlewood inequality, the Riesz–
Thorin theorem, and the Hausdorff–Young theorem are widely used. It turns out
that in addition to these classical results we need new fundamental results in
order to solve the multivariate approximation problems for mixed smoothness
classes. We develop and demonstrate this new technique here in detail. Let us
mention three examples of this new fundamental technique: (I) an embedding-type
inequality proved in §3.3.3 (see Theorem 3.3.6); (II) the volume estimates of sets
of coefficients of trigonometric polynomials (see §3.2.6 and Chapter 7); (III) the
greedy approximation (see Chapters 8 and 9).
Motivated by practical applications we study the following theoretical problem:
how do we replace in an optimal way an infinite-dimensional object (a function
class) by a finite or finite-dimensional object? The theory of widths, in particular,
the Kolmogorov widths, addresses the problem of approximation from a finite-
dimensional subspace. In addition to the Kolmogorov width we employ the linear
width and the orthowidth (Fourier width). In this study we use techniques (I) and
(II). Technique (II) has proved to be very useful in the proof of the lower bounds.
There are still open problems in finding the correct orders of decay of the above
widths in case of the mixed smoothness classes.
Discretization is a very important step in making a continuous problem compu-
tationally feasible. The problem of the construction of satisfactory sets of points in
a multidimensional domain is a fundamental problem of mathematics and, in par-
ticular, computational mathematics. We note that the problem of arranging points
in a multidimensional domain is also a fundamental problem in coding theory. It is
a problem on optimal spherical codes. This problem is equivalent to the problem
from compressed sensing on building large incoherent dictionaries in Rd . A very
interesting and difficult problem is to provide an explicit (deterministic) construc-
tion of a large system with small coherence. The optimal rate in this problem is
still unknown (see Temlyakov, 2011, Chapter 5, for further discussion).
A prominent example of a discretization problem, discussed in detail in this
book, is the problem of numerical integration. It turns out that, contrary to numer-
ical integration in the univariate case (see §2.4) and in the multivariate case of
anisotropic smoothness classes (see §3.6), where regular grid methods are optimal
xii Preface

(in the sense of order), in the case of the numerical integration of functions with
mixed smoothness regular grid methods are very far from being optimal. Numerical
integration in mixed smoothness classes requires deep number-theoretical results
for constructing optimal (in the sense of order) cubature formulas (see Chapter 6).
In addition to number-theoretical methods, technique III is also of use here.
Another example of a classical discretization problem is the problem of metric
entropy (covering numbers and entropy numbers). Bounds for the ε -entropy of
function classes are important in themselves and also have important connections
to other fundamental problems. For instance, the problem of the ε -entropy of some
classes of functions with bounded mixed derivatives is equivalent to the funda-
mental small ball problem from probability theory. This problem is still unresolved
in dimensions greater than two (see Temlyakov, 2011, Chapter 3, and Dinh Dung
et al., 2016 for further discussion). We obtain the correct orders of decay of the
entropy numbers of mixed smoothness classes in Chapter 7. Technique II plays a
fundamental role in proving the lower bounds.
The above discussion demonstrates that multivariate approximation theory in a
classical setting has close connections with other areas of mathematics and has
many applications in numerical computation. Recently, driven by applications in
engineering, biology, medicine, and other areas of science, new and challenging
problems have appeared. The common feature of these problems is very high
dimensions. Classical methods developed in multivariate approximation theory
may work for moderate dimensions, say, up to 40 dimensions. Many contemporary
numerical problems, however, have dimensions which are really large – sometimes
in the millions. Classical methods do not work for such enormous dimensions. This
is a rapidly developing and hot area of mathematics and numerical analysis, where
researchers are trying to understand which new approaches may work. A promising
contemporary approach is based on the concept of sparsity and nonlinear m-term
approximation. We present the corresponding results in Chapters 8 and 9 of this
book.
The fundamental question of nonlinear approximation is how to devise effec-
tive constructive methods (algorithms) of nonlinear approximation. This problem
has two levels of nonlinearity. The first is m-term approximation with regard to
bases. In this problem one can use the unique function expansion with respect to
a given basis to build an approximant. Nonlinearity enters by looking for m-term
approximants with terms (i.e. basis elements in the approximant) that are allowed
to depend on a given function. Since the elements of the basis used in the m-term
approximation are thus allowed to depend on the function being approximated,
this type of approximation is very efficient. On the second level of nonlinearity,
we replace a basis by a more general system, which is not necessarily minimal
(for example, a redundant system, or a dictionary). This setting is much more
 Preface xiii

complicated than the first (the bases case); however, there is a solid justification
due to the importance of redundant systems in both theoretical questions and in
practical applications. Technique III turns out to be very useful for approximation
at both levels of nonlinearity. In this book we are primarily interested in the trigono-
metric approximation. A very interesting phenomenon was observed recently. It
turns out that nonlinear algorithms, in particular the Chebyshev greedy algorithms,
designed for approximation with respect to redundant systems, work better than
algorithms, in particular the thresholding greedy algorithm, designed for bases
when they are applied to a trigonometric system. We discuss this phenomenon
in detail in Chapter 8.
Above we discussed a strategy based on the optimization principle, which we
will apply for finding optimal (in the sense of order) finite dimensional subspaces
(theory of the widths) and optimal (in the sense of order) discretization (numerical
integration, entropy). In addition to the optimization principle we study another
fundamental principle – universality. In Chapters 5 and 6 we illustrate the following
general observation. Methods of approximation and numerical integration which
are optimal in the sense of order for classes with mixed smoothness are universally
applicable for the collection of anisotropic smoothness classes. This gives an a
posteriori justification for the thorough study of classes of functions with mixed
smoothness. The phenomenon of saturation is well known in approximation theory
(DeVore and Lorentz, 1993, Chapter 11). The classical example of a saturated
method is the Fejér operator for approximation of the univariate periodic functions.
In the case of a sequence of Fejér operators, saturation means that the approx-
imation order by Fejér operators of order n does not improve over the rate 1/n
even if we increase the smoothness of the functions under approximation. Methods
(algorithms) that do not have the saturation property are called unsaturated. The
reader can find a detailed discussion of unsaturated algorithms in approximation
theory and in numerical analysis in the survey paper Babenko (1985). We point
out that the concept of smoothness becomes more complicated in the multivariate
case than it is in the univariate case. In the multivariate case a function may have
different smoothness properties in different coordinate directions. In other words,
functions may belong to different anisotropic smoothness classes (see Chapter 3).
It is known (see Chapter 3) that the approximation characteristics of anisotropic
smoothness classes depend on the average smoothness and that optimal approx-
imation methods depend on the anisotropy of classes. This motivated a study,
in Temlyakov (1988c) of the existence of an approximation method that works
well for all anisotropic smoothness classes. The problem is that of the existence
of a universal method of approximation. We note that the universality concept
in learning theory is very important and is close to the concepts of adaptation
and distribution-free estimation in nonparametric statistics (Györfy et al., 2002,
xiv Preface

Binev et al., 2005, Temlyakov, 2006). We discuss universality in approximation

theory in §5.4 and universality in numerical integration in §6.8.
We now give a brief description of the book by chapters.

Chapter 1. Approximation of Univariate Functions This chapter contains classi-

cal results of approximation theory: the properties of trigonometric polynomials
and approximation by trigonometric polynomials. The selection of the material for
this chapter was dictated by further developments and applications in the multi-
variate approximation.

Chapter 2. Optimality and Other Properties of the Trigonometric System This

chapter contains classical results on the Kolmogorov, linear, and Fourier widths
of the univariate smoothness classes. Discretization techniques and the fundamen-
tal finite-dimensional results are discussed in detail in this chapter. Also, classi-
cal results on the convergence of Fourier series are presented here. The book by
DeVore and Lorentz (1993) contains a comprehensive presentation of the univariate
approximation. The presentation in Chapters 1 and 2 is aimed towards multivariate
generalizations. It is somewhat close to the presentation in Temlyakov (1993b).

Chapter 3. Approximation of Functions from Anisotropic Sobolev and Nikol’skii

Classes This chapter is the first step from the univariate approximation to the
multivariate approximation. The approximation technique discussed here is mostly
similar to the univariate technique. Results of this type are typically presented in
books, such as Nikol’skii (1969), on function spaces. However, we include in this
chapter some nontrivial embedding-type inequalities and estimates for the volumes
of sets of Fourier coefficients of the multivariate trigonometric polynomials, which
are frequently used in further chapters.

Chapter 4. Hyperbolic Cross Approximation This is one of the main chapters

on the linear approximation theory of functions with mixed smoothness. In the
sense of its settings it is parallel to Chapter 1. This parallelism in settings allows
us to demonstrate a deep difference in technique between univariate polynomial
approximation and hyperbolic cross polynomial approximation.

Chapter 5. The Widths of Classes of Functions with Mixed Smoothness This is

the other main chapter on the linear approximation theory of functions with mixed
smoothness. The relation between this chapter and Chapter 2 is the same as that
between Chapters 4 and 1.
 Preface xv

Chapter 6. Numerical Integration and Approximate Recovery This is the third

main chapter on the linear approximation theory of functions with mixed smooth-
ness. This chapter is important from the point of view of applications. Also, the
technique for numerical integration developed in this chapter is very different from
that for univariate numerical integration. It is based on deep number-theoretical
constructions and on the general theory of greedy algorithms. The numerical inte-
gration of classes of functions with mixed smoothness has attracted a lot of atten-
tion recently. There are many books on discrepancy theory that are related to this
chapter. However, the development in terms of numerical integration is more gen-
eral than a discrepancy-type presentation. Roughly speaking, discrepancy theory
corresponds to the case of smoothness equal to 1 and equal weights in cubature
formulas, while numerical integration theory considers the whole range of smooth-
ness and general cubature formulas.

Chapter 7. Entropy This is the first chapter devoted to nonlinear approxima-

tion theory. We include here classical results on the entropy numbers of finite-
dimensional compacts. The main new ingredient of this chapter is a study of the
entropy numbers of classes of functions with mixed smoothness.

Chapter 8. Greedy Approximation This chapter contains a brief introduction to

greedy approximation in Banach spaces and a recent result on the Lebesgue-type
inequality for the Chebyshev greedy algorithm (CGA) with respect to special dic-
tionaries. In particular, this result implies that the CGA provides almost ideal (up to
a log m factor) m-term trigonometric approximation for all functions. Our intro-
duction to greedy approximation in Banach spaces follows the lines of Temlyakov
(2011). The Lebesgue-type inequality is also a recent result (see Temlyakov, 2014).

Chapter 9. Sparse Approximation This is one of the most important chapters

of the book. Our main interest in this chapter is to study sparse approximation
problems for classes of functions with mixed smoothness. We discuss in detail m-
term approximation with respect to the trigonometric system. We use techniques
based on a combination of results from the hyperbolic cross approximation, which
were obtained in the 1980s and 1990s (and are presented in Chapters 3–5 and 7),
and recent results on greedy approximation (given in Chapter 8) to obtain sharp
estimates for the best m-term approximation.

Appendix. The Appendix contains classical inequalities and results from har-
monic analysis that are often used in this text.
The book is devoted to the linear and nonlinear approximation of functions
with mixed smoothness. Both Temlyakov (1986c) and Temlyakov (1993b) contain
results on the linear approximation theory of such classes. At present there are
xvi Preface

no books on the nonlinear approximation theory of these classes. In addition to

the results treated in Temlyakov (1993b), we describe in this book substantial new
developments in the linear approximation theory of classes with mixed smoothness.
This makes the book the most complete text on the linear approximation theory of
these classes. Further, it is the first book on the nonlinear approximation theory
of such classes. The background material included in Chapters 1–3 makes the
book self-contained and accessible for readers with graduate or even undergraduate
level mathematical education. The theory of the approximation of these classes and
related questions are important and actively developing areas of research. There are
still many unresolved fundamental problems in the theory. Many open problems are
formulated in the book.

Acknowledgement
The work was supported by the Russian Federation Government Grant No.
14.W03.31.0031.
 1
Approximation of Univariate Functions

1.1 Introduction
The primary problem in approximation theory is the choice of a successful method
of approximation. In this chapter and in Chapter 2 we test various approaches,
based on the concept of width, to the evaluation of the quality of a method of
approximation. We take as an example the approximation of periodic functions of
a single variable. The two main parameters of a method of approximation are its
accuracy and complexity. These concepts may be treated in various ways depend-
ing on the particular problems involved. Here we start from classical ideas about the
approximation of functions by polynomials. After Fourier’s 1807 article the repre-
sentation of a 2π -periodic function by its Fourier series became natural. In other
words, the function f (x) is approximately represented by a partial sum Sn ( f , x) of
its Fourier series:
n
Sn ( f , x) := a0 /2 + ∑ (ak cos kx + bk sin kx),
k=1
 π  π
1 1
ak := f (x) cos kx dx, bk := f (x) sin kx dx.
π −π π −π

We are interested in the approximation of a function f by a polynomial Sn ( f ) in

some L p -norm, 1 ≤ p ≤ ∞. In the case p = ∞ we assume that we are dealing with
the uniform norm. As a measure of the accuracy of the method of approximating
a periodic function by means of its Fourier partial sum we consider the quantity
 f − S( f ) p . The complexity of this method of approximation contains the fol-
lowing two characteristics. The order of the trigonometric polynomial Sn ( f ) is the
quantitative characteristic. The following observation gives us the qualitative char-
acteristic. The coefficients of this polynomial are found by the Fourier formulas,
which means that the operator Sn is the orthogonal projector onto the subspace of
trigonometric polynomials of order n.
2 Approximation of Univariate Functions

In 1854 Chebyshev suggested representing continuous function f by its polyno-

mial of best approximation, namely, by the polynomial tn ( f ) such that
 
 n 
 
 f − tn ( f )∞ = En ( f )∞ := inf  f (x) − ∑ (αk cos kx + βk sin kx) .
αk ,βk  k=0

∞

He proved the existence and uniqueness of such a polynomial. We consider this

method of approximation not only in the uniform norm but in all L p -norms, 1 ≤
p < ∞. The accuracy of the Chebyshev method can be easily compared with the
accuracy of the Fourier method:
 
En ( f ) p ≤  f − Sn ( f ) p .

However, it is difficult to compare the complexities of these two methods. The

quantitative characteristics coincide but the qualitative characteristics are different
(for example, it is not difficult to understand that for p = ∞ the mapping f → tn ( f )
is not a linear operator).
 The Du Bois–Reymond 1873 example of a continuous
function f such that f − Sn ( f )∞ → ∞ when n → ∞, and the Weierstrass theorem

which says that for each continuous function f we have En ( f )∞ → 0 as n → ∞,
showed the advantage of the Chebyshev method over the Fourier method from the
point of view of accuracy.
The desire to construct methods of approximation which have the advantages of
both the Fourier and Chebyshev methods has led to the study of various methods
of summation of Fourier series. The most important among them from the point of
view of approximation are the de la Vallée Poussin, Fejér, and Jackson methods,
which were constructed early in the twentieth century. All these methods are linear.
For example, in the de la Vallée Poussin method a function f is approximated by
the polynomial

1 2n−1
Vn ( f ) := ∑ Sl ( f )
n l=n

of order 2n − 1.
From the point of view of accuracy this method is close to the Chebyshev
method; de la Vallée Poussin proved that
 
 f −Vn ( f ) ≤ 4En ( f ) p , 1 ≤ p ≤ ∞.
p

From the point of view of complexity it is close to the Fourier method, and the
property of linearity essentially distinguishes it from the Chebyshev method.
We see that common to all these methods is approximation by trigonometric
polynomials. However, the methods of constructing these polynomials differ: some
 1.1 Introduction 3

methods use orthogonal projections on to the subspace of trigonometric polynomi-

als of fixed order, some use best-approximation operators, and some use linear
operators.
Thus, the approximation of periodic functions by trigonometric polynomials
is natural and this problem has been thoroughly studied. The approximation of
functions by algebraic polynomials has been studied in parallel with approximation
by trigonometric polynomials. We now point out some results, which determined
the style of investigation of a number of problems in approximation theory. These
problems are of interest even today.
It was proved by de la Vallée Poussin (1908) that, for best approximation of the
function |x| in the uniform norm on [−1, 1] by algebraic polynomials of degree n,
the following upper estimate or bound holds:
 
en |x| ≤ C/n.
He raised the question of the possibility of an improvement of this estimate in the
sense of order. In other words, could the function C/n be replaced by a function
that decays faster to zero? Bernstein (1912) proved that this order estimate is sharp.

Moreover, he then established the asymptotic behavior of the sequence en |x|
(see Bernstein, 1914):
 
en |x| = μ /n + o(1/n), μ = 0.282 ± 0.004.
These results initiated a series of investigations into best approximations of
individual functions having special singularities.
At this stage of investigation the natural conjecture arose that the smoother a
function, the more rapidly its sequence of best approximations decreases.
In 1911 Jackson proved the inequality
En ( f )∞ ≤ Cn−r ω ( f (r) , 1/n)∞ .
The relations which give upper estimates for the best approximations of a func-
tion in terms of its smoothness are now called the Jackson inequalities, and in a
wider sense such relations are called direct theorems of approximation theory.
As a result of Bernstein’s (1912) and de la Vallée Poussin’s (1908, 1919) inves-
tigations we can formulate the following assertion, which is now called the inverse
theorem of approximation theory. If
En ( f )∞ ≤ Cn−r−α , 0 ≤ r integer, 0 < α < 1,
then f has a continuous derivative of order r which belongs to the class Lip α ; that
is, f ∈ W r H α (in the notation of this book it is the class H∞r+α ). Thus, the results
of Jackson, Bernstein, and de la Vallée Poussin show that functions from the class
W r H α , 0 < α < 1, can be characterized by the order of decrease of its sequences
of best approximations.
4 Approximation of Univariate Functions

We remark that at that time, early in the twentieth century, classes similar to
W r H α were used in other areas of mathematics for obtaining the orders of decrease
of various quantities. As an example we formulate a result of Fredholm (1903). Let
f (x, y) be continuous on [a, b] × [a, b] and
 
max f (x, y + t) − f (x, y) ≤ C|t|α , 0 < α ≤ 1.
x,y

Then for eigenvalues λ (J f ) of the integral operator

 b
(J f ψ )(x) = f (x, y)ψ (y)dy
a

the following relation is valid for any ρ > 2/(2α + 1):

∞  ρ
∑ λn (J f ) < ∞.
n=1

The investigation of the upper bounds or estimates of errors of approximation of

functions from a fixed class by some method of approximation began with an article
by Lebesgue (1910). In particular, Lebesgue proved that
 
Sn (Lip α )∞ := sup  f − Sn ( f )∞  n−α ln n.
f ∈Lip α

Here and later we write an  bn for two sequences an and bn if there are two positive
constants C1 and C2 such that C1 bn ≤ an ≤ C2 bn for all n.
The problem of approximation of functions in the classes W r H α by trigonomet-
ric polynomials was so natural that a tendency to find either asymptotic or exact
values of the following quantities appeared:
Sn (W r H α )∞ := sup  f − Sn ( f )∞ , En (W r H α )∞ := sup En ( f )∞ .
f ∈W r H α f ∈W r H α

We now formulate the first results in this direction. Kolmogorov (1936) proved the
r , see §1.4)
relation (in our notation W r = W∞,r
4 ln n
Sn (W r )∞ = + O(n−r ), n → ∞.
π 2 nr
Independently, Favard (1937) and Akhiezer and Krein (1937) proved the equality
En (W r )∞ = Kr (n + 1)−r ,
where Kr is a number depending on the natural number r.
In 1936 Kolmogorov introduced the concept of the width dn of a class F in a
space X:
 
 n 
 
dn (F, X) := infn sup infn  f − ∑ c j φ j  .
{φ j } j−1 f ∈F {c j } j−1  j=1

X
 1.1 Introduction 5

This concept allows us to find, for a fixed n and for a class F, a subspace of
dimension n that is optimal with respect to the construction of a best approxi-
mating element. In other words, the concept of width allows us to choose from
among various Chebyshev methods having the same quantitative characteristic of
complexity (the dimension of the approximating subspace) the one which has the
greatest accuracy.
The first result about widths (Kolmogorov, 1936), namely
d2n+1 (W2r , L2 ) = (n + 1)−r ,
showed that the best subspace of dimension 2n + 1 for the approximation of classes
of periodic functions is the subspace of trigonometric polynomials of order n. This
result confirmed that the approximation of functions in the class W2r by trigonomet-
ric polynomials is natural. Further estimates of the widths d2n+1 (Wq,r α , L p ), 1 ≤ q,
p ≤ ∞, some of which are discussed in §2.1 below, showed that, for some values
of the parameters q, p, the subspace of trigonometric polynomials of order n is
optimal (in the sense of the order of decay) but for other values of q, p this subspace
is not optimal.
The Ismagilov (1974) estimate for the quantity dn (W1r , L∞ ) gave the first exam-
ple, where the subspace of trigonometric polynomials of order n is not optimal.
This phenomenon was thoroughly studied by Kashin (1977).
In analogy to the problem of the Kolmogorov width, that is, to the problem con-
cerning the best Chebyshev method, problems concerning the best linear method
and the best Fourier method were considered.
Tikhomirov (1960b) introduced the linear width:
λn (F, L p ) := inf sup  f − A f  p ,
A:rank A≤n f ∈F

and Temlyakov (1982a) introduced the orthowidth (Fourier width):

 
 n 
 
ϕn (F, L p ) := inf sup 
orthonormal system {ui }ni=1 f ∈F 
f − ∑ i i  .
 f , u u
i=1 p

A discussion and comparison of results concerning λn (Wqr , L p ) and

dn (Wqr , L p ),
ϕn (Wq , L p ) can be found in §2.1. Here we remark that, from the point of view of the
r

orthowidth, the Fourier operator Sn is optimal (in the sense of order) for all 1 ≤ q,
p ≤ ∞ with the exception of the two cases (1, 1) and (∞, ∞).
Keeping in mind the primary question about the selection of an optimal sub-
space of approximating functions, we now draw some conclusions from this brief
historical survey.
(1) The trigonometric polynomials have been considered as a natural means of
approximation of periodic functions during the whole period of development
of approximation theory.
6 Approximation of Univariate Functions

(2) In approximation theory (as well as in other fields of mathematics) it has turned
out that it is natural to unite functions with the same smoothness into a class.
(3) The subspace of trigonometric polynomials has been obtained in many cases as
the solution of problems regarding the most precise method for the classes of
smooth functions: the Chebyshev method (which uses the Kolmogorov width),
the linear method (which uses the linear width), or the Fourier method (which
uses the orthowidth).
On the basis of these remarks we may formulate the following general strategy
for investigating approximation problems; we remark that this strategy turns out to
be most fruitful in those cases where we do not know a priori a natural method
of approximation. First, we solve the width problem for a class of interest in the
simplest case, that of approximation in Hilbert space, L2 . Second, we study the
system of functions obtained and apply it to approximation in other spaces L p .
This strategy will be used in Chapters 3, 4, and 5.

1.2 Trigonometric Polynomials

Functions of the form
n
t(x) = ∑ ck eikx = a0 /2 + ∑ (ak cos kx + bk sin kx)
|k|≤n k=1

(ck , ak , bk are complex numbers) will be called trigonometric polynomials of order

n. We denote the set of such polynomials by T (n) and the subset of T (n) of real
polynomials by RT (n).
We first consider a number of concrete polynomials that play an important role
in approximation theory.

1.2.1 The Dirichlet Kernel of Order n

The classical univariate Dirichlet kernel of order n is defined as follows:
Dn (x) := ∑ eikx = e−inx (ei(2n+1)x − 1)(eix − 1)−1
|k|≤n
sin(n + 1/2)x
= . (1.2.1)
sin(x/2)
The Dirichlet kernel is an even trigonometric polynomial with the majorant
   
Dn (x) ≤ min 2n + 1, π /|x| , |x| ≤ π . (1.2.2)
The estimate
Dn 1 ≤ C ln n, n = 2, 3, . . . , (1.2.3)
follows from (1.2.2).
 1.2 Trigonometric Polynomials 7

We mention the well-known relation (see Dzyadyk, 1977, p. 112)

4
Dn 1 = ln n + Rn , |Rn | ≤ 3, n = 1, 2, 3, . . .
π2
For any trigonometric polynomial t ∈ T (n) we have

Dn ∗ t := (2π )−1 Dn (x − y)t(y)dy = t.
T

Denote
xl := 2π l/(2n + 1), l = 0, 1, . . . , 2n.

Clearly, the points xl , l = 1, . . . , 2n, are zeros of the Dirichlet kernel Dn on [0, 2π ].
For any |k| ≤ n we have
2n 2n
∑ eikx Dn (x − xl ) = ∑ eimx ∑ ei(k−m)x = eikx (2n + 1).
l l

l=0 |m|≤n l=0

Consequently, for any t ∈ T (n),

2n
t(x) = (2n + 1)−1 ∑ t(xl )Dn (x − xl ). (1.2.4)
l=0

Further, it is easy to see that for any u, v ∈ T (n) we have

 π 2n
u, v := (2π ) −1
u(x)v(x)dx = (2n + 1)−1 ∑ u(xl )v(xl ) (1.2.5)
−π l=0

and, for any t ∈ T (n),

2n  2
t22 = (2n + 1)−1 ∑ t(xl ) . (1.2.6)
l=0

For 1 < q ≤ ∞ the estimate

Dn q ≤ C(q)n1−1/q (1.2.7)

follows from (1.2.2). Applying the Hölder inequality (see (A.1.1) in the Appendix)
for estimating Dn 22 we get

2n + 1 = Dn 22 ≤ Dn q Dn q . (1.2.8)

Relations (1.2.7) and (1.2.8) imply for 1 < q < ∞ the relation

Dn q  n1−1/q . (1.2.9)

Relation (1.2.9) for q = ∞ is obvious.

8 Approximation of Univariate Functions

We denote by Sn the operator taking a partial sum of order n. Then for f ∈ L1

we have
 π
Sn ( f ) := Dn ∗ f = (2π )−1 Dn (x − y) f (y)dy.
−π

Theorem 1.2.1 The operator Sn does not change polynomials from T (n) and for
p = 1 or ∞ we have
Sn  p→p ≤ C ln n, n = 2, 3, . . . ,
and for 1 < p < ∞ for all n we have
Sn  p→p ≤ C(p).
This theorem follows from (1.2.3) and the Marcinkiewicz multiplier theorem
(see Theorem A.3.6).
For t ∈ T (n),
n
t(x) = a0 /2 + ∑ (ak cos kx + bk sin kx),
k=1

we call the polynomial t˜ ∈ T (n), where

n
t˜(x) := ∑ (ak sin kx − bk cos kx)
k=1

the polynomial conjugate to t.

Corollary 1.2.2 For 1 < p < ∞ and all n we have
t˜ p ≤ C(p)t p .

Proof Let t ∈ T (n). It is not difficult to see that t˜ = t ∗ Dn , where

n
Dn (x) := 2 ∑ sin kx.
k=1

Clearly, it suffices to consider the case of odd n. Let this be the case and set m :=
(n + 1)/2, l := (n − 1)/2. Representing D̃n (x) in the form
−1
1 n
1  imx 
D̃n (x) =
i ∑ eikx − ∑ eikx =
i
e Dl (x) − e−imx Dl (x) ,
k=1 k=−n

we obtain the corollary.

A trigonometric conjugate operator maps a function f (x) to a function

∑(sign k) fˆ(k)eikx .
k
 1.2 Trigonometric Polynomials 9

The Marcinkiewicz multiplier theorem A.3.6 implies that this operator is bounded
as an operator from L p to L p for 1 < p < ∞. We denote by f˜ the conjugate function.

1.2.2 The Fejér Kernel of Order n − 1

The classical univariate Fejér kernel of order n − 1 is defined as follows:
n−1  
Kn−1 (x) := n−1 ∑ Dm (x) = ∑ 1 − |m|/n eimx
m=0 |m|≤n
 2
sin(nx/2)
=  2 .
n sin(x/2)
The Fejér kernel is an even nonnegative trigonometric polynomial in T (n − 1)
with majorant
   
Kn−1 (x) = Kn−1 (x) ≤ min n, π 2 /(nx2 ) , |x| ≤ π . (1.2.10)
From the obvious relations
Kn−1 1 = 1, Kn−1 ∞ = n
and the inequality, see (A.1.6),
1/q
 f q ≤  f 1  f 1−1/q
∞

we get in the same way as we obtained (1.2.9),

Cn1−1/q ≤ Kn−1 q ≤ n1−1/q , 1 ≤ q ≤ ∞. (1.2.11)

1.2.3 The de la Vallée Poussin Kernels

The classical univariate de la Vallée Poussin kernel with parameters m, n is defined
as follows:
n−1
Vm,n (x) := (n − m)−1 ∑ Dl (x), n > m.
l=m

It is convenient to represent this kernel in terms of Fejér kernels:

 
Vm,n (x) = (n − m)−1 nKn−1 (x) − mKm−1 (x)
  2 −1
= (cos mx − cos nx) 2(n − m) sin(x/2) .
The de la Vallée Poussin kernels Vm,n are even trigonometric polynomials of order
n − 1 with majorant
    
Vm,n (x) ≤ C min n, 1/|x|, 1/ (n − m)x2 ) , |x| ≤ π . (1.2.12)
10 Approximation of Univariate Functions

Relation (1.2.12) implies the estimate

 
Vm,n 1 ≤ C ln 1 + n/(n − m) .
We often use the de la Vallée Poussin kernel with n = 2m and denote it by
Vm (x) := Vm,2m (x), m ≥ 1, V0 (x) := 1.
Then for m ≥ 1 we have
Vm = 2K2m−1 − Km−1 ,
which, with the properties of Kn , implies
Vm 1 ≤ 3. (1.2.13)
In addition,
Vm ∞ ≤ 3m.
Consequently, in the same way as above, see (1.2.9) and (1.2.11), we get
Vm q  m1−1/q , 1 ≤ q ≤ ∞. (1.2.14)
Denote
x(l) := π l/2m, l = 1, . . . , 4m.
Then, analogously to (1.2.4), for each t ∈ T (m) we have
4m    
t(x) = (4m)−1 ∑ t x(l) Vm x − x(l) . (1.2.15)
l=1

The operator Vm defined on L1 by the formula

Vm ( f ) := f ∗ Vm
is called the de la Vallée Poussin operator.
The following theorem is a corollary of the definition of the kernels Vm and the
relation (1.2.13).
Theorem 1.2.3 The operator Vm does not change polynomials from T (m), and
for all 1 ≤ p ≤ ∞ we have
Vm  p→p ≤ 3, m = 1, 2, . . .
In addition, we formulate two properties of the de la Vallée Poussin kernels.
(1) Relation (1.2.12) with n = 2m implies the inequality
   
Vm (x) ≤ C min m, 1/(mx2 ) , |x| ≤ π .
It is easy to derive from this inequality the following property.
 1.2 Trigonometric Polynomials 11

(2) For h satisfying the condition C1 ≤ mh ≤ C2 we have

 
∑ Vm (x − lh) ≤ Cm.
0≤l≤2π /h

We remark that property (2) is valid for the Fejér kernel Km .

1.2.4 The Jackson Kernel

The classical univariate Jackson kernel with parameters n, a is defined as follows:
2a
−1 sin(nx/2)
Jna (x) := γa,n , a ∈ N,
sin(x/2)
where γa,n is selected in such a way that

Jna 1 = 1. (1.2.16)

Let us estimate γa,n from below. We have

 π 2a
sin(nx/2)
γa,n = (2π )−1 dx
−π sin(x/2)
 π /n 2a
−1 nx/π
≥π dx ≥ Cn2a−1 . (1.2.17)
0 x/2
The Jackson kernel is an even nonnegative trigonometric polynomial of order
a(n − 1). It follows from (1.2.17) that

Jna (x) ≤ C min(n, n1−2a x−2a ), |x| ≤ π . (1.2.18)

Relation (1.2.18) implies that for 0 ≤ r < 2a − 1,

 π
Jna (x)xr dx ≤ C(r)n−r . (1.2.19)
0

1.2.5 The Rudin–Shapiro Polynomials

We define recursively pairs of trigonometric polynomials Pj (x) and Q j (x) of order
2 j − 1:

P0 := Q0 := 1,
i2 j x j
Pj+1 (x) := Pj (x) + e Q j (x), Q j+1 (x) := Pj (x) − ei2 x Q j (x).
12 Approximation of Univariate Functions

Then at each point x we have

|Pj+1 |2 + |Q j+1 |2 = (Pj + ei2 x Q j )(P j + e−i2 x Q j )

j j

+ (Pj − ei2 x Q j )(P j − e−i2 x Q j )

j j


= 2 |Pj |2 + |Q j |2 ).

Therefore, for all x

   
Pj (x)2 + Q j (x)2 = 2 j+1 .

Thus, for example,

Pn ∞ ≤ 2(n+1)/2 . (1.2.20)

It is clear from the definition of the polynomials Pn that

2n −1
Pn (x) = ∑ εk eikx , εk = ±1, ε0 = 1.
k=0

Let N be a natural number and

m
N= ∑ 2n , j
n1 > n2 > · · · > nm ≥ 0,
j=1

its binary representation. We set

m n j−1
RN (x) := Pn1 (x) + ∑ Pn j (x)ei(2 1 +···+2
n )x
,
j=2

RN (x) := RN (x) + RN (−x) − 1.

Then RN (x) has the form

RN (x) = ∑ εk eikx , εk = ±1,

|k|<N

and for this polynomial the estimate

RN ∞ ≤ CN 1/2 (1.2.21)

holds.

1.2.6 A Modification of the Fejér Kernel

We consider the polynomials
 1/2 imx
Gn (x) := ∑ 1 − |m|/n e .
|m|<n
 1.2 Trigonometric Polynomials 13

These are even trigonometric polynomials of order n − 1 with the properties

 2π
(2π )−1 Gn (x − a)Gn (x − b)dx = Kn−1 (a − b), (1.2.22)
0

Gn 1 ≤ C. (1.2.23)

Relation (1.2.22) is obvious. It implies that the system of polynomials

Gn (x − 2π l/n), l = 1, . . . , n is an orthogonal system in T (n − 1).
Let us prove the relation (1.2.23). Denote
 1/2
φ (u) := 1 − |u| , |u| ≤ 1.

Then we have on [−1, 1]

φ (u) = ∑ al eiπ lx
l

and it is not hard to prove that

 −3/2
|al | ≤ C |l| + 1 . (1.2.24)

Further,

Gn (x) = ∑ φ (m/n)eimx
|m|≤n

= ∑ al ∑ eim(x+π l/n) = ∑ al Dn (x + π l/n). (1.2.25)

l |m|≤n l

Let us consider the function

gn,l (x) := Dn (x + π l/n) − (−1)l Dn (x).

Using the representation (1.2.1) one can obtain the estimate

 
gn,l 1 ≤ C ln |l| + 2 . (1.2.26)

Further, owing to the equality

φ (1) = ∑ al (−1)l = 0,
l

relation (1.2.25) can be rewritten in the form

Gn (x) = ∑ al gn,l (x).

l

Using relations (1.2.24) and (1.2.26) we then obtain relation (1.2.23).

14 Approximation of Univariate Functions

1.2.7 A Generalization of the Rudin–Shapiro Polynomials

The trigonometric polynomials considered above (see §§1.2.1–1.2.6) were con-
structively obtained: either they are given by a formula (§§1.2.1–1.2.4, 1.2.6) or
a method of construction is supplied (§1.2.5). In this subsection we formulate a
theorem that proves the existence of polynomials with given properties.

Theorem 1.2.4 Let ε > 0, and let a subspace Ψ ⊂ T (n) be such that dim Ψ ≥
ε (2n + 1). Then there exists a t ∈ Ψ such that

t∞ = 1,

and
t2 ≥ C(ε ) > 0.

An analogous statement is valid for the multivariable trigonometric polynomials,

will be proved in Chapter 3 (see Theorem 3.2.1).
We remark that the polynomial t from Theorem 1.2.4, by virtue of the inequality

t22 ≤ t1 t∞ ,

satisfies the condition

t1 ≥ C(ε )2 > 0. (1.2.27)

1.2.8 An Application of the Gaussian Sums

In this subsection we construct polynomials that we will use in studying linear
widths. This construction is based on properties of the Gaussian sums:
q
∑ ei2π l j /q ,
2
S(q, l) :=
j=1

where q is a natural number and l, q are coprime; that is, (l, q) = 1. We confine
ourselves to the case where q is an odd prime.

Theorem 1.2.5 Let q > 2 be a prime, l = 0 an integer, and k an integer. Then, for
q
∑ ei2π (l j +k j)/q ,
2
S(q, l, k) :=
j=1

the following equality is true:

 
S(q, l, k) = q1/2 .
 1.2 Trigonometric Polynomials 15

Proof We first consider the case k = 0. Note that the quantity S(q, l) does not
change if we sum over the complete system of remainders modulo q instead of the
segment [1, q]. Consequently, for any integer h,
q
∑ ei2π l( j+h) /q .
2
S(q, l) = (1.2.28)
j=1

Further,
  q q
S(q, l)2 = ∑ e−i2π lh
2 /q
∑ ei2π l j /q
2
.
h=1 j=1

Using (1.2.28), we see that this is equal to

q q q q
∑e −i2π lh2 /q
∑e i2π l( j+h)2 /q
∑ ∑ ei2π l( j +2 jh)/q .
2
= (1.2.29)
h=1 j=1 h=1 j=1

Taking into account that


q
q for j = q,
∑e i2π l2 jh/q
=
0 for j ∈ [1, q),
h=1

we get from (1.2.29),

 
S(q, l)2 = q. (1.2.30)

Now let k be nonzero. Since q is a prime different from 2, the numbers 2lb,
b = 1, . . . , q, run through a complete system of remainders modulo q. Consequently,
there is a b such that
2lb ≡ k (mod q).

Then
l j2 + k j ≡ l( j + b)2 − lb2 (mod q)

and, consequently,
   
S(q, l, k) = S(q, l) = q1/2 .

The theorem is proved.

Theorem 1.2.6 Let q be a prime and q = 2a + 1. For any n ∈ [1, a] there is a

trigonometric polynomial tn ∈T (a)  such that only n Fourier coefficients of tn are
nonzero and for all k we have t (k) ≤ 1 and in addition
 ˆ
 
tn (0) ≥ (n + 1)/2, t(2π l/q) ≤ Cq1/2 , l = 1, . . . , 2a.
16 Approximation of Univariate Functions

Proof The proof of this theorem can easily be derived from a deep number the-
oretical result due to Hardy and Littlewood about estimating incomplete Gaussian
sums: for any n ∈ [1, q]
 
 n 
 i2π l j2 /q 
∑ e  ≤ Cq1/2 , (l, q) = 1. (1.2.31)
 j=1 

Indeed, let k j denote the smallest nonnegative remainder of the number j2

modulo q, j = 1, . . . , n, and let

G := {k j − a, j = 1, . . . , n}.

We set
tn (x) := ∑ eikx .
k∈G

Then
   n 
     
tn (2π l/q) =  ∑ ei2π lk/q  =  ∑ ei2π l j2 /q ,
   
k∈G j=1

which by (1.2.31) implies the required estimates for tn (2π l/q). The bound
tn (0) = n ≥ (n + 1)/2 is obvious.

For the sake of completeness we will prove Theorem 1.2.6 using Theorem 1.2.5.
Instead of (1.2.31) we prove the inequality
 
   i2π l j2 /q 
 
∑ (1 − | j − a|/n + e  ≤ q1/2 , (l, q) = 1. (1.2.32)
 j 

Let l ∈ [1, q − 1]. Consider the trigonometric polynomial

q−1
∑ ei2π l j /q ei( j−a)x .
2
t(x) :=
j=0

Then at the points xk = 2π k/(2a + 1) = 2π k/q we have

 k   
t(x ) = S(q, l, k) = q1/2 , k = 0, . . . , 2a. (1.2.33)

We set
un (x) := t(x) ∗ Kn−1 (x).

Then by (1.2.5),
2a
un (x) = q−1 ∑ t(xk )Kn−1 (x − xk ),
k=0
 1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz Theorem 17

and, using (1.2.33) we find that

  2a
un (x) ≤ q−1/2 ∑ Kn−1 (x − xk ) = q1/2 . (1.2.34)
k=0

Further,
 
un (0) = ∑ 1 − | j − a|/n + ei2π l j /q ,
2

where (a)+ := max(a, 0). By (1.2.34) this implies (1.2.32).

Setting
 
tn (x) := ∑ 1 − | j − a|/n + ei(k j −a)x ,
j

where the k j are the same as in the beginning of the proof of this theorem, we get
   
     i2π lk /q     i2π l j2 /q 
tn (2π l/q) = ∑ 1 − | j − a|/n e j
 = ∑ 1 − | j − a|/n + e ,
 j +   j 

which by (1.2.32) implies the conclusion of Theorem 1.2.6, with 2n − 1 nonzero

Fourier coefficients instead of n.

1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz Theorem

The Bernstein–Nikol’skii inequalities connect the L p -norms of a derivative of some
polynomial with the Lq -norm, 1 ≤ q ≤ p ≤ ∞, of this polynomial. We obtain here
inequalities for a derivative that is slightly more general than the Weyl fractional
derivative. We first make some auxiliary considerations.
For a sequence {aν }∞ ν =0 we write

Δaν := aν − aν +1 ; Δ2 aν := Δ(Δaν ) = aν − 2aν +1 + aν +2 .

Theorem 1.3.1 We have
 
 π  n  n
 
(π )−1 a0 /2 + ∑ aν cos ν x dx ≤ ∑ (ν + 1)|Δ2 aν |.
−π  ν =1
 ν =0

Proof Applying twice the Abel transformation (see (A.1.18) in the Appendix)
with aν = 0 for ν > n, we obtain
n n
t(x) := a0 + ∑ aν 2 cos ν x = ∑ Dν (x)Δaν
ν =1 ν =0
n ν n
= ∑ ∑ Dμ (x) Δ2 aν = ∑ (ν + 1)Kν (x)Δ2 aν . (1.3.1)
ν =0 μ =0 ν =0
18 Approximation of Univariate Functions

From (1.3.1), using Kν 1 = 1 we find

n
t1 ≤ ∑ (ν + 1)|Δ2aν |,
ν =0

as required.

1.3.1 The Bernstein inequality

We first prove the Bernstein inequality. Let us consider the following special
trigonometric polynomials. Let s be a nonnegative integer. We define
A0 (x) := 1, A1 (x) := V1 (x) − 1, As (x) := V2s−1 (x) − V2s−2 (x), s ≥ 2,
where the Vm are the de la Vallée Poussin kernels (see §1.2.3). Then As ∈ T (2s )
and by (1.2.13),
As 1 ≤ 6. (1.3.2)
Let r ≥ 0 and α be real numbers. We consider the polynomials
n
Vnr (x, α ) := 1 + 2 ∑ kr cos(kx + απ /2)
k=1
2n−1  
+2 ∑ kr 1 − (k − n)/n cos(kx + απ /2).
k=n+1

Let us prove that, for all r > 0 and α ,

 r 
Vn (x, α ) ≤ C(r)nr , n = 1, 2, . . . (1.3.3)
1

Since for an arbitrary α

   
Vnr (x, α ) − 1 = Vnr (x, 0) − 1 cos(απ /2) + Vnr (x, 1) − 1 sin(απ /2),
it suffices to prove (1.3.3) for α = 0 and for α = 1. We first consider the case
α = 0. Let vk be the Fourier cosine coefficients of the function Vnr (x, 0). Then, by
Theorem 1.3.1,
 r  2n−1
Vn (x, 0) ≤ ∑ (k + 1)|Δ2 vk |. (1.3.4)
1
k=0

It is easy to see that, for 1 ≤ k ≤ n − 2,

|Δ2 vk | ≤ C(r)kr−2 . (1.3.5)
By the identity
Δ2 (ak bk ) = (Δ2 ak )bk + 2(Δak+1 )(Δbk ) + ak+2 (Δ2 bk )
 1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz Theorem 19

with ak = kr and bk = 1 − (k − n)/n, we see that the inequality (1.3.5) will be valid
for n ≤ k ≤ 2n − 3 too. For the remaining values of k = 0 we have
|Δ2 vk | ≤ |Δvk | + |Δvk+1 | ≤ C(r)nr−1 . (1.3.6)
From the inequality |Δ2 v0 | ≤ C(r) and relations (1.3.4)–(1.3.6) we get the rela-
tion (1.3.3) for r > 0 and α = 0.
Let α = 1 and let A˜s (x) denote the polynomial which is the trigonometric con-
jugate to As (x), which means that in the expression for As (x) the functions cos kx
are substituted by sin kx. We prove that
A˜s 1 ≤ C. (1.3.7)
Clearly, it suffices to consider s ≥ 3. It is not difficult to see that the equality
   s−1 s−3 
A˜s (x) = 2 Im As (x) ∗ 4K2s−1 −1 (x) − 3K2s−1 −2s−3 −1 (x) ei(2 +2 )x ,
holds. From this equality, by virtue of the Young inequality with p = q = a = 1
(see A.1.16)) and the properties of the functions Kn and As , we obtain (1.3.7).
Further, for n = 2m , we have
 
Vnr (x, 1) − 1 = V2nr (x, 0) − 1 ∗ Vn0 (x, 1)
m+1 m+1
=− ∑ V2nr (x, 0) ∗ A˜s (x) = − ∑ V2r (x, 0) ∗ A˜s (x).
s (1.3.8)
s=1 s=1

From (1.3.8) by means of the Young inequality and using (1.3.7) and relation
(1.3.3), which has been proved for α = 0, we get
 r m+1
Vn (x, α )1 ≤ C(r) ∑ 2rs ≤ C(r)nr . (1.3.9)
s=0

Now let 2m−1 ≤ n < 2m ; then

Vnr (x, 1) = V2rm+1 (x, 1) ∗ Vn (x),
which by (1.3.9) and the Young inequality gives the required estimate for all n.
Relation (1.3.3) is proved.
We define the operator Drα , r ≥ 0, α ∈ R, on the set of trigonometric polynomials
as follows. Let f ∈ T (n); then
Drα f := f (r) (x, α ) := f (x) ∗ Vnr (x, α ), (1.3.10)
and f (r) (x, α ) is called the (r, α ) derivative. It is clear that for f (x) such that
fˆ(0) = 0 we have for natural numbers r,
dr
Drr f = f.
dxr
20 Approximation of Univariate Functions

The operator Drα is defined in such a way that it has an inverse for each T (n).
This property distinguishes Drα from the differential operator and it will be conve-
nient for us. On the other hand it is clear that
dr f
= Drr f − fˆ(0).
dxr
Theorem 1.3.2 For any t ∈ T (n) we have, for r > 0, α ∈ R, 1 ≤ p ≤ ∞,
 (r) 
t (x, α ) ≤ C(r)nr t p , n = 1, 2, . . .
p

Proof By the definition (1.3.10),

t (r) (x, α ) = t(x) ∗ Vnr (x, α ).
Therefore, by the Young inequality (A.1.16) with p = q, a = 1 for all 1 ≤ p ≤ ∞
and r we have
 (r)   
t (x, α ) ≤ t p Vnr (x, α ) .
p 1

To conclude the proof we just use inequality (1.3.3).

Let us discuss the case r = 0, which is excluded from Theorem 1.3.2. In the case
where r = 0 and α is an even integer we have
 (0)  
t (x, α )| = t(x)

and, consequently,
 (0) 
t (x, α ) = t p , 1 ≤ p ≤ ∞. (1.3.11)
p

To investigate the general case it suffices to study the trigonometric conjugate

operator. Theorem 1.2.1 and its corollary show that for all α and 1 < p < ∞ the
inequality
 (0) 
t (x, α ) ≤ C(p)t p

holds.
It remains to consider the cases p = 1, ∞. It is sufficient to consider α = 1.
We have for t ∈ T (n),
t (0) (x, 1) = tˆ(0) − t˜(x) = tˆ(0) − t(x) ∗ D2n+1 (x).
Further,
2n+1
D̃2n+1 (x) = 2 ∑ sin kx = 2 Im Dn (x)ei(n+1)x ;
k=1

consequently,
D2n+1 1 ≤ C ln(n + 2).
 1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz Theorem 21

Thus, for t ∈ T (n),

 (0) 
t (x, 1) ≤ C ln(n + 2)t p , p = 1, ∞. (1.3.12)
p

The relation (1.3.11) with α = 0 and (1.3.12) imply for all α the inequality
 (0) 
t (x, α ) ≤ C ln(n + 2)t p , p = 1, ∞. (1.3.13)
p

Remark 1.3.3 We have the relation

 
sup t (0) (x, 1) p t p  ln(n + 2), p = 1, ∞.
t∈T (n)

The upper estimate follows from (1.3.12). Let us prove the lower estimate. We
first consider the case p = ∞. Let f (x) = (π − x)/2, 0 < x < 2π , be a 2π -periodic
function; then
∞
f (x) = ∑ (sin kx)/k.
k=1

Let m = [n/2]. Then

t(x) := f (x) ∗ Vm (x)

has the following properties: t ∈ T (n),

m
t∞ ≤ 3π /2, t (0) (0, 1) ≥ ∑ 1/k ≥ C ln(m + 2), (1.3.14)
k=1

which imply the required lower estimate in the case p = ∞.

Let p = 1 and m = [n/2]. Then the function Vm ∈ T (n) has the following
properties:

Vm 1 ≤ 3, (1.3.15)
 (0) 
Vm (x, 1) ≥ C ln(m + 2). (1.3.16)
1

Let us prove (1.3.16). For t we have from the above consideration for p = ∞,
 (0) 
σ = |Vm (x, 1),t| ≤ Vm (x, 1)1 t∞
(0)
(1.3.17)

and
m
σ≥ ∑ 1/k ≥ C ln(m + 2). (1.3.18)
k=1

From relations (1.3.14), (1.3.17) and (1.3.18) we obtain (1.3.16). Then (1.3.15) and
(1.3.16) give the required lower estimate for p = 1.
22 Approximation of Univariate Functions

1.3.2 The Nikol’skii Inequality

Let us now prove the Nikol’skii inequality.
Theorem 1.3.4 For any t ∈ T (n), n > 0, we have the inequality
t p ≤ Cn1/q−1/p tq , 1 ≤ q < p ≤ ∞.
Proof First let p = ∞; then
t = t ∗ Vn
and by the Hölder inequality (A.1.1) we have
t∞ ≤ tq Vn q ,
which, by (1.2.14), implies that
t∞ ≤ Ctq n1/q . (1.3.19)
Further, let q < p < ∞. Then by (A.1.6),
q/p
t p ≤ tq t1−q/p
∞ . (1.3.20)
The theorem follows from relations (1.3.19) and (1.3.20).
We now formulate a corollary of Theorems 1.3.2 and 1.3.4.
Corollary 1.3.5 (The Bernstein–Nikol’skii inequality) For t ∈ T (n) and arbi-
trary r > 0, α , 1 ≤ q ≤ p ≤ ∞, we have the inequality
 (r) 
t (x, α ) ≤ C(r)nr+1/q−1/p tq , n = 1, 2, . . .
p

1.3.3 The Marcinkiewicz Theorem

The set T (n) of trigonometric polynomials is a space of dimension 2n
 + 1.Each
polynomial t ∈ T (n) is uniquely defined by its Fourier coefficients tˆ(k) |k|≤n ,
and by the Parseval identity we have
 2
t22 = ∑ tˆ(k) , (1.3.21)
|k|≤n

which means that the set T (n) as a subspace of L2 is isomorphic to 2n+1

2 . Relation
(1.2.6) shows that a similar isomorphism can be
 set up in another way: by mapping
2n
a polynomial t ∈ T (n) to the vector m(t) := t(xl ) l=0 of its values at the points
xl := 2π l/(2n + 1), l = 0, . . . , 2n.
Relation (1.2.6) gives
 
t2 = (2n + 1)−1/2 m(t)2 .
The following statement is the Marcinkiewicz theorem.
 1.3 The Bernstein–Nikol’skii Inequalities. The Marcienkiewicz Theorem 23

Theorem 1.3.6 Let 1 < p < ∞; then for t ∈ T (n), n > 0, we have the relation
 
C1 (p)t p ≤ n−1/p m(t) p ≤ C2 (p)t p .

Proof We first prove a lemma.

Lemma 1.3.7 Let 1 ≤ p ≤ ∞; then, for n > 0,
 
 2n 
 
 ∑ al Vn (x − xl ) ≤ Cn1−1/p a2n+1 , a := (a0 , . . . , a2n ).
l=0  p
p

Proof Let V be an operator on 2n+1

p defined as follows:
2n
V (a) := ∑ al Vn (x − xl ).
l=0

It is obvious that (see (1.2.13))

V 2n+1 →L1 ≤ 3. (1.3.22)

1

Using the estimate (see (1.2.12))

  
Vn (x) ≤ C min n, (nx2 )−1 )

it is not hard to prove that

V ∞2n+1 →L∞ ≤ Cn. (1.3.23)

From relations (1.3.22) and (1.3.23), using the Riesz–Torin theorem (see Theorem
A.3.2) we find that
V 2n+1
p →L p ≤ Cn
1−1/p
,

which implies the lemma.

We now continue the proof of Theorem 1.3.6. Let Sn be the operator that takes
the partial Fourier sum of order n. Using Theorem 1.2.1 we derive from Lemma
1.3.7 the upper estimate (the first inequality in Theorem 1.3.6):
2n
t(x) = (2n + 1)−1 ∑ t(xl )Dn (x − xl )
l=0
2n
= Sn (2n + 1)−1 ∑ t(xl )Vn (x − xl ) .
l=0

Consequently,
 
t p ≤ C(p)n−1/p m(t) p .
24 Approximation of Univariate Functions

We now prove the lower estimate (the second inequality in Theorem 1.3.6) for
1 ≤ p < ∞. We have
  2n   2n  
m(t) p = ∑ t(xl ) p = ∑ t(xl )εl t(xl ) p−1
p
l=0 l=0
 2π 2n   p−1
= (2π )−1 t(x) ∑ εl t(xl ) Vn (x − xl )dx
0 l=0
 
 2n   p−1 
 
≤ t p  ∑ εl t(x ) Vn (x − x ) ,
l l
l=0 
p

using Lemma 1.3.7 we see that the last expression is

  p−1
≤ Ct p n1/p m(t) , p

which implies the required lower estimate and the theorem is proved.

Remark 1.3.8 In the proof of Theorem 1.3.6 we also proved the inequality

m(t)1 ≤ Cnt1 .

We now prove a statement that is analogous to Theorem 1.3.6 but, in contrast to

it, includes the cases p = 1 and p = ∞. Instead of the vector m(t) we now consider
the vector
   
M(t) := t x(1) , . . . ,t x(4n) , x(l) := π l/(2n), l = 1, . . . , 4n.

Theorem 1.3.9 For an arbitrary t ∈ T (n), n > 0, 1 ≤ p ≤ ∞, we have

 
C1 t p ≤ n−1/p M(t) p ≤ C2 t p .

Proof In the same way as for Lemma 1.3.7 one can prove:

Lemma 1.3.10 Let 1 ≤ p ≤ ∞, then, for n > 0,

 
 4n  
 
 ∑ al Vn x − x(l)  ≤ Cn1−1/p a4n .
l=1  p
p

Lemma 1.3.10 with a = M(t) and relation (1.2.15) implies the upper estimate
 
t p ≤ Cn−1/p M(t) p .

The corresponding lower estimate for 1 ≤ p < ∞ can be proved in the same way as
above for m(t), substituting xl by x(l).
The lower estimate for p = ∞ is obvious.
 1.4 Approximation of Functions in the Classes Wq,r α and Hqr 25

1.4 Approximation of Functions in the Classes Wq,r α and Hqr

1.4.1 Some Properties of the Bernoulli Kernels
For r > 0 and α ∈ R the functions
∞
Fr (x, α ) = 1 + 2 ∑ k−r cos(kx − απ /2)
k=1

are called Bernoulli kernels.

We define the following operator in the space L1 ,
 2π
(Iαr φ )(x) := (2π )−1 Fr (x − y, α )φ (y)dy. (1.4.1)
0

Let us prove that the definition of this operator is reasonable. To establish this it
suffices to prove that Fr ∈ L1 .
Theorem 1.4.1 For r > 0, α ∈ R we have

Fr ∈ L1 , En (Fr )1 ≤ C(r)(n + 1)−r , n = 0, 1, . . .

Proof Let us consider the functions

2s
f rs (x, α ) := As (x) ∗ 1 + 2 ∑ k−r cos(kx − απ /2) ,
k=1

where the As are defined in §1.3.

We first consider the case α = 0. Using Theorem 1.3.1 in the same way as in the
proof of inequality (1.3.3) we get
 r 
 f s (x, 0) ≤ C(r)2−rs . (1.4.2)
1

Further,
fsr (x, α ) = Dr−α fs2r (x, 0),

and, consequently, from (1.4.2) and Theorem 1.3.2 we find that

 r 
 f s (x, α ) ≤ C(r)2−rs . (1.4.3)
1

Thus the series

∞
∑ f rs (x, α )
s=0

converges in L1 to some function f (x) and

 ∞ 
 
 ∑ f rs (x, α ) ≤ C(r)2−rm . (1.4.4)
 
s=m 1
26 Approximation of Univariate Functions

From the definition of the function f rs (x, α ) we get

n
Sn ( f ) = 1 + 2 ∑ k−r cos(kx − απ /2)
k=1
and
  ∞   
 f − Sn ( f ) ≤ ∑  f rs (x, α ) − Sn f rs (x, α ) 1
1
s=0
 r 
≤ ∑  f s − Sn ( f rs ) ≤ C ln(n + 2) ∑  f rs 1
1
s:2s >n 2s >n
−r
≤ C(r)n ln(n + 2). (1.4.5)
Here we have used Theorem 1.2.1 and relation (1.4.3). Relation (1.4.5) shows that
the series defining the function Fr (x, α ) converges in L1 to f (x). The first part
of the theorem is proved. The second part of the theorem follows from relation
(1.4.4).
We now proceed to formulate some properties of the operators Drα and Iαr . From
the equality (φ ∈ L1 )
 2π  2π    
−1
π φ (u) cos k(y − u) + απ /2 cos k(x − y) + β π /2 dy du
0 0
 2π 
= φ (u) cos k(x − u) + (α + β )π /2)du,
0
which is valid for any nonzero k, the equalities
Drα11 Drα22 = Drα11+r2
+ α2 , (1.4.6)
Iαr11 Iαr22 = Iαr11+r2
+ α2 , (1.4.7)
Drα Iαr = Iαr Drα = I (1.4.8)
follow (we assume that the operators act on a set of trigonometric polynomials).
Denote by Wq,r α B, r > 0, α ∈ R, 1 ≤ q ≤ ∞, the class of functions f (x) repre-
sentable in the form
f = Iαr φ , φ q ≤ B. (1.4.9)
For such functions, with some q and B. we define (see (1.4.8))
Drα f = φ .
Let 1 < q < p < ∞, β := 1/q−1/p. From Corollary A.3.8 of the Hardy–Littlewood
inequality (see the Appendix) and the boundedness of the trigonometric conjugate
operator as an operator from L p to L p for 1 < p < ∞ (see Corollary 1.2.2), it follows
that
β
Iα q→p ≤ C(q, p). (1.4.10)
 1.4 Approximation of Functions in the Classes Wq,r α and Hqr 27

Relations (1.4.7) and (1.4.10) imply the following embedding theorem.

Theorem 1.4.2 Let 1 < q < p < ∞, β = 1/q − 1/p, r > β ; then
r−β
Wq,r α1 ⊂ Wp,α2 B, α1 , α2 ∈ R.

1.4.2 Approximation for Smoothness Classes

Let us define the classes Hqr B, r > 0, 1 ≤ q ≤ ∞ as follows:
 
 a 
Hq B := f ∈ Lq :  f q ≤ B, Δt f (x)q ≤ B|t| , a = [r] + 1 ,
r r

Δt f (x) := f (x) − f (x + t), Δta := (Δt )a .

For the case B = 1 we simply write Hqr := Hqr 1, i.e., we drop the constant B.
Let us study these classes from the point of view of their approximation by
trigonometric polynomials.

Theorem 1.4.3 Let r > 0, 1 ≤ q ≤ ∞, then

En (Hqr )q  (n + 1)−r , n = 0, 1, . . .

Proof Let us prove the upper estimate. Clearly, it suffices to consider the case
n > 0. Let f ∈ Hqr . We consider (see §1.2.4)
 π 
t(x) := (2π )−1 f (x) − Δay f (x) Jna (y)dy.
−π

Then t ∈ T (an) and

 π
f (x) − t(x) = (2π )−1 Δay f (x)Jna (y)dy.
−π

By a generalization of the Minkowskii inequality, (A.1.9), we have

 π 
 f − tq ≤ (2π )−1 Δay f (x) Jna (y)dy,
q
−π

which by the definition of the class Hqr and relation (1.2.19) implies that

 f − tq ≤ C(r)n−r .

The upper estimate is proved.

We now prove the lower estimate. We construct functions which will be used in
the proof of the more general Theorem 1.4.9. Let n > 0 be given and s be such that

4n ≤ 2s ≤ 8n.
28 Approximation of Univariate Functions

We consider
f (x) := 2−(r+1−1/q)s As (x) (1.4.11)
and remark that to prove the theorem it suffices to consider the simpler function
f (x) = (n + 1)−r ei(n+1)x . Then, for any t ∈ T (n), we have on the one hand
 f − t, As  =  f , As  = 2−(r+1−1/q)s As 22 ≥ C2−(r−1/q)s . (1.4.12)
On the other hand using the definition of As and (1.2.14) we get
 f − t, As  ≤  f − tq As q ≤ C2s/q  f − tq . (1.4.13)
From relations (1.4.12) and (1.4.13) we obtain
En ( f )q ≥ C2−rs ≥ Cn−r .
To show that f ∈ Hqr B, we prove the following auxiliary statement.
Lemma 1.4.4 Let g(x) be an a-times continuously differentiable 2π -periodic
function. Then for all 1 ≤ q ≤ ∞ we have
 a   
Δy g(x) ≤ | y|a g(a) (x) .
q q

Proof Clearly it suffices to consider the case a = 1. We have

 x+y   y 
     
Δy g(x) =  g (u)du =  g (x + u)du
 
q  x 0
 ≤ |y|g q ,
q q

as required.
From (1.4.11), (1.2.14), and the Bernstein inequality (Theorem 1.3.2) we get
 f (a) q ≤ C(a)2(a−r)s . (1.4.14)
Using Lemma 1.4.4 and the simple inequality
 a 
Δy f (x) ≤ 2a  f q ,
q

we obtain
 a   
Δy f (x) ≤ C(a) min |y|a na−r , n−r , (1.4.15)
q

which implies that f ∈ Hqr B with some B that is independent of n, and this proves
the lower estimate.
Let us now prove a representation theorem for the class Hqr B. Let
As ( f ) := As ∗ f
and denote the value of As ( f ) at a point x by As ( f , x).
 1.4 Approximation of Functions in the Classes Wq,r α and Hqr 29

Theorem 1.4.5 Let f ∈ Lq , 1 ≤ q ≤ ∞,  f q ≤ 1. For Δta f q ≤ |t|r , a = [r] + 1

it is necessary and sufficient that the following conditions be satisfied:
 
As ( f ) ≤ C(r, q)2−rs , s = 0, 1, . . . .
q

(The constants C(r, q) may be different for the cases of necessity and sufficiency.)
Proof

Necessity. Let f ∈ Hqr ; then for any ts ∈ T (2s−2 ), s ≥ 2 we have

A s ( f ) = A s ( f − ts )
and

As ( f )q ≤ As 1  f − ts q .

Applying Theorem 1.4.3 and using relation (1.3.2) we get

 
As ( f ) ≤ C(r, q)2−rs .
q

Sufficiency. Let
 
As ( f ) ≤ γ 2−rs , (1.4.16)
q

then using Corollary 2.2.7 we get

∞
f= ∑ As ( f ),
s=0

in the sense of convergence in Lq , and

∞  
Δta f q ≤ ∑ Δta As ( f )q . (1.4.17)
s=1

From Lemma 1.4.4 we find, in the same way as in (1.4.15),

 a     
Δt As ( f ) ≤ C(a)2−rs min 1, |t|2s a . (1.4.18)
q

From (1.4.17) and (1.4.18) we obtain

Δta f q ≤ C(r)γ |t|r ,
which concludes the proof of the theorem if we take γ < 1/C(r).
Denote
δ0 ( f ) := S0 ( f ), δs ( f ) := S2s −1 ( f ) − S2s−1 −1 ( f ), s = 1, 2, . . .
Corollary 1.4.6 In the case 1 < q < ∞ the functions As ( f ) in Theorem 1.4.5 can
be replaced by δs ( f ).
30 Approximation of Univariate Functions

Proof For 1 < q < ∞ the conditions

 
(1) As ( f )q ≤ C(q)2−rs ,
 
(2) δs ( f ) ≤ C(q)2−rs
q

are equivalent for all s. Indeed,

 
As ( f ) = As ∗ δs−1 ( f ) + δs ( f ) ,
 
δs ( f ) = δs As ( f ) + As+1 ( f ) ,
which by (1.3.2) and the boundedness of the operator δs as an operator from Lq
to Lq , 1 < q < ∞ (see Corollary A.3.4) implies the equivalence of conditions (1)
and (2).
Corollary 1.4.7 Let 1 ≤ q ≤ ∞,  f q ≤ 1 and
En ( f )q  (n + 1)−r , n = 0, 1, . . . ;
then f ∈ Hqr B for some B.
Indeed, in the same way as in the proof of the necessity in Theorem 1.4.5 we get

As ( f )q  2−rs ,

which by Theorem 1.4.5 (regarding the sufficiency) implies that f ∈ Hqr B.

Statements of the type of Theorem 1.4.3 are called direct theorems of approx-
imation theory, and statements of the type of Corollary 1.4.7 are called inverse
theorems of approximation theory.
Theorem 1.4.1 and Corollary 1.4.7 imply that
Fr (x, α ) ∈ H1r B. (1.4.19)
Consequently, for f ∈ Wq,r α we have
 a   
Δt f (x) ≤ Δta Fr (x, α ) Drα f q ≤ B|t|r ;
q 1

that is, f ∈ Hqr B.

Thus, we have proved that
Wq,r α ⊂ Hqr B. (1.4.20)

Let us prove an embedding theorem for the H classes.

Theorem 1.4.8 Let 1 ≤ q ≤ p ≤ ∞, β := 1/q − 1/p, r > β . We have the inclusion
Hqr ⊂ Hpr−β B
r−β
(in the case p = ∞ this means that for any f ∈ Hqr there is an equivalent g ∈ H∞ B).
 1.4 Approximation of Functions in the Classes Wq,r α and Hqr 31

Proof Let f ∈ Hqr . By Theorem 1.4.5

 
As ( f ) ≤ C(r, q)2−rs .
q

Therefore, by the Nikol’skii inequality (Theorem 1.3.4) we have

 
As ( f ) ≤ C(r, q)2−(r−β )s . (1.4.21)
p

Let g(x) denote the sum of the series ∑∞s=0 As ( f , x) in the sense of convergence in
L p . From Corollary 2.2.7 below it follows that f and g are equivalent. From (1.4.21)
r−β
and the equality As ( f ) = As (g), by Theorem 1.4.5 we obtain g ∈ Hp B.
The theorem is proved.
With the aid of Theorem 1.4.8 we can prove the following statement.
Theorem 1.4.9 Let 1 ≤ q, p ≤ ∞, r > (1/q − 1/p)+ . Then
En (Wq,r α ) p  En (Hqr ) p  n−r+(1/q−1/p)+ .
Proof By relation (1.4.20) it suffices to prove the upper estimate for the H classes
and the lower estimate for the W classes. We first prove the upper estimate. Let
1 ≤ q ≤ p ≤ ∞. Then Theorems 1.4.8 and 1.4.3 give
En (Hqr ) p  n−r+1/q−1/p . (1.4.22)
For 1 ≤ p < q ≤ ∞ we have, by the monotonicity of the L p -norms and Theorem
1.4.3,
En (Hqr ) p ≤ En (Hqr )q  n−r .
From this and relation (1.4.22) the required upper estimates follow.
Let us prove the lower estimate. Let n and s be the same as in the proof of the
lower estimate in Theorem 1.4.3 and let f be defined by (1.4.11). Then by the
Bernstein inequality,
Drα f q ≤ C(r),
and f ∈ Wq,r α C(r).
Let 1 ≤ q ≤ p ≤ ∞. From relation (1.4.12) and relation (1.4.13) with p instead
of q we get
En ( f ) p ≥ Cn−r+1/q−1/p . (1.4.23)
For 1 ≤ p ≤ q ≤ ∞ it suffices to consider as an example
f (x) = 2(n + 1)−r cos(n + 1)x.
r and, for any t ∈ T (n),
Then f ∈ W∞, α

σ =  f (x) − t(x), cos(n + 1)x = (n + 1)−r , σ ≤  f − t1 ,

32 Approximation of Univariate Functions

which implies the estimate

−r
α )1 ≥ (n + 1) .
r
En (W∞, (1.4.24)
The required lower estimates follow from (1.4.23) and (1.4.24) and the theorem is
proved.
Remark 1.4.10 Theorem 1.2.3 implies that for any f ∈ L p the de la Vallée
Poussin inequality holds:
 
 f −Vn ( f ) ≤ 4En ( f ) p , 1 ≤ p ≤ ∞. (1.4.25)
p

This inequality and Theorem 1.4.9 show that, for all 1 ≤ q, p ≤ ∞,

 
Vn (Hqr ) p := sup  f −Vn ( f ) p  E2n (Hqr ) p , (1.4.26)
f ∈Hqr

and an analogous relation is valid for the W classes.

Thus, for the classes Wq,r α and Hqr there exist linear methods giving an approxi-
mation of the same order as the best approximation.
Remark 1.4.11 From Theorem 1.2.1 it follows that for all 1 < p < ∞ and f ∈ L p ,
 
 f − Sn ( f ) ≤ C(p)En ( f ) p . (1.4.27)
p

Consequently, if we are interested only in the dependence of the approximation

of a function f ∈ L p on n then it suffices, in the case 1 < p < ∞, to consider the
simplest method of approximation, namely, the Fourier method.
This remains true for the classes Wq,r α and Hqr for all 1 ≤ q, p ≤ ∞, excepting the
cases q = p = 1 and q = p = ∞. For the function class F let us denote
 
Sn (F) p := sup f − Sn ( f ) p .
f ∈F

Theorem 1.4.12 Let 1 ≤ q, p ≤ ∞, (q, p) = (1, 1) or (∞, ∞), and r > (1/q −
1/p)+ . Then
Sn (Wq,r α ) p  Sn (Hqr ) p  n−r+(1/q−1/p)+ .
Proof In the case 1 < p < ∞ the theorem follows from Theorem 1.4.9 and relation
(1.4.27). It remains to consider the cases p = 1, q > 1 and 1 ≤ q < p = ∞. In the
case p = 1, q > 1 we have
Sn (Hqr )1 ≤ Sn (Hqr∗ )q∗  n−r ,
where q∗ = min(q, 2).
Now let 1 ≤ q < p = ∞. In the case 1 ≤ q < 2, by Theorem 1.4.8 we have
r−(1/q−1/2)
Hqr ⊂ H2 B,
Random documents with unrelated
content Scribd suggests to you:
The Project Gutenberg eBook of Shell-shock
and other neuropsychiatric problems
 This ebook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.

Title: Shell-shock and other neuropsychiatric problems

Presented in five hundred and eighty-nine case
histories from the war literature, 1914-1918

Author: Elmer Ernest Southard

Release date: May 19, 2016 [eBook #52105]

Language: English

Credits: E-text prepared by Brian Coe and the Online Distributed

Proofreading Team (http://www.pgdp.net) from page
images generously made available by Internet Archive
(https://archive.org)

*** START OF THE PROJECT GUTENBERG EBOOK SHELL-SHOCK

AND OTHER NEUROPSYCHIATRIC PROBLEMS ***
 E-text prepared by Brian Coe
and the Online Distributed Proofreading Team
(http://www.pgdp.net)
from page images generously made available by
Internet Archive
(https://archive.org)

Note: Images of the original pages are available through

Internet Archive. See
https://archive.org/details/shellshockother00sout

THE
CASE HISTORY SERIES

CASE HISTORIES IN MEDICINE

BY
Richard C. Cabot, M.D.
Third edition, revised and enlarged
 DISEASES OF CHILDREN
BY
John Lovett Morse, M.D.
Third edition, revised and enlarged
Presented in two hundred Case Histories

ONE HUNDRED SURGICAL PROBLEMS

BY
James G. Mumford, M.D.
Second Printing

CASE HISTORIES IN NEUROLOGY

BY
E. W. Taylor, M.D.
Second Printing

CASE HISTORIES IN OBSTETRICS

BY
Robert L. DeNormandie, M.D.
Second Edition
 DISEASES OF WOMEN
BY
Charles M. Green, M.D.
Second Edition
Presented in one hundred and seventy-three Case Histories

NEUROSYPHILIS
MODERN SYSTEMATIC DIAGNOSIS AND TREATMENT
Presented in one hundred and thirty-seven Case Histories
BY
E. E. Southard, M.D., Sc.D.
AND
H. C. Solomon, M.D.
Being Monograph Number Two of the Psychopathic Department of
the Boston State Hospital, Massachusetts. (Monograph Number One
was A Point Scale for Measuring Mental Ability by Robert M. Yerkes,
James W. Bridges and Rose S. Hardwick. Published by Warwick and
York. Baltimore 1915.)

SHELL SHOCK and other NEUROPSYCHIATRIC PROBLEMS

Printed in five hundred and eighty-nine Case Histories
BY
E. E. Southard, M.D., Sc.D.
Being Monograph Number Three of the Psychopathic Department of
the Boston State Hospital, Massachusetts
HORSLEY, 1857-1916
DEJERINE, 1849-1917
 VAN GEHUCHTEN, 1861-1914

IN MEMORIAM

SHELL-SHOCK
AND OTHER
NEUROPSYCHIATRY
PROBLEMS
 PRESENTED IN FIVE HUNDRED AND
EIGHTY-NINE
CASE HISTORIES

FROM THE
WAR LITERATURE, 1914-1918

BY
E. E. SOUTHARD, M.D., Sc.D.
Director (1917-1918), U. S. Army Neuropsychiatric Training School
(Boston Unit); Late
Major, Chemical Warfare Service, U. S. Army; Bullard Professor of
Neuropathology,
Harvard Medical School; Director, Massachusetts State Psychiatric
Institute (of the Massachusetts Commission on Mental Diseases);
Late President, American Medico-Psychological Association

WITH A BIBLIOGRAPHY BY
NORMAN FENTON, S.B., A.M.
Sergeant Medical Corps, U. S. Army (Assistant in Psychology to the
Medical Director,
Base Hospital 117 A. E. F.); late interne in Psychology, Psychopathic
Department,
Boston State Hospital; Assistant in Reconstruction, National
Committee for
Mental Hygiene
 AND AN INTRODUCTION BY
CHARLES K. MILLS, M.D., L.L.D.
Emeritus Professor of Neurology, University of Pennsylvania

BY VOTE OF THE TRUSTEES OF THE BOSTON STATE HOSPITAL

MONOGRAPH NUMBER THREE
OF THE
PSYCHOPATHIC DEPARTMENT

BOSTON
W. M. LEONARD, Publisher
1919

COPYRIGHT, 1919, BY
W. M. LEONARD

To
THE NATIONAL COMMITTEE FOR
MENTAL HYGIENE
AND
ITS WORK IN
WAR AND PEACE
 PREFACE
This compilation was begun in the preparedness atmosphere of
the U. S. Army Neuropsychiatric Training School at Boston, 1917-18.
This particular school had to adapt itself to the clinical material of
the Psychopathic Hospital. Although war cases early began to drift
into the wards (even including some overseas material), it was
thought well to supplement the ordinary “acute, curable, and
incipient” mental cases of the hospital wards and out-patient service
with representative cases from the literature.
As time wore on, this “preparedness” ideal gave place to the ideal
of a collection of cases to serve as a source-book for
reconstructionists dealing with neuroses and psychoses. Shortage of
medical staff and delays incidental to the influenza epidemic held the
book back still further, and, as meantime Brown and Williams had
served the immediate need with their Neuropsychiatry and the War,
it was determined to make the compilation the beginning of a case-
history book on the neuropsychiatry of the war, following in part the
traditions of various case-books in law and medicine.
With the conclusion of the armistice, there is by no means an end
of these problems. Peace-practice in neuropsychiatry is bound to
undergo great changes and improvements, if only from the influx
into the peace-community of many more trained neuropsychiatrists
than were ever before available. This is particularly true in the
American community by reason of the many good men specially
trained in camp and hospital neuropsychiatry, both at home and in
the A. E. F., through the enlightened policy of our army in
establishing special divisions of the Surgeon-General’s Office dealing
separately with those problems.
Though a book primarily for physicians, some of its material has
interest for line-officers, who may see how much “criming” is matter
for medical experts, by running through the boxed headings
(especially of Sections A and B) and reading the simulation cases. As
Chavigny remarks, “shooting madmen neither restrains crime nor
sets a good example.”
But parts of the book look ahead to Reconstruction. Surely
occupation-workers, vocationalists, war risk insurance experts, and
in fact all reconstructionists, medical and lay, must find much to their
advantage in the data of Section D (Treatment and Results). Had
time permitted, the whole old story of “Railway Spine”—Shell-shock’s
congener—might have been covered in a series of cases from last
century’s literature, together with others illustrating the effects of
suggestion and psychotherapy; but this must be a post-bellum task.
The compiler, who has personally dictated (and as a rule
redictated and twice condensed) all the cases from the originals (or
in a few instances, e.g., Russian, from translations), hopes he has
not added anything new to the accounts. The cases are drawn from
the literature of the belligerents, 1914-1917, English, French, Italian,
Russian, and—so far as available here—German and Austrian.
I would call the collection not so much a posey of other men’s
flowers as a handful of their seeds. For I have constantly not so
much transcribed men’s general conclusions as borrowed their
specific fine-print and footnotes. The lure of the 100 per cent has
been very strong in many authors; but the test of fine-print, viz., of
the actual case-protocols, saves us from premature conclusions, and
the plan of the book allows us to confront actualities with actualities.
One gets the impression of a dignified debate from the way in which
case-histories automatically confront each other, say in Section C
(Diagnosis).
Obligations to the books of Babinski and Froment, Eder, Hurst,
Mott (Lettsomian Lectures), Roussy and Lhermitte, Elliot Smith and
Pear, and others are obvious. Yealland’s book came too late for
sampling its miracles, though cases of his in the periodical literature
had already been incorporated in my selection.
 Some of the cases in Section A, i, had already been abstracted in
Neurosyphilis: Modern Systematic Diagnosis and Treatment
(Southard and Solomon, 1917).
What we actually have made is a case-history book in the newly
combined fields now collectively termed neuropsychiatry. The more
general the good general practitioner of medicine, the more of a
neuropsychiatrist! And this is no pious wish or counsel of perfection.
Neuropsychiatry, mental hygiene, psychotherapy and somatotherapy
—all these will flourish intra-bellum and post-bellum, in days of
destruction and in days of reconstruction. And who amongst us,
medical or lay, will not have to deal in reconstruction days with cases
like some here compiled? A minor blessing of the war will be the
incorporation of mental hygiene in general medical practice and in
auxiliary fields of applied sociology, e.g., medico-social work.
Subsidies aiding publication are due to the National Committee for
Mental Hygiene; the Permanent Charity Foundation (Boston Safe
Deposit and Trust Company); Mrs. Zoe D. Underhill of New York; Mr.
H. T. White of New York; and Dr. W. N. Bullard of Boston—to all of
these the various military recipients of the book will be under
obligations, as well as others who would otherwise have had to pay
the great majoration de prix due to war times.
Of those great dead contributors to neurology laid (in the
Epicrisis) at the feet of the neo-Attila, perhaps only Sir Victor was in
a narrow sense the Kaiser’s victim: still, but for the war, they might
all remain to us.
By the way, just as I found John Milton had said things that fitted
neurosyphilis, so also Dante is observed in the chosen mottoes to
have had inklings even of Shell-shock. To the Inferno it was natural
to turn for fitting mottoes (Carlyle’s renderings mainly used). The
pages might have been strewn with them. A glint of too great
optimism might seem to shine—in the pre-Epicrisis motto—from the
lance of Achilles with its “sad yet healing gift;” but out of Shell-shock
Man may get to know his own mind a little better, how under stress
and strain the mind lags, blocks, twists, shrinks, and even splits, but
on the whole is afterwards made good again.
E. E. Southard.
Washington,
November, 1918.
 INTRODUCTION
The duties of an introducer, whether of a platform speaker to an
audience, or of a writer to his anticipated readers, are not always
clearly defined. It has been sometimes said that the critic or
reviewer may meet with better success if he has not acquainted
himself too thoroughly with the contents of the book about which he
writes, as in that case he will have a larger opportunity to indulge his
imagination, but a critique thus produced may have the
disadvantage of possible shortcoming or unfairness. In the case of
this volume, however, I have felt it worth while to acquaint myself
with its contents, no light task when one is confronted with a
thousand pages.
The great war just closing has done much to enlighten us as to
the causes, nature, outcome, and treatment of injuries and diseases
to which its victims have been subjected. The object of this book is
to present both the data and the principles involved in certain
neuropsychiatry problems of the war. These are presented in a
wealth of detail through an extraordinary series of case records (589
in all) drawn from current medical literature, during the first three
years of the conflict. Case reporting is here seen at its best, and the
experiences recorded are largely allowed to speak for themselves,
although comments are not wanting and are often illuminating.
Many criticisms have been heard on the use of the term Shell-
shock as applied to some of the most important psychiatric and
neurological problems of the recent war; but that the designation
has meaning will be evident if Dr. Southard’s book is not simply
skimmed over by the reader, but is studied in its entirety. The
symptoms of a very large number, if not the majority, of the cases
recorded, had for their initiating influence the psychic and physical
horrors of life among exploding shells. As the author and those from
whom he has received his clinical supply not infrequently point out,
in many cases it would appear that purely psychic influences have
played the chief rôle, but in others physical injuries have not been
lacking. Much more than this is true: in many instances the soil was
prepared by previous defect, disease, or injury, or to use one of Dr.
Southard’s favorite expressions, “weak spots” were present before
martial causes became operative.
While the contributions to the medical and surgical history of the
war have been somewhat numerous in current medical journals and
in monographs, few comprehensive volumes have appeared. The
reasons for this are not far to seek. The conflict has been of such
magnitude, and the demands on the bodily and mental activity of
the medical profession have been so intense and continuous, that
time and opportunity for the careful and complete recording of
experiences have not been often available; but works are beginning
to appear in the languages of all the belligerent countries and these
will increase in number and value during the next lustrum and
decade, although it may be that some of the most important
contributions will come after a decade or more is past. The great
work before me is one that will leave its lasting impress, not only
upon military but on civil medicine, for the lessons to be drawn from
its pages are in large part as applicable to the one as to the other.
Looking backward to our Civil War, one is strongly impressed with
the fact that the present volume, one of the earliest works of its kind
to appear in book form, deals largely with psychiatry and functional
nervous diseases, whereas during and after the American conflict
the most important contributions to neurology related to organic
disease, especially as illustrated by the work of Weir Mitchell and his
collaborators on injuries of nerves. This is the more interesting when
it is remembered that Mitchell not very long after the close of the
Civil War became the most prominent exponent of functional
neurology, from the diagnostic and therapeutic sides. To him the
profession the world over has been indebted for the development of
new views as to the nature of neurasthenia and hysteria and new
methods for combating these disorders. In this fact is to be found
matter for thought. Those who handled best the neuropsychiatric
problems of the present war were in large part qualified not merely
by a knowledge of psychology and psychiatry, but far more by a
thorough training in organic neurology. The problems of psychiatry
can be grasped fully only by those who have a fundamental
knowledge of the anatomy, physiology, and diseases of the nervous
system.
Dr. Southard, preëminently a neuropathologist, is well grounded in
organic neurology, and shows at every turn his capabilities for
considering the neuroses, psychoses, and insanities from the
standpoint of the neurologist. Moreover, he clearly shows training
and insight into the problems of non-neurological internal medicine.
The ideal method of training a student for neuropsychiatric work—
if one had the opportunity of directing his course from the time of
his entry into medicine—would be to see to it, after a good
grounding in the fundamental sciences like anatomy, physiology, and
chemistry, that medicine and surgery in their broadest phases first
received school and hospital attention; that the fields of neurology,
pure and applied, were then fully explored; and that psychology and
psychiatry received late but thorough consideration. When after
America’s entrance into the world war the writer assisted in
preparing medical reserve officers for neuropsychiatric service, those
men did best both during their postgraduate work and in base
hospitals and in the field, who had built from the bottom after the
manner indicated.
At the outset of Dr. Southard’s book, for more than two hundred
and fifty pages, the author considers under ten subdivisions the
acquired diseases and constitutional defects which may predispose
the soldier to functional and reflex nervous disease. Neurosyphilis,
on which Dr. Southard and Dr. Solomon have already given us a
valuable treatise, the pharmacopsychoses, especially alcoholism, and
the somatopsychoses covering fevers like typhoid and paratyphoid,
are considered in numerous carefully chosen case reports. The
reader needs only to look closely into the case records of the first
quarter of the volume to get a knowledge of the affections chiefly
predisposing the soldier or civilian to functional and reflex nervous
diseases. To those familiar with the medical history of the war it is
well known that one of the reasons for the efficiency of the
American Expeditionary Force resided in the fact that the preliminary
examinations of the recruits received the fullest attention not only
from the points of view of acquired and inherited disease, but also
from those of special psychiatric and even psychological deficiencies.
Our country, however, had for its guidance the experience of nations
which were fighting for three years before we entered the arena and
in addition had a large surplus of material from which to cull out the
weaklings.
Among the predispositional affections considered—besides
syphilis, alcohol, and other drug habits, and the somatopsychoses—
are the feeble-mindednesses or hypophrenoses, the epilepsies, the
psychoses due to focal brain lesions, the presenile and senile
disorders, the schizophrenoses including dementia præcox and allied
affections, the cyclothymoses like manic depressive insanity, the
psychoneuroses, and the psychopathoses. The last two subjects
indicated, considered in special chapters, seem to some extent to be
receptacles for affections which cannot well be otherwise placed,—
hallucinoses, hysteria, neurasthenia, and psychasthenia,—and under
the psychopathoses, pathological lying, Bolshevism, delinquencies of
various sorts, homosexuality, suicide and self-mutilation, nosophobia,
and even claustrophobia with its exemplar who preferred exposure
to shell-fire to remaining in a tunnel.
Under the encephalopsychoses are found interesting illustrations
of focal lesions and the general effects of infection and toxemia.
Cases of brain abscess, of spinal focal lesions, and meningeal
hemorrhage are in evidence, aphasias, monoplegias, Jacksonian
spasm, and thalamic disease receiving consideration.
All neurologists know the difficulties in diagnosticating epilepsy in
the absence of opportunities to see attacks and to receive the
carefully analyzed statement of the observers of the patient. All this
and much more is well brought out in the chapter on the
epileptoses. Many epileptics found their way into the armies either
through the carelessness of examiners or by suppression of the facts
on the part of those who desired to serve.
The fact that an imbecile can shoot straight and face fire comes
out in one or two places, but this does not seem to prove that a
good rifleman is necessarily an all-round good soldier.
A book like Dr. Southard’s could be made of much use in teaching
students, especially postgraduates, by having them, when a
particular subject like epilepsy or schizophrenia, for instance, is
under discussion, use as collateral reading the case reports of this
work.
Dr. Southard’s book will prove useful to many workers—to the
medical officer whose duty it is to examine recruits for the service or
to pass upon and treat them while in service; almost equally to the
medical officer in time of peace; to authors of textbooks and
treatises and to contributors to neurological and psychiatric journals;
to lecturers and clinical demonstrators; to the examiner for the
juvenile courts; and to members of the psychopathic, psychiatric,
and neurological staffs of our hospitals.
One is not called upon in an introduction to review at length the
contents of the volume, but it may prove of value to the reader to
dip here and there into the pages of the work to which his attention
is being invited.
It will be remembered that fifty years ago and much later, down to
the time of Babinski’s active propaganda in favor of the theories of
suggestion, counter-suggestion, and persuasion in hysteria, various
affections of a vasomotor and thermic type were included in the list
of hysterical phenomena. These and some other phenomena
sometimes classed as hysterical, Babinski and those who accord with
him now find it necessary to sweep entirely from the domain of
hysteria, which being produced by suggestion and cured by counter-
suggestion or persuasion cannot include symptoms which are
beyond the control of the will and intellect of the patient.
According to the new or rather revived pronouncement, these
must be due either to definite organic lesion, or to a disorder of
reflex origin, connoting the occurrence of changes in the nervous
centers as long ago taught by Vulpian and Charcot. In the records of
cases and in the discussions thereon this differentiation receives
much consideration.
It is held that the paralysis in the reflex cases is more limited,
more persistent, and assumes special forms not observable in
hysteria. The attitudes in hysterical palsies conform more to the
natural positions of the limbs than do those observed in reflex
paralysis. Probably the presence of marked amyotrophies in the
reflex nervous disorders is the most convincing factor in separating
these from pithiatic affections. These atrophies correspond to the
arthritic muscular atrophies of Vulpian, Charcot, Gowers, and others,
and cannot for a moment be regarded as caused by suggestion or as
removable by counter-suggestion or persuasion. They are
influenced, discounting the effect of time and natural recuperation,
only by methods of treatment designed to improve the peripheral
and central nutrition of the patient. Pithiatic atrophies are slight and
probably always to be accounted for by disuse or the association of
some peripheral neural disorder with the hysteria. Affections of the
sudatory and pilatory systems are more definitely pronounced in
reflex cases than in those of a strictly hysterical character.
Some of the facts brought forward by Babinski and Froment to
demonstrate the differentiation of reflex paralyses from pithiatic
disorders of motion are challenged in the records of this volume by
others, as for instance, by Dejerine, Roussy, Marie, and Guillain.
Babinski tells us that in pithiatism, properly so designated, the
tendon reflexes are not affected. He believes that even in
pronounced anesthesia of the lower extremities the plantar reflexes
can always be elicited and are not abnormal in exhibition. Dejerine,
however, produces cases to illustrate the fact that in marked
hysterical anesthesia of the feet plantar responses cannot be
produced. I have personally studied cases which lend some strength
to either contention. In some of these I was not able to conclude
that either the use of the will or the presence of contractions in
extension was sufficient to exclude the normal responses.
Differences in muscle tonicity, in mechanical irritability of the
muscles, and the presence or absence of fibrotendinous contractions
are indications of a separation between the reflex and purely
functional cases, as apparently demonstrated in some of the case
records. True trophic disorders of the skin, hair, and bones observed
in the reflex cases are also said to have no place in the illustrations
of pithiatism.
The delver into the case histories of this volume will find
numerous instructive combinations of hystero-reflex and organo-
hysterical associations which are not to be enumerated in an
introduction. The great importance of what all recognize as
pathognomonic signs of organic disease—Babinski extensor toe
response, persistent foot clonus, reactions of degeneration, marked
atrophy, lost tendon jerks, etc.—is, of course, continuously in
evidence. Extraordinary associations of hysterical, organic, and reflex
disorders with other affections due to direct involvement of bone,
muscle, and vessels and with the secondary effects of cicatrization
and immobilization are brought out on many pages. In quitting this
branch of our subject it might be remarked that considerable
changes must be made in our textbook descriptions of nervous
diseases in the light of the contributions to the neurology of the
present war.
One is reminded in the details of some of the cases of the
discussions some decades since on the subject of spinal
traumatisms; of the work of Erichsen which resulted in giving his
name and that of “railway spine” to many of the cases now
commonly spoken of as traumatic hysteria and traumatic
neurasthenia; of the rejoinders of Page and his views regarding
spinal traumatisms; and of Oppenheim’s development of the
symptom complex of what he prefers to term the traumatic neurosis.
One who has taken part in much court work cannot but read these
case records with interest, for the neurology of the war as presented
in this volume and in numerous monographs which are now
appearing, throws much light upon many often mooted medicolegal
problems. I recall how many able and honest neurological observers
have changed their points of view since the early days of Erichsen’s
“railway spine,” a pathological suggestion which is said to have cost
the corporations of England an almost fabulous sum during a score
of years. I recall also that a certain Court of Appeals in one of our
states even felt itself called upon to promulgate an opinion intended
to exorcise entirely the plea for damages for alleged injuries if it
could be shown that these were due to fright. The data of this book
do not put weapons entirely into the hands of the attorney and the
expert for either the plaintiff or the defendant.
Some of the French writers on the neurology of the war, as
illustrated in the records collected by Dr. Southard, have brought to
our attention distinctions which they draw between états
commotionnels and états émotionnels—happy terms, and yet not
sufficient in their invention or in the explanations which accompany
them, fully to satisfy the requirements of the facts presented. These
writers seem to think of the commotional states as denoting some
real disease or condition of the brain, and yet one which is really
curable and reversible. They explicitly tell us, however, that these
commotions fall short of being lésionnel. After all, is this not
somewhat obscure? Is it not something of a return to the period of
“railway spine” when one of the comparisons sometimes made was
that the injury suffered by the nervous tissues produced in them a
state comparable to that of a magnet which had been subjected to a
severe blow? At any rate, in commotion thus discussed the nervous
structures are supposed to sustain some real injury of a
physiochemical character, whereas in the emotional states the
neurones are, as Southard puts it, affected somewhat after the
manner of normal emotional functioning, except perhaps that they
are called upon to deliver an excessive stream of impulses. The
latter would be classed among the psychopathic, the former among
the physiopathic affections, and yet the distinction between the two
is not always quite clear.
In not a few instances of Shell-shock—although these are not
numerous, so far as records have been obtained—actual structural
lesions have been recorded even in cases in which no direct external
injury of a material kind was experienced as a result of the explosion
of shells. In others the evidences of external injury were relatively
unimportant. Various lesions, in some cases recognizable even by
the naked eye, were present. Mott, for example, found not only
minute hemorrhages, but in one instance a bulbar extravasation of
moderate massiveness, the patient not showing external signs of
injury. Cases are also recorded of hematomyelia; others with
edematous or necrotic areas in the cord; and still others with lesions
of the ependyma or even with splitting of the spinal canal, reminding
one of the classical experiments of Duret on cerebral and
cerebrospinal traumatisms.
It has been argued that too much stress should not be laid on a
few cases of this sort—but are they as few as they seem to be? The
fact is that necropsical opportunities are not often afforded. May not
such scattered lesions often be present without resulting in death or
even in long continued disturbance? There is no essential reason
why minute hemorrhages into the brain and spinal cord, and
especially into their membranes, may not undergo rapid absorption
or even remain unchanged for some time without dire results.
One of the reported cases in which lung splitting occurred from
severe concussion without external injury is not without interest in
this connection, reminding one, as the commentator says, of those
cases of severe concussion in which the interior of a building is
injured while the exterior escapes. In the same connection also the
cited experiments of Mairet and Durante on rabbits are not without
instructiveness. As a result of explosives set off close to these
animals, pulmonary apoplexy, spinal cord and root hemorrhages,
and extravasations, perivascular and ependymal, and into the
cortical and bulbar gray were found. Russca obtained direct and
contrecoup brain lesions, etc., in a similar way.
Here and there throughout the book will be found references to
symptoms and syndromes which will have a particular interest for
the reader—soldier’s heart, trench foot, congealed hand, tics,
tremors, convulsions, sensory areas variously mapped, and forms of
local tetanus, the last being distinctly to be differentiated from
pithiatic contractures and those due to organic lesions of the
nervous system. Cases of an affection described by Souques as
camptocormia, from Greek words meaning to bend the trunk, were
shown to the Neurological Society of Paris in 1914 and later, the
main features of this affection being pronounced incurvation forward
of the trunk from the dorsolumbar region, with extreme abduction
and outward rotation of the lower limbs, pain in the back, and
difficult and tremulous walking. In some of these cases, organic
lesions of the trunkal tissues were present, but in addition psychic
elements played a not unimportant part, and the cases were
restored to health by a combination of physical measures with
psychotherapy, enforced by electrical applications.
The part of this book given over to the discussion of treatment will
doubtless to some prove the most interesting section. The
presentation of the subject of therapeutics is in some degree a
discussion also of diagnosis and prognosis; and so it happens in
various parts of the volume that the particular subject under
consideration is more or less a reaffirmation or anticipation of
remarks under other headings.
Similar results are brought about by various therapeutic
procedures. Nonne, Myers, and a few others bring hypnosis into the
foreground, although non-hypnotic suggestion plays a larger rôle by
far.
Miracle cures are wrought through many pages. Mutism, deafness
and blindness, palsies, contractures, and tics disappear at times as if
by magic under various forms of suggestion. Ether or chloroform
narcosis drives out the malady at the moment when it reveals its
true nature. Verbal suggestion has many adjuvants and collaborators
—electricity, sometimes severely administered, lumbar puncture,
injections of stovaine into the cerebrospinal fluid, injections of saline
solution, colored lights, vibrations, active mechanotherapy,
hydrotherapy, hot air baths and blasts, massage, etc. Painful and
punitive measures have their place—one is inclined to think a less
valuable place than is given them by some of the recorders. In some
instances the element of suggestion, while doubtless present, is
overshadowed by the material methods employed. Persuasion and
actual physical improvement are in these cases highly important.
Reëducation is not infrequently in evidence. The patient in one way
or another is taught how to do things which he had lost the way of
doing.
It is interesting to American neurologists to note how frequently in
the reports, especially of French observers, the “Weir Mitchell
treatment” was the method employed, including isolation, the faradic
current, massage, and Swedish movements, hydrotherapy, dietetic
measures, reëducative processes, and powerful suggestion variously
exhibited, especially through the mastery of the physician over the
patient. It is rather striking that few records of Freudian
psychoanalytic therapy are presented.
When all is said, however, counter-suggestion and persuasion, in
whatever guise made use of, were not always sufficient and this not
only in the clearly organic cases, but in those which are ranked
under the head of reflex nervous disorders. In these the long-
continued use of physical agencies was found necessary to
supplement the purely psychic procedures, these facts sometimes
giving rise in the Paris Society of Neurology and elsewhere to
animated discussion as to the real nature of the cases. The pithiatic
features of the case at times disappear, but leave behind much to be
explained and more to be accomplished. The cures wrought are not
always permanent and in some cases post-bellum experiences may
be required to prove the real value of the measures advocated. The
reader must study well the detailed records in order to arrive at just
conclusions; nevertheless, the tremendous efficacy of suggestion
and persuasion stands out in many of the recitals.
Perhaps the author may permit the introducer a little liberty of
comment. His non-English interpellations, especially Latin and
French, may be regarded by some as overdone or perhaps pedantic,
but are rather piquant, giving zest to the text. Diagnosis per
exclusionem in ordine is sonorous and has a scholarly flavor, but
does not prevent the reader who lives beyond the faubourgs of
Boston from understanding that the author is speaking of an ancient
and well-tried method of differential diagnosis. Passim may be more
impressive or thought-fixing than its English translation, but this to
the reader will simply prove a matter of individual opinion.
Psychopathia martialis is not only mouth-filling like Senegambia or
Mesopotamia, but really has a claim to appreciation through its
evident applicability. It is agreeable to note that the book seems
nowhere to indicate that psychopathia sexualis and psychopathia
martialis are convertible terms.
The bibliography of the volume challenges admiration because of
its magnitude and thoroughness and is largely to be credited, as the
author indicates, to the energy and efficiency of Sergeant Norman
Fenton, who did the work in connection with the Neuropsychiatric
Training School at Boston, resorting first-hand to the Boston Medical
Library and the Library of the New York Academy of Medicine. After
Sergeant Fenton joined the American Expeditionary Force, Dr.
Southard greatly increased the value of the bibliography by his
personal efforts.
This bibliography covers not only the 589 case histories of the
book, but it goes beyond this, especially in the presentation of
references for 1917, 1918, and even 1919. Owing to the time when
our country entered the war, American references are, in the main,
of later date than the case histories. They will be found none the
less of value to the student of neuropsychiatric problems.
The references in the bibliography number in all more than two
thousand, distributed so far as nationalities are concerned about as
given below, although some mistakes may have crept into this
enumeration for various reasons, like the publication of the same
articles in the journals of different countries. The list of references
includes French, 895; British (English and Colonial), 396; Italian, 77;
Russian, 100; American, 253; Spanish, 5; Dutch, 5; Scandinavian, 5;
and Austrian and German, 476. It will be seen, therefore, that the
bibliography covers in number nearly four times the collected case
studies, most of these records being from reports made during the
first three years of the war. The author has wisely made an effort to
bring the bibliographic work up to and partially including 1919.
The manner in which the French neurologists and alienists
continued their work during the strenuous days of the terrible
conflict is worthy of all praise. The labors of the Society of Neurology
of Paris never flagged, its contributions in current medical journals
having become familiar to neurologists who have followed closely
the trend of medical events during the war. Cases and subjects were
also frequently presented and discussed at the neurological centers
connected with the French and allied armies in France.
It may be almost invidious to specify names, the work done by
many was of so much interest and value. Dejerine in the early days
of the war, before his untimely sickness and death, contributed his
part. Marie from the beginning to the end of the conflict continued
to make the neurological world his debtor. The name of Babinski
stands out in striking relief. Other names frequently appearing
among the French contributors are those of Froment, Clovis Vincent,
Roussy and Lhermitte, Léri, Guillain, Souques, Laignel-Lavastine,
Courbon, Grasset, Claude, Barre, Benisty, Foix, Chavigny,
Charpentier, Meige, Thomas, and Sollier.
For a work of this character not only as complete a bibliography as
possible, but a thorough index is absolutely necessary, and this has
been supplied. The author has not made the index too full, but with
enough cross-references to enable those in all lines of medical work
interested to cull out the cases and comments which most concern
them.
 My prologue finished, I step aside for the play and the player, with
the recommendation to the reader that he give close heed to the
performance—to the recital of the cases, the comments thereon, and
the general discussion of subjects—knowing that such attention will
be fully rewarded, for in this wonderful collection of Dr. Southard is
to be seen an epitome of war neurology not elsewhere to be found.
Charles K. Mills.
Philadelphia, May, 1919.
 TABLE OF CONTENTS

SECTION A. PSYCHOSES INCIDENTAL IN THE WAR

I. The Syphilitic Group (Syphilopsychoses)

Case Page
1. Desertion of an officer Briand, 1915 8
2. Visions of a naval officer Carlill, Fildes, 9
Baker, 1917
3. Aggravation of Weygandt, 1915 10
neurosyphilis by war
4. Same Hurst, 1917 10
5. Same Beaton, 1915 10
6. Same Boucherot, 1915 11
7. Same Todd, 1917 12
8. Same Farrar, 1917 13
9. Same Marie, Chatelin, 14
Patrikios,
1917
10. Root-sciatica Long, 1916 15
11. Disciplinary Kastan, 1916 17
12. Same Kastan, 1916 18
13. Same? Kastan, 1916 19
14. Hysterical chorea versus de Massary, du 20
neurosyphilis Sonich, 1917
15. Traumatic general paresis Hurst, 1917 22
16. Head trauma; shell-shock; Babonneix, 23
mania; W. R. positive David, 1917
17. Head trauma in a syphilitic Babonneix, 24
David, 1917
18. Shell wound: general Boucherot, 1915 25
paresis
19. “Shell-shock” ocular palsy: Schuster, 1915 26
syphilitic
20. Shell-shock: general Donath, 1915 27
paresis
21. Shell-shock: tabes Logre, 1917 28
22. Same Duco, Blum, 28
1917
23. Pseudotabes (Shell-shock) Pitres, Marchand, 29
1916
24. Shell-shock neurosyphilis Hurst, 1917 30
25. Shell-shock neurosyphilis Hurst, 1917 31
26. Pseudoparesis (Shell- Pitres, Marchand, 32
shock) 1916
27. War strain and Shell-shock Karplus, 1915 34
in a syphilitic
28. Shell-shock recurrence of Mairet, Piéron, 36
syphilitic hemiplegia 1915
29. Shell-shock (functional!) Laignel- 37
amaurosis in a Lavastine,
neurosyphilitic Courbon,
1916
30. Shell-shock (functional) Babonneix, 39
phenomena in a David, 1917
neurosyphilitic
31. Vestibular symptoms in a Guillain, Barré, 40
neurosyphilitic 1916
32. Syphilophobic suicidal Colin, Lautier, 41
attempts 1917
33. Simulated chancre Pick, 1916 42
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com