Principles of Functional Analysis Second Edition Martin Schechter

Full download ebook at ebookgate.
com
Principles of Functional Analysis Second

Edition Martin Schechter
https://ebookgate.com/product/principles-of-
functional-analysis-second-edition-martin-
schechter/
Download more ebook from https://ebookgate.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Principles of Functional Verification 1st Edition

Andreas Meyer
https://ebookgate.com/product/principles-of-functional-
verification-1st-edition-andreas-meyer/
Hydrology Principles Analysis and Design Second Edition

H.M. Raghunath
https://ebookgate.com/product/hydrology-principles-analysis-and-
design-second-edition-h-m-raghunath/
Obstinate Hebrews Ronald Schechter
https://ebookgate.com/product/obstinate-hebrews-ronald-schechter/
The Functional Analysis of English 3rd Edition Thomas

Bloor
https://ebookgate.com/product/the-functional-analysis-of-
english-3rd-edition-thomas-bloor/
Functional Analysis 2nd Edition Balmohan Vishnu Limaye
https://ebookgate.com/product/functional-analysis-2nd-edition-
balmohan-vishnu-limaye/
Principles of Insurance Second Edition Loma
https://ebookgate.com/product/principles-of-insurance-second-
edition-loma/
Phonological Analysis A Functional Approach Donald A.

Burquest
https://ebookgate.com/product/phonological-analysis-a-functional-
approach-donald-a-burquest/
PRINCIPLES OF MATHEMATICAL ANALYSIS 3rd Edition Walter

Rudin
https://ebookgate.com/product/principles-of-mathematical-
analysis-3rd-edition-walter-rudin/
Trace Environmental Quantitative Analysis Principles

Techniques and Applications Second Edition Paul R.
Loconto (Author)
https://ebookgate.com/product/trace-environmental-quantitative-
analysis-principles-techniques-and-applications-second-edition-
paul-r-loconto-author/
Principles of
Functional Analysis
SECOND EDITION
Martin Schechter
Graduate Studies
in Mathematics
Volume 36
American Mathematical Society

Selected Titles in This Series
36 Martin Schechter, Principles of functional analysis, second edition, 2002
35 James F. Davis and Paul Kirk, Lecture notes in algebraic topology, 2001
34 Sigurdur Helgason, Differential geometry, Lie groups, and symmetric spaces, 2001
33 Dmitri Burago, Yuri Burago, and Sergei Ivanov, A course in metric geometry, 2001
32 Robert G. Bartle, A modern theory of integration, 2001
31 Ralf Korn and Elke Korn, Option pricing and portfolio optimization: Modern methods
of financial mathematics, 2001
30 J. C. McConnell and J. C. Robson, Noncommutative Noetherian rings, 2001
29 Javier Duoandikoetxea, Fourier analysis, 2001
28 Liviu I. Nicolaescu, Notes on Seiberg-Witten theory, 2000
27 Thierry Aubin, A course in differential geometry, 2001
26 Rolf Berndt, An introduction to symplectic geometry, 2001
25 Thomas Friedrich, Dirac operators in Riemannian geometry, 2000
24 Helmut Koch, Number theory: Algebraic numbers and functions, 2000
23 Alberto Candel and Lawrence Conlon, Foliations I, 2000
22 Günter R. Krause and Thomas H. Lenagan, Growth of algebras and Gelfand-Kirillov
dimension, 2000
21 John B. Conway, A course in operator theory, 2000
20 Robert E. Gompf and András I. Stipsicz, 4-manifolds and Kirby calculus, 1999
19 Lawrence C. Evans, Partial differential equations, 1998
18 Winfried Just and Martin Weese, Discovering modern set theory. II: Set-theoretic
tools for every mathematician, 1997
17 Henryk Iwaniec, Topics in classical automorphic forms, 1997
16 Richard V. Kadison and John R. Ringrose, Fundamentals of the theory of operator
algebras. Volume II: Advanced theory, 1997
15 Richard V. Kadison and John R. Ringrose, Fundamentals of the theory of operator
algebras. Volume I: Elementary theory, 1997
14 Elliott H. Lieb and Michael Loss, Analysis, 1997
13 Paul C. Shields, The ergodic theory of discrete sample paths, 1996
12 N. V. Krylov, Lectures on elliptic and parabolic equations in Hölder spaces, 1996
11 Jacques Dixmier, Enveloping algebras, 1996 Printing
10 Barry Simon, Representations of finite and compact groups, 1996
9 Dino Lorenzini, An invitation to arithmetic geometry, 1996
8 Winfried Just and Martin Weese, Discovering modern set theory. I: The basics, 1996
7 Gerald J. Janusz, Algebraic number fields, second edition, 1996
6 Jens Carsten Jantzen, Lectures on quantum groups, 1996
5 Rick Miranda, Algebraic curves and Riemann surfaces, 1995
4 Russell A. Gordon, The integrals of Lebesgue, Denjoy, Perron, and Henstock, 1994
3 William W. Adams and Philippe Loustaunau, An introduction to Gröbner bases,
1994
2 Jack Graver, Brigitte Servatius, and Herman Servatius, Combinatorial rigidity,
1993
1 Ethan Akin, The general topology of dynamical systems, 1993
Principles of
Functional Analysis
Principles of
Functional Analysis
SECOND EDITION
Martin Schechter
Graduate Studies
in Mathematics
Volume 36
American Mathematical Society

Providence, Rhode Island
Editorial Board
Steven G. Krantz
David Saltman (Chair)
David Sattinger
Ronald Stern
2000 Mathematics Subject Classification. Primary 46–01, 47–01, 46B20, 46B25, 46C05,
47A05, 47A07, 47A12, 47A53, 47A55.
Abstract. The book is intended for a one-year course for beginning graduate or senior under-
graduate students. However, it can be used at any level where the students have the prerequisites
mentioned below. Because of the crucial role played by functional analysis in the applied sciences
as well as in mathematics, the author attempted to make this book accessible to as wide a spec-
trum of beginning students as possible. Much of the book can be understood by a student having
taken a course in advanced calculus. However, in several chapters an elementary knowledge of
functions of a complex variable is required. These include Chapters 6, 9, and 11. Only rudimen-
tary topological or algebraic concepts are used. They are introduced and proved as needed. No
measure theory is employed or mentioned.
Library of Congress Cataloging-in-Publication Data

Schechter, Martin.
Principles of functional analysis / Martin Schechter.—2nd ed.
p. cm. — (Graduate studies in mathematics, ISSN 1065-7339 ; v. 36)
Includes bibliographical references and index.
ISBN 0-8218-2895-9 (alk. paper)
1. Functional analysis. I. Title. II. Series.
QA320 .S32 2001
515.7—dc21 2001031601
Copying and reprinting. Individual readers of this publication, and nonprofit libraries
acting for them, are permitted to make fair use of the material, such as to copy a chapter for use
in teaching or research. Permission is granted to quote brief passages from this publication in
reviews, provided the customary acknowledgment of the source is given.
Republication, systematic copying, or multiple reproduction of any material in this publication
is permitted only under license from the American Mathematical Society. Requests for such
permission should be addressed to the Assistant to the Publisher, American Mathematical Society,
P. O. Box 6248, Providence, Rhode Island 02940-6248. Requests can also be made by e-mail to
[email protected].

c 2002 by the American Mathematical Society. All rights reserved.
The American Mathematical Society retains all rights
except those granted to the United States Government.
Printed in the United States of America.

∞ The paper used in this book is acid-free and falls within the guidelines
established to ensure permanence and durability.
Visit the AMS home page at URL: http://www.ams.org/
10 9 8 7 6 5 4 3 2 17 16 15 14 13 12
BS D
To my wife, children, and grandchildren.
May they enjoy many happy years.

Contents
PREFACE TO THE REVISED EDITION xv

FROM THE PREFACE TO THE FIRST EDITION xix
Chapter 1. BASIC NOTIONS 1

§1.1. A problem from differential equations 1
§1.2. An examination of the results 6
§1.3. Examples of Banach spaces 9
§1.4. Fourier series 17
§1.5. Problems 24
Chapter 2. DUALITY 29
§2.1. The Riesz representation theorem 29
§2.2. The Hahn-Banach theorem 33
§2.3. Consequences of the Hahn-Banach theorem 36
§2.4. Examples of dual spaces 39
§2.5. Problems 51
Chapter 3. LINEAR OPERATORS 55
§3.1. Basic properties 55
§3.2. The adjoint operator 57
§3.3. Annihilators 59
§3.4. The inverse operator 60
§3.5. Operators with closed ranges 66
§3.6. The uniform boundedness principle 71
ix
x Contents
§3.7. The open mapping theorem 71

§3.8. Problems 72
Chapter 4. THE RIESZ THEORY FOR COMPACT OPERATORS 77
§4.1. A type of integral equation 77
§4.2. Operators of finite rank 85
§4.3. Compact operators 88
§4.4. The adjoint of a compact operator 95
§4.5. Problems 98
Chapter 5. FREDHOLM OPERATORS 101
§5.1. Orientation 101
§5.2. Further properties 105
§5.3. Perturbation theory 109
§5.4. The adjoint operator 112
§5.5. A special case 114
§5.6. Semi-Fredholm operators 117
§5.7. Products of operators 123
§5.8. Problems 126
Chapter 6. SPECTRAL THEORY 129
§6.1. The spectrum and resolvent sets 129
§6.2. The spectral mapping theorem 133
§6.3. Operational calculus 134
§6.4. Spectral projections 141
§6.5. Complexification 147
§6.6. The complex Hahn-Banach theorem 148
§6.7. A geometric lemma 150
§6.8. Problems 151
Chapter 7. UNBOUNDED OPERATORS 155
§7.1. Unbounded Fredholm operators 155
§7.2. Further properties 161
§7.3. Operators with closed ranges 164
§7.4. Total subsets 169
§7.5. The essential spectrum 171
§7.6. Unbounded semi-Fredholm operators 173
§7.7. The adjoint of a product of operators 177
Contents xi
§7.8. Problems 179

Chapter 8. REFLEXIVE BANACH SPACES 183
§8.1. Properties of reflexive spaces 183
§8.2. Saturated subspaces 185
§8.3. Separable spaces 188
§8.4. Weak convergence 190
§8.5. Examples 192
§8.6. Completing a normed vector space 196
§8.7. Problems 197
Chapter 9. BANACH ALGEBRAS 201
§9.1. Introduction 201
§9.2. An example 205
§9.3. Commutative algebras 206
§9.4. Properties of maximal ideals 209
§9.5. Partially ordered sets 211
§9.6. Riesz operators 213
§9.7. Fredholm perturbations 215
§9.8. Semi-Fredholm perturbations 216
§9.9. Remarks 222
§9.10. Problems 222
Chapter 10. SEMIGROUPS 225
§10.1. A differential equation 225
§10.2. Uniqueness 228
§10.3. Unbounded operators 229
§10.4. The infinitesimal generator 235
§10.5. An approximation theorem 238
Chapter 11. HILBERT SPACE 243
§11.1. When is a Banach space a Hilbert space? 243
§11.2. Normal operators 246
§11.3. Approximation by operators of finite rank 252
§11.4. Integral operators 253
§11.5. Hyponormal operators 257
xii Contents
Chapter 12. BILINEAR FORMS 265

§12.1. The numerical range 265
§12.2. The associated operator 266
§12.3. Symmetric forms 268
§12.4. Closed forms 270
§12.5. Closed extensions 274
§12.6. Closable operators 278
§12.7. Some proofs 281
§12.8. Some representation theorems 284
§12.9. Dissipative operators 285
§12.10. The case of a line or a strip 290
§12.11. Selfadjoint extensions 294
Chapter 13. SELFADJOINT OPERATORS 297
§13.1. Orthogonal projections 297
§13.2. Square roots of operators 299
§13.3. A decomposition of operators 304
§13.4. Spectral resolution 306
§13.5. Some consequences 311
§13.6. Unbounded selfadjoint operators 314
Chapter 14. MEASURES OF OPERATORS 325
§14.1. A seminorm 325
§14.2. Perturbation classes 329
§14.3. Related measures 332
§14.4. Measures of noncompactness 339
§14.5. The quotient space 341
§14.6. Strictly singular operators 342
§14.7. Norm perturbations 345
§14.8. Perturbation functions 350
§14.9. Factored perturbation functions 354
Chapter 15. EXAMPLES AND APPLICATIONS 359
§15.1. A few remarks 359
Contents xiii
§15.2. A differential operator 360

§15.3. Does A have a closed extension? 363
§15.4. The closure of A 364
§15.5. Another approach 369
§15.6. The Fourier transform 372
§15.7. Multiplication by a function 374
§15.8. More general operators 378
§15.9. B-Compactness 381
§15.10. The adjoint of Ā 383
§15.11. An integral operator 384
Appendix A. Glossary 393
Appendix B. Major Theorems 405
Bibliography 419
Index 423
PREFACE TO THE
REVISED EDITION
The first edition of Principles of Functional Analysis enjoyed a successful

run of 28 years. In revising the text, we were confronted with a dilemma.
On the one hand, we wanted to incorporate many new developments, but
on the other, we did not want to smother the original flavor of the book.
As one usually does under such circumstances, we settled for a compromise.
We considered only new material related to the original topics or material
that can be developed by means of techniques existing within the original
framework. In particular, we restricted ourselves to normed vector spaces
and linear operators acting between them. (Other topics will have to wait
for further volumes.) Moreover, we have chosen topics not readily available
in other texts.
We added sections to Chapters 3, 5, 7, 9, and 13 and inserted a new

chapter – Chapter 14. (The old Chapter 14 now becomes Chapter 15.)
Added topics include products of operators (Sections 5.7 and 7.7), a more
general theory of semi-Fredholm operators (Sections 5.6 and 7.6), Riesz op-
erators (Section 9.6), Fredholm and semi-Fredholm perturbations (Sections
9.6 and 9.7), spectral theory for unbounded selfadjoint operators (Section
13.6), and measures of operators and perturbation functions (Chapter 14).
We attempted to strengthen those areas in the book that demonstrate

its unique character. In particular, new material introduced concerning
Fredholm and semi-Fredholm operators requires minimal effort since the
xv
xvi PREFACE TO THE REVISED EDITION
required machinery is already in place. By these means we were able to

provide very useful information while keeping within our guidelines.
The new chapter (Chapter 14) deserves some additional remarks. It

is designed to show the student how methods that were already mastered
can be used to attack new problems. We gathered several topics which are
new, but relate only to those concepts and methods emanating from other
parts of the book. These topics include perturbation classes, measures of
noncompactness, strictly singular operators and operator constants. This
last topic illustrates in a very surprising way how a constant associated with
an operator can reveal a great deal of information concerning the operator.
No new methods of proof are needed, and, again, most of this material
cannot be readily found in other books.
We went through the entire text with a fine toothed comb. The presen-
tation was clarified and simplified whenever necessary, and misprints were
corrected. Existing lemmas, theorems, corollaries and proofs were expanded
when more elaboration was deemed beneficial. New lemmas, theorems and
corollaries (with proofs) were introduced as well. Many new problems were
added.
We have included two appendices. The first gives the definitions of

important terms and symbols used throughout the book. The second lists
major theorems and indicates the pages on which they can be found.
The author would like to thank Richard Jasiewicz for installing LATEX 2ε
into his computer. He would also like to thank the editors and staff of the
AMS for helpful suggestions.
Irvine, California
March, 2001
T V SLB O
PREFACE TO THE REVISED EDITION xvii
The following are a few excerpts from a review of the original edition by
Einar Hille in the American Scientist.1
“ ‘Charming’ is a word that seldom comes to the mind of a science

reviewer, but if he is charmed by a treatise, why not say so? I am charmed
by this book.”
“Professor Schechter has written an elegant introduction to functional

analysis including related parts of the theory of integral equations. It is
easy to read and is full of important applications. He presupposes very little
background beyond advanced calculus; in particular, the treatment is not
burdened by topological ‘refinements’ which nowadays have a tendency of
dominating the picture.”
“The book can be warmly recommended to any reader who wants to

learn about this subject without being deterred by less relevant introductory
matter or scared away by heavy prerequisites.”
1From Hille, Einar, Review of Principles of Functional Analysis, American Scientist {Vol.
60}, No. 3, 1972, 390.
FROM THE
PREFACE TO THE
FIRST EDITION
Because of the crucial role played by functional analysis in the applied sci-
ences as well as in mathematics, I have attempted to make this book ac-
cessible to as wide a spectrum of beginning students as possible. Much of
the book can be understood by a student having taken a course in advanced
calculus. However, in several chapters an elementary knowledge of functions
of a complex variable is required. These include Chapters 6, 9, and 11. Only
rudimentary topological or algebraic concepts are used. They are introduced
and proved as needed. No measure theory is employed or mentioned.
The book is intended for a one-year course for beginning graduate or

senior undergraduate students. However, it can be used at any level where
the students have the prerequisites mentioned above.
I have restricted my attention to normed vector spaces and their impor-

tant examples, Banach and Hilbert spaces. These are venerable institutions
upon which every scientist can rely throughout his or her career. They are
presently the more important spaces met in daily life. Another considera-
tion is the fact that an abundance of types of spaces can be an extremely
confusing situation to a beginner.
xix
xx FROM THE PREFACE TO THE FIRST EDITION
I have also included some topics which are not usually found in textbooks
on functional analysis. A fairly comprehensive treatment of Fredholm op-
erators is given in Chapters 5 and 7. I consider their study elementary.
Moreover, they are natural extensions of operators of the form I − K, K
compact. They also blend naturally with other topics. Additional top-
ics include unbounded semi-Fredholm operators and the essential spectrum
considered in Chapter 7. Hyponormal and seminormal operators are treated
in Chapter 11, and the numerical range of an unbounded operator is studied
in Chapter 12. The last chapter is devoted to the study of three types of
operators on the space L2 (−∞, ∞).
One will notice that there are few applications given in the book other
than those treated in the last chapter. In general, I used as many illus-
trations as I could without assuming more mathematical knowledge than is
needed to follow the text. Moreover, one of the basic philosophies of the
book is that the theory of functional analysis is a beautiful subject which
can be motivated and studied for its own sake. On the other hand, I have
devoted a full chapter to applications that use a minimum of additional
knowledge.
The approach of this book differs substantially from that of most other
mathematics books. In general one uses a “tree” or “catalog” structure, in
which all foundations are developed in the beginning chapters, with later
chapters branching out in different directions. Moreover, each topic is in-
troduced in a logical and indexed place, and all the material concerning
that topic is discussed there complete with examples, applications, refer-
ences to the literature and descriptions of related topics not covered. Then
one proceeds to the next topic in a carefully planned program. A descriptive
introduction to each chapter tells the reader exactly what will be done there.
In addition, we are warned when an important theorem is approaching. We
are even told which results are of “fundamental importance.” There is much
to be said for this approach. However, I have embarked upon a different
path. After introducing the first topic, I try to follow a trend of thought
wherever it may lead without stopping to fill in details. I do not try to de-
scribe a subject fully at the place it is introduced. Instead, I continue with
my trend of thought until further information is needed. Then I introduce
the required concept or theorem and continue with the discussion.
This approach results in a few topics being covered in several places in

the book. Thus, the Hahn-Banach theorem is discussed in Chapters 2 and
FROM THE PREFACE TO THE FIRST EDITION xxi
9, with a complex form given in Chapter 6, and a geometric form in Chapter

7. Another result is that complex Banach spaces are not introduced until
Chapter 6, the first place that their advantage is clear to the reader.
This approach has further resulted in a somewhat unique structure for

the book. The first three chapters are devoted to normed vector spaces, and
the next four to arbitrary Banach spaces. Chapter 8 deals with reflexive
Banach spaces, and Chapters 11 – 13 cover Hilbert spaces. Chapters 9 and
10 discuss special topics.
Chapter 1
BASIC NOTIONS
1.1. A problem from differential equations

Suppose we are given the problem of finding a solution of
(1.1) f (x) + f (x) = g(x)
in an interval a ≤ x ≤ b with the solution satisfying
(1.2) f (a) = 1, f (a) = 0.
(We shall not enter into a discussion as to why anyone would want to solve
this problem, but content ourselves with the statement that such equations
do arise in applications.) From your course in differential equations you will
recall that when g = 0, equation (1.1) has a general solution of the form
(1.3) f (x) = A sin x + B cos x,
where A and B are arbitrary constants. However, if we are interested in
solving (1.1) for g(x) an arbitrary function continuous in the closed inter-
val, not many of the methods developed in the typical course in differential
equations will be of any help. A method which does work is the least pop-
ular and would rather be forgotten by most students. It is the method of
variation of parameters which states, roughly, that one can obtain a solution
of (1.1) if one allows A and B to be functions of x instead of just constants.
Since we are only interested in a solution of (1.1), we shall not go into any
justification of the method, but merely apply it and then check to see if
what we get is actually a solution. So we differentiate (1.3) twice, substitute
into (1.1) and see what happens. Before proceeding, we note that we shall
get one equation with two unknown functions. Since we were brought up
from childhood to believe that one should have two equations to determine
1
2 1. BASIC NOTIONS
two functions, we shall feel free to impose a further restriction on A and B,

especially if such action will save labor on our part. So here we go:
f = A cos x − B sin x + A sin x + B cos x.
Now it becomes clear to us that further differentiation will yield eight terms,
a circumstance that should be avoided if possible. Moreover, the prospect
of obtaining higher order derivatives of A and B does not appeal to us. So
we make the perfectly natural assumption
(1.4) A sin x + B cos x = 0.
Thus,
f = A cos x − B sin x − f,
showing that we must have
g = A cos x − B sin x.
Combining this with (1.4), we get
A = g cos x, B = −g sin x.
From the initial conditions (1.2) we see that B(a) = cos a, A(a) = sin a.
Thus,
x x
A(x) = sin a + g(t) cos t dt, B(x) = cos a − g(t) sin t dt
a a
and
x
f (x) = cos(x − a) + [sin x cos t − cos x sin t]g(t) dt
a x
(1.5) = cos(x − a) + sin(x − t)g(t) dt.
a
So far so good. Since we made no claims concerning the method, we really
should verify that (1.5) truly is a solution of (1.1) and (1.2). Differentiating
twice, we have
x
f (x) = − sin(x − a) + cos(x − t)g(t) dt
a
x

f (x) = − cos(x − a) − sin(x − t)g(t) dt + g(x)
a
= −f (x) + g(x).
(Make sure to check back in your advanced calculus text about differentiat-
ing an integral with respect to a parameter appearing in the integrand and
in the limits of integration.)
Encouraged by our success, we generalize the problem. Suppose in place
of (1.1) we want to solve
(1.6) f (x) + f (x) = σ(x)f (x), a ≤ x ≤ b,
1.1. A problem from differential equations 3
under the initial conditions (1.2). Here σ(x) is a given function continuous
in the closed interval. One thing is certain. Any solution of (1.6) satisfies
x
(1.7) f (x) = cos(x − a) + sin(x − t)σ(t)f (t) dt.
a
Thus, by what was just shown, a function f (x) is a solution of (1.6), (1.2) if,
and only if, it is a solution of (1.7). Do not believe that this has improved
the situation much. It has merely transformed a differential problem into
one of solving the integral equation (1.7).
A more disturbing fact is that equation (1.7) is rather complicated, and
matters appear to be getting worse unless taken in hand. We must introduce
some shorthand. If we write
x
(1.8) Kh = sin(x − t)σ(t)h(t) dt, u(x) = cos(x − a),
a
then (1.7) takes on the more manageable form
(1.9) f = u + Kf.
The “object” K is called an “operator” or “transformation,” since it acts on
continuous functions and transforms them into other continuous functions.
(We will give a more precise definition in Chapter 3.) Thus, we are looking
for a continuous function f such that Kf added to u gives back f.
Now that (1.7) has been simplified to (1.9), we can think about it a bit
more clearly. It really does seem like a difficult equation to solve. To be sure,
if one takes an arbitrary function f0 and plugs it into the right hand side of
(1.9), one would have to be extremely lucky if u + Kf0 turned out to be f0 .
In general, we would only get some other function f1 , which is probably no
closer to an actual solution (should one exist) than f0 . On second thought,
perhaps it is, in some way. After all, it is obtained by means of the right
hand side of (1.9). Well, if this is the case, let us plug in f1 . This gives
another function f2 = u + Kf1 . Continuation of the procedure leads to a
sequence {fn } of continuous functions defined by
(1.10) fn = u + Kfn−1 ,
where it is hoped that fn is “closer” to a solution than fn−1 . This suggests
that the sequence {fn } might even converge to some limit function f. Would
such an f be a solution? Well, if f is continuous and Kfn converges to Kf,
we then have that (1.9) holds, showing that f is indeed a solution. Thus,
the big question: is our operator K such that these things will happen?
Now before we go further, we must consider what type of convergence we
want. Since we want the limit function to be continuous, it is quite natural
to consider uniform convergence. (Here you are expected to know that the
uniform limit of continuous functions is continuous, and to suspect that it
4 1. BASIC NOTIONS
may not be so if the convergence is otherwise.) Now you may recall that a
continuous function has a maximum on a closed interval and that a sequence
{fn (x)} of continuous functions converges uniformly if and only if
max |fn (x) − fm (x)|
a≤x≤b
can be made as small as we like by taking m and n large enough. Again I

am compelled to pause a moment. The expression above is too tedious to
write often, and since we expect it to occur frequently, more shorthand is in
order. One idea is to put
(1.11) h = max |h(x)|.
a≤x≤b
This is the most appealing to me, so we shall use it. So we now want
(1.12) fn − fm −→ 0 as m, n −→ ∞.
Now let us carry out our program. By (1.10),
f1 = u + Kf0
f2 = u + Kf1
= u + K(u + Kf0 )
= u + Ku + K 2 f0 ,
where K 2 h = K(Kh), and we have used the property
(1.13) K(v + w) = Kv + Kw.
If we define
K n h = K(K n−1 h)
by induction, we have
f3 = u + Kf2 = u + Ku + K 2 u + K 3 f0 ,
(1.14) fn = u + Ku + · · · + K n−1 u + K n f0 .
Thus, for say n > m,
fn − fm = K m u + · · · + K n−1 u + K n f0 − K m f0 .
Now we want (1.12) to hold. Since
(1.15) fn − fm ≤ K m u + · · · + K n−1 u + K n f0 + K m f0 ,
the limit (1.12) would be guaranteed if the right hand side of (1.15) went to
zero as m, n → ∞. Note that we have used the properties
(1.16) v + w ≤ v + w,
(1.17) − v = v,
1.1. A problem from differential equations 5
which are obvious consequences of (1.11). Now the right hand side of (1.15)
converges to zero as m, n → ∞ if for each continuous function v
∞
(1.18) K n v < ∞,
0
where we defined K 0v = v. This follows from the fact that
K m u + · · · + K n−1 u
is contained in the tail end of such a series, and K n f0 is the n-th term of
another such series (perhaps a review of convergent series is in order).
Now before attempting to verify (1.18), we note that it implies that
{fn (x)} is a uniformly convergent Cauchy sequence and hence has a contin-
uous limit f (x). Thus, all we need is
(1.19) Kfn − Kf −→ 0 as n −→ ∞.
Now an examination of (1.8) reveals that there is a constant M such that
(1.20) Kh ≤ M h
for all continuous functions h. In fact, we can take M = (b − a)σ. This
coupled with the further property
(1.21) K(−h) = −Kh,
shows that
Kfn − Kf = K(fn − f ) ≤ M fn − f −→ 0
as n → ∞. This gives (1.19). Thus, we are in business if we can verify (1.18).
As it happens, (1.18) is an easy consequence of (1.8). In fact, we have
x
|Kv| ≤ σ |v(t)| dt ≤ σ v(x − a)
a
for all x in the interval [a, b]. Thus,
x x
|K v| ≤ σ
2
|Kv| dt ≤ σ v
2
(t − a) dt
a a
= σ v(x − a) /2,
2 2
x x
1
|K 3 v| ≤ σ |K 2 v| dt ≤ σ3 v (t − a)2 dt
a 2 a
1
= σ v(x − a) ,
3 3
3!
and by induction
1
|K n v| ≤ σn v(x − a)n .
n!
Thus,
1
K n v ≤ σn v(b − a)n ,
n!
6 1. BASIC NOTIONS
and
∞
∞
1
K v ≤ v
n
σn (b − a)n = veσ(b−a) ,
n!
0 0
giving (1.18).
Now we tie the loose ends together. From property (1.18), we see by
(1.15) that the sequence {fn } defined by (1.10) is a Cauchy sequence con-
verging uniformly. Thus, the limit f is continuous in [a, b]. Moreover, by
(1.20) and (1.21) we see that (1.19) holds, showing that f is a solution of
(1.9), i.e., of (1.7). Differentiation now shows that f is a solution of (1.6),
(1.2). Note that the choice of f0 did not play a role, and we did not have to
know anything about the functions σ or u other than the fact that they are
continuous.
Those of you who are familiar with the Picard iteration method that is
used to solve differential equations will notice a similarity to the presentation
given here.
Let k(x, t) be a function continuous in the triangle a ≤ t ≤ x ≤ b. Then
the method above gives a solution of the integral equation
x
f (x) = u(x) + k(x, t)f (t) dt.
a
This is known as a Volterra equation.
1.2. An examination of the results

We now sit back and contemplate what we have done. We have solved a
problem. We have shown that for any function σ(x) continuous in [a, b] we
can find a solution of (1.6), (1.2). But I claim we have done more. We have
also shown that if K is any operator which maps continuous functions into
continuous functions and satisfies (1.13), (1.18), (1.20) and (1.21), then the
equation (1.9) has a solution. Other properties of K were not needed. Now
one may ask: why did the method work? Could it work in other situations?
What properties of continuous functions were used?
To get some further insight, let us examine some of the well-known
properties of continuous functions. Let C ≡ C[a, b] be the set of functions
continuous on the closed interval [a, b]. We note the following properties:
(1) They can be added. If f and g are in C, so is f + g.

(2) f + (g + h) = (f + g) + h, f, g, h ∈ C.
(3) There is an element 0 ∈ C such that h + 0 = h for all h ∈ C.
(4) For each h ∈ C there is an element −h ∈ C such that h + (−h) = 0.
(5) g + h = h + g, g, h ∈ C.
1.2. An examination of the results 7
(6) For each real number α, αh ∈ C.

(7) α(g + h) = αg + αh.
(8) (α + β)h = αh + βh.
(9) α(βh) = (αβ)h.
(10) To each h ∈ C there corresponds a real number h (called a norm
with the following properties:
(11) αh = |α| h.
(12) h = 0 if, and only if, h = 0.
(13) g + h ≤ g + h.
(14) If {hn } is a sequence of elements of C such that hn − hm → 0 as
m, n → ∞, then there is an element h ∈ C such that hn − h → 0
as n → ∞.
Note that we haven’t used all of the properties (1)–(14) in solving our prob-
lem, but all of the properties used are among them.
Now for some definitions. A collection of objects which satisfies state-
ments (1)–(9) and the additional statement
(15) 1h = h
is called a vector space (VS) or linear space. We will be using real scalars
until we reach a point where we will be forced to allow complex scalars. A
set of objects satisfying statements (1)–(13) is called a normed vector space
(NVS), and the number h is called the norm of h. Although statement (15)
is not implied by statements (1)–(9), it is implied by statements (1)–(13).
A sequence satisfying
hn − hm → 0 as m, n → ∞,
is called a Cauchy sequence. Property (14) states that every Cauchy se-
quence converges in norm to a limit. Property (14) is called completeness,
and a normed vector space satisfying it is called a complete normed vector
space or a Banach space. Thus, we have proved the following
Theorem 1.1. Let X be a Banach space, and assume that K is an operator
on X (i.e., maps X into itself ) such that
a) K(v + w) = Kv + Kw,
b) K(−v) = −Kv,
8 1. BASIC NOTIONS
c) Kv ≤ M v,

d) ∞ 0 K v < ∞
n
for all v, w ∈ X. Then for each u ∈ X there is a unique f ∈ X such that

(1.22) f = u + Kf.
The uniqueness in Theorem 1.1 is trivial. In fact, suppose there were

two solutions f1 and f2 of (1.22). Set f = f1 − f2 . Then by a) and b) we
have
f = Kf.
From this we get
f = K 2f = K 3f = · · · = K nf
for each n. Thus,
f = K n f → 0
as n → ∞ by d). Since f does not depend on n, we have f = 0, and
conclude that f = 0.
A special case of Theorem 1.1 is very important. If K satisfies a) and
b) and
(1.23) Kv ≤ θv, v ∈ X,
for some θ satisfying 0 < θ < 1, then c) and d) are both satisfied. For, we
have
K n v ≤ θK n−1 v ≤ θ2 K n−2 v ≤ · · · ≤ θn v.
Thus,
∞
∞

K v ≤ v
n
θn = v/(1 − θ).
0 0
As an example, let k(x, t) be a continuous function in the square a ≤

x, t ≤ b. Then the equation
b
f (x) = u(x) + k(x, t)f (t) dt
a
is called a Fredholm integral equation of the second kind. If
max |k(x, t)| < 1/(b − a),
a≤x,t≤b
then the operator

b
Kh = k(x, t)h(t) dt
a
satisfies (1.23), and Theorem 1.1 applies.
1.3. Examples of Banach spaces 9
1.3. Examples of Banach spaces

We now consider some other Banach spaces.
Example 1. The most familiar is Rn , Euclidean n-dimensional real space.
It consists of sequences of n real numbers
f = (α1 , · · · , αn ), g = (β1 , · · · , βn ),
where addition is defined by
f + g = (α1 + β1 , · · · , αn + βn ),
and multiplication by a scalar is defined by
γf = (γα1 , · · · , γαn ).
Under these definitions, Rn is a vector space. If we set
1
(1.24) f = (α12 + · · · + αn2 ) 2 ,
the only axioms of a Banach space which are not immediately verified are
the triangle inequality (property (13)) and completeness (property (14)).
We verify completeness first.
Let
(j)
fj = (α1 , · · · , αn(j) )
be a Cauchy sequence in Rn , i.e., assume
fj − fk −→ 0 as j, k −→ ∞.
Thus, for any ε > 0, one can find a number N so large that
(j) (k)
(1.25) fj − fk 2 = (α1 − α1 )2 + · · · + (αn(j) − αn(k) )2 < ε2
whenever j, k > N. In particular,
(j) (k)
|α1 − α1 | < ε, j, k > N.
(j)
This means that the sequence {α1 } is a Cauchy sequence of real numbers,
which according to a well known theorem of advanced calculus has a limit.
Thus, there is a real number α1 such that
(j)
α1 −→ α1 as j −→ ∞.
The same reasoning shows that for each l = 1, · · · , n,
(j)
αl −→ αl as j −→ ∞.
Set f = (α1 , · · · , αn ). Then f ∈ Rn . Now letting k → ∞ in (1.25), we have
(j)
(α1 − α1 )2 + · · · + (αn(j) − αn )2 ≤ ε2 f or j > N.
But this is precisely
fj − f 2 ≤ ε2 .
10 1. BASIC NOTIONS
Thus,
fj − f −→ 0 as j −→ ∞.
Now for the triangle inequality. If it is to hold, we must have
f + g2 ≤ (f + g)2 = f 2 + g2 + 2f g.
In other words, we want to prove
1 1
(αi2 + 2αi βi + βi2 ) ≤ αi2 + βi2 + 2( αi2 ) 2 ( βi2 ) 2 ,
or
1 1
(1.26) αi βi ≤ ( αi2 ) 2 ( βi2 ) 2 .
Now, before things become more complicated, set

n
(1.27) (f, g) = αi βi .
1
This expression has the following obvious properties:
i. (αf, g) = α(f, g)
ii. (f + g, h) = (f, h) + (g, h)
iii. (f, g) = (g, f )
iv. (f, f ) > 0 unless f = 0,
and we want to prove

(1.28) (f, g)2 ≤ (f, f )(g, g).
Lemma 1.2. Inequality (1.28) follows from the aforementioned properties.
Proof. Let α be any scalar. Then

(αf + g, αf + g) = α2 (f, f ) + 2α(f, g) + (g, g)

(f, g) (f, g)2 (f, g)2
2
= (f, f ) α + 2α + + (g, g) −
(f, f ) (f, f )2 (f, f )
2 2
(f, g) (f, g)
= (f, f ) α + + (g, g) − ,
(f, f ) (f, f )
where we have completed the square with respect to α and tacitly assumed
that (f, f ) = 0. This assumption is justified by the fact that if (f, f ) = 0,
then (1.28) holds vacuously. We now note that the left-hand side of (1.29) is
nonnegative by property iv listed above. If we now take α = −(f, g)/(f, f ),
this inequality becomes
(f, g)2
0 ≤ (g, g) − ,
(f, f )
which is exactly what we want.
Returning to Rn , we see that (1.26) holds and hence, the triangle in-
equality is valid. Thus, Rn is a Banach space.
An expression (f, g) that assigns a real number to each pair of elements
of a vector space and satisfies the aforementioned properties is called a scalar
(or inner) product. We have essentially proved
Lemma 1.3. If a vector space X has a scalar product (f, g), then it is a
1
normed vector space with norm f = (f, f ) 2 .
Proof. Again, the only thing that is not immediate is the triangle inequality.
This follows from (1.28) since
f + g2 = f 2 + g2 + 2(f, g)
≤ f 2 + g2 + 2f g = (f + g)2 .
This gives the desired result.
A vector space which has a scalar product and is complete with respect to
the induced norm is called a Hilbert space. By Lemma 1.3, every Hilbert
space is a Banach space, but we shall soon see that the converse is not true.
Inequalities (1.26) and (1.28) are known as the Cauchy-Schwarz inequalities.
Now that we know that Rn is a Banach space, we can apply Theorem

1.1. In particular, if
1 1 1
Kf = ( α1 , α2 , · · · , αn ),
2 3 n+1
we know that K satisfies the hypotheses of that theorem and we can always
solve
f = u + Kf
for any u ∈ R . The same is true for the operator
n
1
Kf = (α2 , · · · , αn , 0).
2
As another application of Theorem 1.1, consider the system of equations
n
(1.29) xj − ajk xk = bj , j = 1, · · · , n.
k=1
We are given the coefficients ajk and the constants bj , and we wish to solve
for the xj . One way of solving is to take b = (b1 , · · · , bn ), x = (x1 , · · · , xn ) ∈
Rn and
n
n
Kx = ( a1k xk , · · · , ank xk ).
k=1 k=1
12 1. BASIC NOTIONS
Then,
n 2

n
Kx2 = ajk xk
j=1 k=1
n

n
n
≤ a2jk x2k
j=1 k=1 k=1
⎛ ⎞

n
≤⎝ a2jk ⎠ x2 .
j,k=1
If we assume that

n
a2jk ≤ θ < 1,
j,k=1
then we can apply Theorem 1.1 to conclude that for every b ∈ Rn there is a
unique x ∈ Rn such that x = b + Kx. This solves (1.29).
Example 2. Another example is given by the space l∞ (the reason for
the odd notation will be given later). It consists of infinite sequences of real
numbers
(1.30) f = (α1 , · · · , αn , · · · )
for which
sup |αi | < ∞.
i
(The sup of any set of real numbers is the least upper bound, i.e., the
smallest number, which may be +∞, that is an upper bound for the set. An
important property of the real numbers is that every set of real numbers has
a least upper bound.) If we define addition and multiplication by a scalar
as in the case of Rn , we come up with a vector space. But we want more.
We want l∞ to be a Banach space with norm
f = sup |αi |.
i
As one will find in most examples, the only properties which are not imme-
diately obvious are the triangle inequality and completeness. In this case
the triangle inequality is not far from it, since
sup |αi + βi | ≤ sup(|αi | + |βi |) ≤ sup |αi | + sup |βi |.
i i i i
It remains to prove completeness. Suppose
(j) (j)
fj = (α1 , · · · , αi , · · · )
is a Cauchy sequence in l∞ . Then for each ε > 0 there is an N so large that
fj − fk < ε when j, k > N.
In particular for each i

(j) (k)
(1.31) |αi − αi | < ε for j, k > N.
(j)
Thus, for each i the sequence {αi } is a Cauchy sequence of real numbers,
so that there is a number αi for which
(j)
αi −→ αi as j −→ ∞.
Set f = (α1 , · · · , αn , · · · ). Is f ∈ l∞ ? Well, let us see. If we fix j and let

k → ∞ in (1.31), we have
(j)
|αi − αi | ≤ ε for j > N.
Thus,
(j) (j)
|αi | ≤ |αi − αi | + |αi | ≤ ε + fj
for j > N. This shows that f ∈ l∞ . But this is not enough. We want
fj − f −→ 0 as j −→ ∞.
But by (1.31), we have

(j)
fj − f = sup |αi − αi | ≤ ε for j > N.
i
This is precisely what we want.

Again if we set
1 1
Kf = ( α1 , · · · , αi , · · · )
2 i+1
or
1 1
Kf = ( α2 , · · · , αi+1 , · · · ),
2 2
we can solve the equation f = u + Kf by Theorem 1.1. We obtain the
solutions as the limits of convergent series.
We can also consider the infinite system of equations
∞

(1.32) xj − ajk xk = bj , j = 1, · · · .
k=1
As before, we are given the coefficients ajk and the constants bj , and we
wish to solve for the xj . One way of solving is to take b = (b1 , · · · ), x =
(x1 , · · · ) ∈ l∞ and

n
Kx = ( a1k xk , · · · ).
k=1
14 1. BASIC NOTIONS
Then,
∞

Kx = sup ajk xk
j
k=1
∞

≤ sup |ajk | · |xk |
j
k=1

≤ sup |ajk | · x∞ .
j
k
If we assume that
∞

sup |ajk | ≤ θ < 1,
j
k=1
then we can apply Theorem 1.1 to conclude that for every b ∈ l∞ there is a
unique x ∈ l∞ such that x = b + Kx. This solves (1.32).
Example 3. A similar example is the space l2 . (Recall the remark con-
cerning l∞ .) It consists of all sequences of the form (1.30) for which
∞

αi2 < ∞.
1
Here we are not immediately sure that the sum of two elements of l2 is in l2 .
Looking ahead a moment, we intend to investigate whether l2 is a Banach
space. Now the candidate most likely to succeed as a norm is
∞ 1
2
f = αi2 .
1
Thus, if we can verify the triangle inequality, we will also have shown that
the sum of two elements of l2 is in l2 . Now, since the triangle inequality
holds in Rn , we have

n 1 1
(αi + βi )2 ≤ αi2 + βi2 + 2( αi2 ) 2 ( βi2 ) 2
1
≤ f 2 + g2 + 2f g = (f + g)2

for any n. Letting n → ∞, we see that f + g ∈ l2 , and that the triangle
inequality holds. Thus, we have a normed vector space. To check if l2 is
complete, let
(j) (j)
fj = (α1 , · · · , αi , · · · )
be a Cauchy sequence in l2 . Thus, for each ε > 0, there is an N such that
∞
(j) (k)
(1.33) (αi − αi )2 < ε2 for j, k > N.
i=1
In particular, (1.31) holds for each i, and hence, there is a number αi which
(j)
is the limit of αi as j → ∞. Now by (1.33), for each n

n
(j) (k)
(1.34) (αi − αi )2 < ε2 for j, k > N.
i=1
Fix j in (1.34) and let k → ∞. Then

n
(j)
(αi − αi )2 ≤ ε2 for j > N.
i=1
Since this is true for each n, we have

∞
(j)
(1.35) (αi − αi )2 ≤ ε2 for j > N.
i=1
Set
(j) (j)
hj = (α1 − α1 , · · · , αi − αi , · · · ).
By (1.35), hj ∈ l2 and hj ≤ ε when j > N. Hence,
f = fj − hj = (α1 , · · · , αi , · · · )
is in l2 , and
fj − f = hj ≤ ε for j > N.
This means that
fj − f −→ 0 as j −→ ∞.
Thus, l2 is a Banach space.
Since Rn is a Hilbert space, one might wonder whether the same is true
of l2 . The scalar product for Rn is given by (1.27), so that the natural
counterpart for l2 should be
∞

(1.36) (f, g) = αi βi ,
1
provided this series converges. By (1.26) we have

n
n
1
n
1
|αi βi | ≤ ( αi2 ) 2 ( βi2 ) 2 ≤ f g,
1 1 1
showing that the series in (1.36) converges absolutely. Thus, (f, g) is defined
for all f, g ∈ l2 , and satisfies i.– iv. (and hence (1.28)). Since (f, f ) = f 2 ,
we see that l2 is a Hilbert space.
16 1. BASIC NOTIONS
We can also solve (1.32) in l2 for b ∈ l2 . In this case

∞
∞ 2

Kx =2
ajk xk
j=1 k=1
∞
∞
∞

≤ a2jk x2k
j=1 k=1 k=1
⎛ ⎞
∞

≤⎝ a2jk ⎠ x2 .
j,k=1
If we assume that
∞

a2jk ≤ θ < 1,
j,k=1
then we can apply Theorem 1.1 to conclude that for every b ∈ l2 there is a
unique x ∈ l2 such that x = b + Kx. This solves (1.32) for this case.
Example 4. Our last example is the set B = B[a, b] of bounded real
valued functions on an interval [a, b]. The norm is
ϕ = sup |ϕ(x)|.
a≤x≤b
The verification that B is a normed vector space is immediate. To check

that it is complete, assume that {ϕj } is a sequence satisfying
ϕj − ϕk −→ 0 as j, k −→ ∞.
Then for each ε > 0, there is an N satisfying
(1.37) sup |ϕj (x) − ϕk (x)| < ε for j, k > N.
Thus, for each x in the interval [a, b], the sequence {ϕj (x)} has a limit cx
as j → ∞. Define the function ϕ(x) to have the value cx at the point x. By
(1.37), for each x ∈ [a, b], we have
|ϕj (x) − ϕk (x)| < ε for j, k > N.
Holding j fixed and letting k → ∞, we get
|ϕj (x) − ϕ(x)| ≤ ε for j > N.
Since this is true for each x, ϕj − ϕ ∈ B for j > N and
ϕj − ϕ ≤ ε for j > N.
Thus, ϕ ∈ B, and it is the limit of the ϕj in B.
Let H be a Hilbert space. For any elements f, g ∈ H, we have
f + g2 = f 2 + 2(f, g) + g2
1.4. Fourier series 17
Figure 1.1
and
f − g2 = f 2 − 2(f, g) + g2 ,
giving
(1.38) f + g2 + f − g2 = 2f 2 + 2g2 .
This is known as the parallelogram law. The name comes from the special
case which states that the sum of the squares of the lengths of the sides
of a parallelogram is equal to the sum of the squares of the lengths of the
diagonals. Thus, it follows that in any Hilbert space, (1.38) must hold for
all elements f, g. This gives us a convenient way of checking whether a given
Banach space is a Hilbert space as well. If one can exhibit two elements f, g
of the Banach space which violate (1.38), then it clearly cannot be a Hilbert
space.
Recall the spaces C and B from Section 1.2 and Example 4. Let us show
that they are not Hilbert spaces. For simplicity take a = 0, b = 3. Define f
to be 1 from 0 to 1, to vanish from 2 to 3 and to be linear from 1 to 2 (see
Figure 1.1). Similarly, let g vanish from 0 to 1, equal 1 from 2 to 3 and be
linear from 1 to 2. Both f and g are continuous in the closed interval [0,3]
and hence are elements of B and C. But
f = g = f + g = f − g = 1,
violating (1.38). Thus, B and C are not Hilbert spaces. In Section 9.1 we
shall show that every Banach space whose elements satisfy (1.38) is also a
Hilbert space.
1.4. Fourier series

Let f (x) be a function having a continuous derivative in the closed interval
[0, 2π] and such that f (2π) = f (0). We can extend f (x) to the whole of R by
making it periodic with period 2π. It will be continuous in all of R, but its
18 1. BASIC NOTIONS
derivative may have jumps at the endpoints of the periods. Then according
to the theory of Fourier series
∞ ∞
a0
(1.39) f (x) = + an cos nx + bn sin nx,
2
1 1
where
2π 2π
1 1
(1.40) an = f (x) cos nx dx, bn = f (x) sin nx dx,
π 0 π 0
and the convergence of (1.39) is uniform. Now our first reaction to all this
is that formulas (1.39) and (1.40) are very complicated. So we recommend
the following simplifications. Set
ϕ0 (x) = (2π)− 2 ,
1 1
α0 = (π/2) 2 a0 ,
ϕ2k (x) = π −1/2 cos kx, ϕ2k+1 (x) = π −1/2 sin kx,
α2k = π 1/2 ak , α2k+1 = π 1/2 bk .
With these definitions, (1.39) and (1.40) become
∞
2π
(1.41) f (x) = αn ϕn , αn = f (x)ϕn (x) dx.
0 0
An important property of the functions ϕn which was used in deriving (1.39)

and (1.40) is
2π
0, if m = n,
(1.42) ϕm (x)ϕn (x) dx =
0 1, if m = n.
Now consider the sequence (α0 , α1 , · · · ) of coefficients in (1.41). A fact which
may or may not be surprising is that it is an element of l2 . For,
n n 2π 2π
2
αj = αj f (x)ϕj (x) dx = f (x)fn (x) dx,
0 0 0 0
where

n
(1.43) fn (x) = αj ϕj (x).
0
Since the Fourier series converges uniformly to f (x) and fn (x) is just a
partial sum, we have
n 2π
αj −→
2
f (x)2 dx as n −→ ∞.
0 0
Thus, (α0 , α1 , · · · ) is an element of l2 , and its norm is

2π 1/2
2
f (x) dx .
0
Figure 1.2
Hence for each f ∈ C 1 (the set of functions with period 2π having continuous
first derivatives), there is a unique sequence (α0 , α1 , · · · ) in l2 such that
∞ 2π
(1.44) αj2 = f (x)2 dx,
0 0
and (1.41) holds. (In general, the symbol C 1 refers to all continuously
differentiable functions. At this point we use it to refer only to periodic
functions having period 2π.)
One might ask if we can go back. If we are given a sequence in l2 , does
there exist a function f ∈ C 1 such that (1.41) and (1.44) hold? The answer
is a resounding no. The reason is simple. Since l2 is complete, every Cauchy
sequence in l2 has a limit in l2 . But does every sequence {fj } of functions
in C 1 such that
2π
(1.45) [fj (x) − fk (x)]2 dx −→ 0 as j, k −→ ∞
0
have a limit in 1
C ? We can see quite easily that this is not the case. In fact
one can easily find a discontinuous function g(x) which can be approximated
by a smooth function h(x) in such a way that
2π
[g(x) − h(x)]2 dx
0
is as small as we like (see Figure 1.2). Another way of looking at it is that
2π 1/2
(1.46) f = 2
f (x) dx
0
is a norm on the vector space C 1 , but C 1 (or even C) is not complete with
respect to this norm. What does one do in the situation when one has a
normed vector space V which is not complete? In general one can complete
the space by inventing fictitious or “ideal” elements and adding them to
the space. This may be done as follows. Consider any Cauchy sequence of
elements of V. If it has a limit in V, fine. Otherwise, we plug the “hole” by
20 1. BASIC NOTIONS
inventing a limit for it. One has to check that the resulting enlarged space
satisfies all of the stipulations of a Banach space.
This can be done as follows: Let Ṽ denote the set of all Cauchy sequences
in V. If
G = {gk }, H = {hk }
are members of Ṽ , we define
G + H = {gk + hk }, αG = {αgk }, G = lim gk .
k→∞
We consider a Cauchy sequence equal to 0 if it converges to 0. With these
definitions, Ṽ becomes a normed vector space. Moreover, it is complete. To
see this, let Gn = ({gnk }) be a Cauchy sequence in Ṽ . Then for each n, there
is an Nn such that
1
gnk − gnl ≤ , k, l ≥ Nn .
n
Pick an element gnl with l ≥ Nn and call it fn . Then
1
fn − gnk ≤ , k ≥ Nn .
n
Let Fn be the element in Ṽ given by the constant sequence (fn , fn , · · · ). (It
qualifies as a Cauchy sequence.) Then
1
Gn − Fn ≤ .
n
Finally, let F = {fn }. Then
Gn − F ≤ Gn − Fn + Fn − F → 0 as n → ∞.
We note that in order to be logically precise, we had to consider equiv-
alence classes of Cauchy sequences. If this sounds artificial, just remember
that it is precisely the way we obtain the real numbers from the rationals.
In Section 8.6 we shall give an “easy” proof of the fact that we can always
complete a normed vector space to make it into a Banach space (without
considering equivalence classes of Cauchy sequences).
In our present case, however, it turns out that we do not have to invent
ideal elements. In fact, for each sequence satisfying (1.45) there is a bona
fide, function f (x) such that
2π
(1.47) [fj (x) − f (x)]2 dx −→ 0 as j −→ ∞.
0
It may be that this function is very discontinuous, but its square is integrable
in a generalized sense. No claim concerning pointwise convergence of the
sequence {fj } is intended, but just that (1.47) holds. To summarize: the
completion of C 1 with respect to the norm (1.46) consists of those functions
having squares integrable in a generalized sense. We denote this space by
L2 = L2 ([0, 2π]). (When no confusion will result, we do not indicate the

region over which the integration is taken. At the present time we are
integrating over the interval [0, 2π]. At other times we shall use different
regions.)
Now suppose (α0 , α1 , · · · ) is a sequence in l2 . Set

n
(1.48) fn (x) = αj ϕj (x).
0
Then f ∈ C 1 , and for m < n

n
n
(1.49) fn − fm 2 = αj ϕj 2 = αj2 −→ 0 as m, n −→ ∞.
m+1 m+1
Thus, there is an f ∈ L2 such that

fn − f −→ 0 as n −→ ∞.
By (1.42), we see for m < n,
2π
αm = fn (x)ϕm (x) dx.
0
We claim that
2π
(1.50) αm = f (x)ϕm (x) dx.
0
This follows from the fact that
2π
(f, g) = f (x)g(x) dx
0
is a scalar product corresponding to (1.46). Hence,
|(fn , ϕm ) − (f, ϕm )| = |(fn − f, ϕm )| ≤ fn − f · ϕm −→ 0 as n −→ ∞.
But αm does not depend on n. Therefore, (1.50) holds and

n
n
αj2 = αj (f, ϕj ) = (f, fn ).
0 0
Letting n → ∞, we get
∞

(1.51) αj2 = f 2 .
0
Conversely, let f be any function in L2 . Then there is a sequence {fn } of

elements in C 1 which converges to f in L2 . Now each fn can be expanded
in a Fourier series. Thus,
∞
(n)
(1.52) fn = αj ϕj .
j=0
22 1. BASIC NOTIONS
Since it is a Cauchy sequence, we have

∞
(n) (m) 2
(1.53) fn − fm = 2
(αj − αj ) −→ 0 as m, n −→ ∞.
j=0
Since l2 is complete, there is a sequence (α0 , α1 , · · · ) in l2 such that

∞
(n)
(1.54) (αj − αj )2 −→ 0 as n −→ ∞.
j=0
Set
∞

(1.55) f˜ = αj ϕj .
0
By what we have just shown, f˜ ∈ L2 . Moreover

∞
(n)
fn − f˜2 = (αj − αj )2 −→ 0 as n −→ ∞.
j=0
Hence, f˜ = f. Thus, we have

Theorem 1.4. There is a one-to-one correspondence between l2 and L2
such that if (α0 , α1 , · · · ) corresponds to f, then

n
(1.56) αj ϕj − f −→ 0 as n −→ ∞,
0
and
∞

(1.57) f 2 = αj2 , αj = (f, ϕj ).
0
Note that this correspondence is linear and one-to-one in both directions.

We shall have more to say about this in Chapter 3. As usual, we are not sat-
isfied with merely proving statements about L2 . We want to know if similar
statements hold in other Hilbert spaces. So we examine the assumptions we
have made. One property of the sequence {ϕn } is

0, m = n,
(1.58) (ϕm , ϕn ) =
1, m = n.
Such a sequence is called orthonormal. So suppose we have an arbitrary
Hilbert space H and an orthonormal sequence {ϕj } of elements in H. If f
is an arbitrary element of H, set αn = (f, ϕn ). Then

n
n
n
n
(1.59) f − αi ϕi = f − 2
2 2
αi (f, ϕi ) + αi2 = f −
2
αi2 .
1 1 1 1
Hence,

n
αi2 ≤ f 2 .
1
Letting n → ∞, we have
∞

(1.60) αi2 ≤ f 2 .
1
Equation (1.59) is called Bessel’s identity, while (1.60) is called Bessel’s

inequality. We also have
Theorem 1.5. Let (α1 , α2 , · · · ) be a sequence of real numbers, and let {ϕn }
be an orthonormal sequence in H. Then
n
αi ϕi
1
converges in H as n → ∞ if, and only if,
∞
αi2 < ∞.
1
Proof. For m < n,

2
n
n

αi ϕi = αi2 .
m m

An orthonormal sequence {ϕj } in a Hilbert space H is called complete if

sums of the form
n
(1.61) S= αi ϕi
1
are dense in H; i.e., if for each f ∈ H and ε > 0, there is a sum S of this
form such that f − S < ε. We have
Theorem 1.6. If {ϕn } is complete, then for each f ∈ H
∞

f= (f, ϕi )ϕi
1
and
∞

(1.62) f =
2
(f, ϕi )2 .
1
24 1. BASIC NOTIONS
Proof. Let f be any element of H, and let

(n)
fn = αj ϕj
be a sequence of sums of the form (1.61) which converges to f in H. In

particular, (1.53) holds. Thus, there is a sequence (α0 , α1 , · · · ) in l2 such
that (1.54) holds. If we define f˜ by (1.55), we know that f˜ ∈ H (Theorem
1.5) and fn → f˜ as n → ∞. Hence, f = f˜. This proves the first statement.
To prove (1.62), we have by Bessel’s identity (1.59),

n
n
f − αi ϕi = f −
2 2
αi2 .
1 1
Letting n → ∞, we obtain (1.62).
Equation (1.62) is known as Parseval’s equality.
Theorem 1.6 has a trivial converse. If (1.62) holds for all f in a Hilbert
space H, then the orthonormal sequence {ϕn } is complete. This follows
immediately from (1.59). We also have
Theorem 1.7. If {ϕn } is complete, then

∞

(f, g) = (f, ϕn )(g, ϕn ).
1
Proof. Set αj = (f, ϕj ). Since

∞

f= αj ϕj ,
1
we have

n
n
(f, g) = lim ( αj ϕj , g) = lim αj (ϕj , g).
n→∞ n→∞
1 1

1.5. Problems
(1) Show that statement (15) of Section 1.2 is implied by statements

(1)–(13) of that section, but not by statements (1)–(9).
1.5. Problems 25
(2) Let ϕ1 , · · · , ϕn be an orthonormal set in a Hilbert space H. Show

that
n
n
f − αk ϕk ≥ f − (f, ϕk )ϕk
1 1
for all f ∈ H and all scalars αk .
(3) Show that an orthonormal sequence {ϕk } is complete if, and only
if, 0 is the only element orthogonal to all of them.
(4) Let c denote the set of all elements (α1 , · · · ) ∈ l∞ such that {αn } is
a convergent sequence, and let c0 be the set of all such elements for
which αn → 0 as n → ∞. Show that c and c0 are Banach spaces.
(5) Show that the operator given at the end of Section 1.1 satisfies the
hypotheses of Theorem 1.1.
(6) Carry out the details of completing a normed vector space by the
method described in Section 1.4.
(7) Prove Theorem 1.4 with L2 replaced by any Hilbert space with a
complete orthonormal sequence.
(8) Show that the norm of an element is never negative.
(9) If x, y are elements of a Hilbert space and satisfy x + y = x +

y, show that either x = cy or y = cx for some scalar c ≥ 0.
(10) If xk → x, yk → y in a Hilbert space, show that (xk , yk ) → (x, y).
(11) If x1 , · · · , xn are elements of a Hilbert space, show that

det[(xj , xk )] ≥ 0.
(12) Show that in a normed vector space

∞

n ∞

vj = vj + vj .
j=1 j=1 j=n+1
26 1. BASIC NOTIONS
(13) Let H be a Hilbert space, and let {ϕν } be any collection of or-
thonormal elements in H (not necessary denumerable). If x ∈ H,
show that there is at most a denumerable number of the ϕν such
that (x, ϕν ) = 0.
(14) Show that B[a, b] is a normed vector space.
(15) For f = (α1 , · · · , αn ) ∈ Rn , show that
f 0 = max |αk |
k
and

n
f 1 = |αk |
k=1
are norms. Is Rn complete with respect to either of them?
(16) If {ϕn } is an orthonormal sequence in a Hilbert space H, and

∞

f =2
(f, ϕn )2
1
holds for each f ∈ H, show that {ϕn } is complete.
(17) Prove: If {ϕn } is an orthonormal sequence in a Hilbert space H

and (α1 , · · · ) ∈ l2 , then there is an f ∈ H such that
∞

αk = (f, ϕk ), k = 1, · · · , and f = 2
αk2 .
k=1
(18) Calculate
2

n
n
n
a2k b2k − ak bk
1 1 1
and use the answer to prove the Cauchy-Schwarz inequality (1.26).
(19) Show that cf ∞ = |c| · f ∞ for f ∈ l∞ .
(20) Show that the shortest distance between two points in a Hilbert
space is along a straight line. Is this true in all Banach spaces?
1.5. Problems 27
(21) Define
f ∗ = inf{t > 0 : f < t}, f ∈ Rn .
Is this a norm on Rn ?
(22) Does there exist a norm · on R2 such that the set

B = {x ∈ R2 : x < 1}
is a rectangle? an ellipse?
(23) Show that (1.29) can be solved in Rn for arbitrary b if

1
|ajk | < , 1 ≤ j, k ≤ n.
n
(24) Show that (1.32) can be solved in l∞ for arbitrary b when

1
|ajk | ≤ 2 , j, k = 1, 2, · · · .
2k
(25) Show that (1.32) can be solved in l2 for arbitrary b if

1
|ajk | ≤ , j, k = 1, 2, · · · .
2jk
Another random document with
no related content on Scribd:
chorus,” on any account, be “the shouting of the morning stars.”
Rather begin, as you propose to end, with “silence,” than in this
melancholy way. Let your thoughts be based on the unalterable
emotions of the heart, not on the wild driftings of the fancy. Observe
all that strongly appeals to the feelings of others and of yourself—let
art assist you to select and to combine—your warm imagination will
give life to the conception, and your powers of fancy and language
will vividly express it. Don’t set down any odd conceit that may strike
you about the relation of the sea and the stars, and the moon; but
when you conceive an image which, besides being fine in itself, shall
bear essential, not accidental, relation to some part of your theme,
put it by till your main subject, in its natural expansion, affords it a
fitting place.
Following this course, we trust that Alexander will prove worthy of
the many illustrious scions of the house of Smith who have
distinguished themselves since Adam, and maintain its precedence
over the houses of Brown, Jones, and Robinson. Sydney the
Reverend—Horace and James of the Rejected Addresses—and
William, of the modest and too obscure dramas (noticed by us
before), might well become prouder of the patronymic to which they
have already lent lustre, when Alexander, mellowed by time, and
taught by thought and experience, shall have produced his next and
riper work.
THE EPIDEMICS OF THE MIDDLE AGES.[8]
This extremely interesting work of Dr Hecker’s consists of three

several treatises, or historical sketches, published at different times,
and here collected in a single volume. They are translated and
published under the direction of the Sydenham Society—a society
which has been the means of introducing to the medical profession,
and to the English reader, some of the most eminent works of
German physicians and physiologists. It is seldom, indeed, that their
publications are of the popular and amusing description of the one
we have selected for notice; but, speaking of them as a series, they
are of that high philosophic character which must render them
acceptable to every man of liberal education. How far they are
accessible to the public at large we have not the means of knowing,
nor whether the purchase of any single volume is a practicable
matter to a non-subscriber; but, at all events, means, we think, ought
to be taken to place the whole series on the shelves of every public
library.
The great plague of the fourteenth century, called in Germany The
Black Death, from the dark spots of fatal omen which appeared on
the bodies of its victims; the Dancing Mania, which afterwards broke
out both in Germany and Italy; and the Sweating Sickness, which
had its origin in England, but extended itself also widely upon the
Continent—these form the three subjects of Dr Hecker’s book. The
dancing mania, known in Germany as St John’s or St Vitus’s Dance,
and in Italy as the poison of the Tarantula or Tarantism, will be most
likely to present us with novel and curious facts, and we shall be
tempted to linger longest upon this topic. Readers of all kinds,
whether of Thucydides, or Boccaccio, or Defoe, are familiar with the
phenomena and events which characterise a plague, and which bear
a great resemblance to each other in all periods of history. We shall,
therefore, refrain from dwelling at any length upon the well-known
terrors of the Great Mortality or the Black Death.
Yet the subject is one of undying interest. The Great Plague is, in
this respect, like the Great Revolution of France; you may read fifty
histories of it, and pronounce it to be a topic thoroughly worn out
and exhausted; and yet when the fifty-first history is put into your
hands, the chance is that you will be led on, and will read to the very
last page with almost undiminished interest. The charm is alike in
both cases. It is that our humanity is seen in its moments of great, if
not glorious excitement—of plenary inspiration of some kind,
though it be of an evil spirit—seen in moments when all its passions,
good and bad, and the bad chiefly, stand out revealed in full
unfettered strength. And the history, in both cases, is of perpetual
value and significance to us. Plagues, as our own generation can
testify, are no more eradicated or banished from the cities of
mankind than political revolutions. They read a lesson to us which,
terrible as it is, we are still slow in learning.
We are often haunted with the dread of over-population. This fear
may perhaps be encountered by another of a quite opposite
description, when we read that in the fourteenth century one quarter
at least of the population of the Old World was swept away in the
short space of four years! Such is the calculation which Dr Hecker
makes, on the best sources of information within his reach. If such
devastating plagues arise, as our author thinks, from great physical
causes over which man has no control, from an atmospheric poison
not traceable to his ignorance or vice, and which no advancement in
science can prevent or expel, there is indeed room for an undefined
dread of periodical depopulations, putting to the rout all human
calculations and all human forethought. But on this point we have
our doubts.
“An inquiry into the causes of the Black Death,” says our author,
“will not be without important results in the study of the plagues
which have visited the world, although it cannot advance beyond
generalisation without entering upon a field hitherto uncultivated,
and, to this hour, entirely unknown. Mighty revolutions in the
organism of the earth, of which we have credible information, had
preceded it. From China to the Atlantic the foundations of the earth
were shaken—throughout Asia and Europe the atmosphere was in
commotion, and endangered, by its baneful influence, both vegetable
and animal life.” When, however, Dr Hecker proceeds to specify the
earthquakes and volcanic eruptions, and other terrific events which
shook the foundations of the earth from China to the Atlantic, we do
not find that the enumeration at all bears out this general
description. A large proportion of such disastrous phenomena as he
has been able to collect relate to China; and although the plague
should be proved to have travelled from the East, it is not traced, as
an identical disease, so far eastward as to China, and therefore is but
vaguely connected with the great droughts and violent rains which
afflicted that region of the earth. Nearer at home, in Europe, we have
mention made of “frequent thunderstorms,” and an eruption of
Ætna, but thunderstorms and a volcanic eruption have not, on other
occasions, given rise to a plague; not to add, that if the atmosphere of
Europe was tainted from causes of this kind, springing from its own
soil and its own climate, it would be quite superfluous to trace the
disease to the East at all. We should merely say that a similar disease
broke out in different countries at the same time, demonstrating
some quite cosmical or universal cause. The most important fact
which is mentioned here, as proving some wide atmospheric
derangement, is the “thick stinking mist seen to advance from the
East and spread itself over Italy.” But Dr Hecker himself adds, that at
such a time natural occurrences would be transformed or
exaggerated into miracles; and we are quite sure that any really
extraordinary event, occurring simultaneously with the plague,
would, without further inquiry, be described as the cause of it. An
unusual mist, just as a comet or any unusual meteor, appearing at
the time, would be charged with the calamity.
On so obscure a subject we have no desire to advance any
dogmatic opinion. There are facts connected with this and other
great epidemics which, to men of cautious research, have seemed to
point to some widespreading poison, some subtle, deleterious matter
diffused through the air, or some abnormal condition of the
atmosphere itself. Such there may be, acting either as immediate or
predisposing cause of the disease. But to our apprehension, all
plagues and pestilences have been bred from two well-known and
sufficient causes—famine and filth. Scanty and unwholesome diet
first disorders and debilitates the frame, fevers ensue, the foul
atmosphere of crowded unventilated dwellings becomes
impregnated by breathings that have passed through putrid lungs;
and thus the disease, especially in a hot climate, attains to that
malignity that the stricken wretch, move him where you will,
becomes the centre of infection to all around him, and from his
pestiferous dwelling there creeps a poison which invades even the
most salubrious portion of the town; which, stealing through the
garden-gate and over the flower-beds, enters even into the very
palace itself. Doubtless other causes may co-operate, as unusual
rains and fogs; the fact that a murrain amongst cattle sometimes
accompanies or precedes a plague, indicates local causes of this
description; but the true source of the disease lies in the city man has
built, in his improvidence or injustice, his ignorance or his sloth.
It is thus that Dr Hecker speaks of the manner in which the disease
may be propagated, so far as the agency of man is concerned:—we do
not seem to want any quite cosmical influence.
“Thus much from authentic sources of the nature of the Black Death. The
descriptions which have been communicated contain, with a few unimportant
exceptions, all the symptoms of the Oriental plague, which have been observed in
more modern times. No doubt can obtain on this point. The facts are placed clearly
before our eyes. We must, however, bear in mind that this violent disease does not
always appear in the same form; and that, while the essence of the poison which it
produces, and which is separated so abundantly from the body of the patient,
remains unchanged, it is proteoform in its varieties, from the almost imperceptible
vesicle, unaccompanied by fever, which exists for some time before it extends its
poison inwardly, and then excites fevers and buboes, to the fatal form in which
carbuncular inflammations fall upon the most important viscera.
“Such was the form which the plague assumed in the fourteenth century, for the
accompanying chest affection, which appeared in all the countries whereof we have
received any account, cannot, on a comparison with similar and familiar
symptoms, be considered as any other than the inflammation in the lungs of
modern medicine, a disease which at present only appears sporadically, and owing
to a putrid decomposition of the fluids is probably combined with hemorrhages
from the vessels of the lungs. Now as every carbuncle, whether it be cutaneous or
internal, generates in abundance the matter of contagion which has given rise to it,
so therefore must the breaths of the affected have been poisonous in this plague,
and on this account its power of contagion wonderfully increased; wherefore the
opinion appears incontrovertible that, owing to the accumulated numbers of the
diseased, not only individual chambers and houses, but whole cities, were infected;
which, moreover, in the middle ages, were, with few exceptions, narrowly built,
kept in a filthy state, and surrounded with stagnant ditches. Flight was in
consequence of no avail to the timid; for some, though they had sedulously avoided
all communication with the diseased and the suspected, yet their clothes were
saturated with the pestifierous atmosphere, and every inspiration imparted to
them the seeds of the destructive malady which, in the greater number of cases,
germinated with but too much fertility. Add to which the usual propagation of the
plague through clothes, beds, and a thousand other things to which the pestilential
poison adheres,—a propagation which, from want of caution, must have been
infinitely multiplied; and since articles of this kind, removed from the access of air,
not only retain the matter of contagion for an indefinite period, but also increase
its activity, and engender it like a living being, frightful ill consequences followed
for many years after the first fury of the pestilence was passed.”
It may be worth noticing that Dr Hecker, or his translator, uses the

terms contagion and infection indiscriminately; nor is the question
entered into whether the disease is capable of being propagated by
mere contact, without inhaling the morbific matter, or becoming
inoculated with it through some puncture in the skin. Dr Hecker
nowhere gives countenance to such a supposition. The poison would
hardly penetrate by mere touch through a sound and healthy skin.
Such a belief, however, was likely enough to prevail at a time when
we are told that “even the eyes of the patient were considered as
sources of contagion, which had the power of acting at a distance,
whether on account of their unwonted lustre or the distortion which
they always suffer in plague, or whether in conformity with an
ancient notion, according to which the sight was considered as the
bearer of a demoniacal enchantment.”
Avignon is here mentioned as the first city in which the plague
broke out in Europe. We have a report of it from a contemporary
physician, Guy de Chauliac, a courageous man, it seems, who
“vindicated the honour of medicine by bidding defiance to danger,
boldly and constantly assisting the affected, and disdaining the
excuse of his colleagues, who held the Arabian notion, that medical
aid was unavailing, and that the contagion justified flight.” The
plague appeared twice in Avignon, first in the year 1348, and twelve
years later, in 1360, “when it returned from Germany.” On the first
occasion it raged chiefly amongst the poor; on the second more
amongst the higher classes, destroying a great many children, whom
it had formerly spared, and but few women. We presume that on the
second occasion the plague was re-introduced at once amongst the
merchant class of the city, and this would account for fewer women
falling victims to it, because men of this class could take precautions
for the safety of their wives and daughters. But why a greater number
of children should have died, when the women were comparatively
spared, is what we will make no attempt to explain.
How fatal it proved at Florence, Boccaccio has recorded. It is from
him we learn with certainty that other animals besides man were
capable of being infected by the disease—a fact of no little interest in
the history of the plague. He mentions that he himself saw two hogs,
on the rags of a person who had died of plague, after staggering
about for a short time, fall down dead as if they had taken poison. A
multitude of dogs, cats, fowls, and other domesticated animals, were,
he tells us, fellow-sufferers with man.
In Germany the mortality was not so great as in Italy, but the
disease assumed the same character. In France, it is said, many were
struck as if by lightning, and died on the spot—and this more
frequently among the young and strong than the old. Throughout
England the disease spread with great rapidity, men dying in some
cases immediately, in others within twelve hours, or at latest in two
days. Here, as elsewhere, the inflammatory boils and buboes were
recognised at once as prognosticating a fatal issue. It first broke out
in the county of Dorset. Few places seem to have escaped; and the
mortality was so great that contemporary annalists have reported
(with what degree of accuracy we cannot say) that throughout the
whole land not more than a tenth part of the inhabitants had
survived.
The north of Europe did not escape, nor did all the snows of Russia
protect her from this invasion. In Norway the disease broke out in a
frightful manner. Nor was the sea a refuge; sailors found no safety in
their ships; vessels were seen driving about on the ocean and drifting
on the shore, whose crews had perished to the last man.
It is a terrible history, this of a plague. Nevertheless, if we were
capable of surveying such events from an elevated position, where
past and future were revealed to our view, and the whole scheme of
creation unfolded to our knowledge, we should doubtless discover
that even plagues and pestilences play their parts for the welfare and
advancement of the human race. Nor are we without some glimpses
of their utility. Viewing the matter, in the first place, in a quite
physiological light, let us suppose that disease has been generated in
a great city, that debilitated parents give birth to feeble offspring,
that the fever, or whatever it may be, is wasting the strength of whole
classes of the population, is it not better that such disease should
attain a power and virulence that will enable it to sweep off at once a
whole infected generation, men, women, and children, leaving the
population to be replaced by the healthier who would survive? would
not this be better than to allow the disease to perpetuate itself
indefinitely, and thus to continue to multiply from an infected stock?
The poison passes on, and searches out other neighbourhoods where
the like terrible remedy is needed. Ay, but it passes, you say, into
cities and districts where no such curative process, no such
restoration of the breed, was called for. But it is always thus with the
great laws of nature, or of Providence. Thus far, and no farther! is
said to the pestilence as well as to the ocean; but the line along the
beach is not kept or measured with that petty precision which a land-
surveyor would assuredly have suggested. Man’s greatness arises in
part from this struggle with an external nature, which threatens from
time to time to overwhelm him. There is, according to his
measurement of things, a dreadful surplus of power and activity,
both in the organic and the inorganic world. Nowhere are the forces
of nature exactly graduated to suit his taste or convenience. Happily
not. Man would sink into the tameness and insipidity of an Arcadian
shepherd, or the sheep he feeds and fondles, if every wind that blew
were exactly tempered to his own susceptibility.
But the moral effects of plague and pestilence—what good thing
can be said of them? A general dissoluteness, an unblushing villany,
for the most part prevails: a few instances of heroic virtue brighten
out above the corrupted mass. Well, is it nothing, then, that from
time to time our nature should be fully revealed to us in its utmost
strength for good or for evil? A very hideous revelation it may
sometimes be, but not the less salutary on this account. The mask of
hypocrisy is torn off a whole city; in one moment is revealed to a
whole people what its morality, what its piety is worth. Of the island
of Cyprus, we are told, that an earthquake shook its foundations, and
was accompanied by so frightful a hurricane that the inhabitants,
who had slain their Mahometan slaves in order that they might not
themselves be subjected by them, fled in dismay in all directions.
Who had slain their Mahometan slaves! Their Christianity had
brought them thus far on the road of moral culture! At Lübeck, the
Venice of the North, the wealthy merchants were not, in this
extremity, unmindful of the safety of their souls; they spent their last
strength in carrying their treasures to monasteries and churches.
Useless for all other purposes, their gold would now purchase
heaven. To such intelligent views of Christianity had they attained!
But the treasure had no longer any charm for the monks; it might be
infected; and even with them the thirst for gold was in abeyance.
They shut their gates upon it; yet still it was cast to them over the
convent walls. “People would not brook an impediment to the last
pious work to which they were driven by despair.”
Did all desert their post, or belie their professions? No; far from it.
Amongst other instances, take that of the Sisters of Charity at the
Hotel Dieu. “Though they lost their lives evidently from contagion,
and their numbers were several times renewed, there was still no
want of fresh candidates, who, strangers to the unchristian fear of
death, piously devoted themselves to their holy calling.”
But how cruel had their fears made the base multitude of
Christendom! They rose against the Jews. They sought an enemy.
The wells were poisoned; the Jews had poisoned them. Sordid
natures invariably strive to lose the sense of their own calamity in a
vindictive passion against some supposed author of it. For this
reason it is, that, whatever the nature of the public distress may be,
they always fasten it upon some human antagonist, whom they can
have the luxury of hating and reviling. If they cannot cure, they can
at least revenge themselves.
“The noble and the mean fearlessly bound themselves by an oath to extirpate the
Jews by fire and sword, and to snatch them from their protectors, of whom the
number was so small, that throughout all Germany but few places can be
mentioned where these unfortunate people were not regarded as outlaws, and
martyred and burnt. Solemn summonses were issued from Berne, to the towns of
Basle, Freyburg, and Strasburg, to pursue the Jews as prisoners. The burgomasters
and senators, indeed, opposed this requisition; but in Basle the populace obliged
them to bind themselves by an oath to burn the Jews, and to forbid persons of that
community from entering their city for the space of two hundred years. Upon this
all the Jews in Basle, whose number could not be inconsiderable, were enclosed in
a wooden building, constructed for the purpose, and burnt together with it, upon
the mere outcry of the people, without sentence or trial, which indeed would have
availed them nothing. Soon after the same thing took place at Freyburg. A regular
diet was held at Bennefeeld, in Alsace, where the bishops, lords, and barons, as
also deputies of the counties and towns, consulted how they should proceed with
regard to the Jews: and when the deputies of Strasburg—not, indeed, the bishop of
this town, who proved himself a violent fanatic—spoke in favour of the persecuted,
as nothing criminal was substantiated against them, a great outcry was raised, and
it was vehemently asked why, if so, they had covered their wells and removed their
buckets?” [The wells were not used in the mere suspicion that they were poisoned,
and then the covering of them up became a proof with these reasoners that they
had been poisoned]. “A sanguinary decree was resolved upon, of which the
populace, who obeyed here the call of the nobles and superior clergy, became but
the too willing executioners. Wherever the Jews were not burnt they were at least
banished, and so being compelled to wander about, they fell into the hands of the
country people, who without humanity, and regardless of all laws, persecuted them
with fire and sword. At Spires the Jews, driven to despair, assembled in their own
habitations, which they set on fire, and thus consumed themselves with their
families.”
The atrocities, in short, that were committed against this unhappy

people were innumerable. At Strasburg 2000 men were burnt in
their own burial-ground. At Mayence, 12,000 are said to have been
put to a cruel death. At Eslingen the whole Jewish community
burned themselves in their own synagogue. Those whom the
Christians saved they insisted upon baptising! And, as fanaticism
begets fanaticism, Jewish mothers were seen throwing their children
on the pile, to prevent their being baptised, and then precipitating
themselves into the flames. From many of the accused the rack
extorted a confession of guilt; and as some Christians also were
sentenced to death for poisoning the wells, Dr Hecker suggests that it
is not improbable the very belief in the prevalence of the crime had
induced some men of morbid imagination really to commit it. When
a faith in witchcraft, he observes, was prevalent, many an old woman
was tempted to mutter spells against her neighbour. The false
accusation had ended in producing, if not the crime itself, yet the
criminal intention.
When we remember what took place in England under the reign of
one Titus Oates, we shall not conclude that these terrible
hallucinations of the public mind are proofs of any very peculiar
condition of barbarism. Then, as at the later epoch to which we have
alluded, a very marvellous plot was devised and thoroughly credited.
All the Jews throughout Christendom were under the control and
government of certain superiors at Toledo—a secret and mysterious
council of Rabbis—from whom they received their commands. These
prepared the poison with their own hands, from spiders, owls, and
other venomous animals, and distributed it in little bags, with
injunctions where it was to be thrown. Dr Hecker gives us, in an
appendix, an official account of the “Confessions made on the 15th
September, in the year of our Lord 1348, in the castle of Chillon, by
the Jews arrested in Neustadt on the charge of poisoning the wells,
springs, and other places, also food, &c., with the design of
destroying and extirpating all Christians.” These confessions were, of
course, produced by the rack, or by the threat of torture, and the
manifest inutility of any defence or denial. Nor must it be forgotten,
that the official report was drawn up after the whole of the Jews at
Neustadt had been burnt on this very charge. Amongst these
confessions is one of Balaviginus, a Jewish physician, arrested at
Chillon “in consequence of being found in the neighbourhood.” He
was put for a short time upon the rack, and, after being taken down,
“confessed, after much hesitation, that, about ten weeks before, the
Rabbi Jacob of Toledo sent him, by a Jewish boy, some poison in the
mummy of an egg: it was a powder sewed up in a thin leathern
pouch, accompanied by a letter, commanding him, on penalty of
excommunication, and by his required obedience to the law, to throw
the poison into the larger and more frequented wells of Thonon.”
Similar letters had been sent to other Jews. All Jews, indeed, were
under the necessity of obeying these injunctions. He, Balaviginus,
had done so; he had thrown the poison into several wells. It was a
powder half red and half black. Red and black spots were produced
by the plague; it was right that this poison should partake of these
two colours.
Conveyed over the lake from Chillon to Clarens to point out the
well into which he had thrown the powder, Balaviginus, “on being
conducted to the spot, and having seen the well, acknowledged that
to be the place, saying, ‘This is the well into which I put the poison.’
The well was examined in his presence, and the linen cloth in which
the poison had been wrapped was found. He acknowledged this to be
the linen which had contained the poison; he described it as being of
two colours—red and black.” We follow in imagination this Jewish
physician. Taken from the rack to his cell, he repeats whatever
absurdity his unrelenting persecutors put into his mouth. Rabbi
Jacob of Toledo—mummy of an egg—what you will. Conducted to the
well—yes, this was the well; shown the very rag—yes, this was the
rag;—and the powder? yes, it was red and black. What scorn and
bitterness must have mingled with the agony of the Jewish
physician!
Amidst all this we hear the scourge and miserable chant of the
Flagellants, stirring up the people to fresh persecutions, and
infecting their minds with a superstition as terrible as the vice it
pretended to expiate. This was not, indeed, their first appearance in
Europe; nor did the Flagellants do more, at the commencement, than
exaggerate the sort of piety their own church had taught them.
Happily, as their fanaticism rose, they put themselves in opposition
to the hierarchy, and were thus the sooner dispersed. In their
spiritual exultation they presumed to reform or to dispense with the
priesthood. They found themselves, therefore, in their turn subjected
to grave denunciations, and pronounced to be one cause of the wrath
of Heaven.
All this time what were the physicians doing? In the history of the
plague, written by a physician, the topic, we may be sure, is not
forgotten. But the information we glean is of a very scanty,
unsatisfactory character. As to the origin of the plague—“A grand
conjunction of the three superior planets, Saturn, Jupiter, and Mars,
in the sign of Aquarius, which took place, according to Guy de
Chauliac, on the 24th March 1345, was generally received as its
principal cause. In fixing the day, this physician, who was deeply
versed in astrology, did not agree with others; wherefore there arose
various disputations of weight in that age, but of none in ours.” The
medical faculty of Paris pronounced the same opinion. Being
commissioned to report on the causes and the remedies of this Great
Mortality, they commence thus: “It is known that in India, and the
vicinity of the Great Sea, the constellations which emulated the rays
of the sun, and the warmth of the heavenly fire, exerted their power
especially against that sea, and struggled violently with its waters.”
Hence vapours and corrupted fogs; hence no wholesome rain, or
hail, or snow, or dew, could refresh the earth. But notwithstanding
this learning, quite peculiar to the age, they were not more at fault
than other learned bodies have been in later times, in the practical
remedies they suggested against the disease. They were not entirely
occupied in fixing the day when Jupiter, Mars, and Saturn, had
combated the sun over the great Indian Ocean. “They did,” as Dr
Hecker says, “what human intellect could do in the actual condition
of the healing art; and their knowledge of the disease was by no
means despicable.” When fevers have attained to that malignancy
that they take the name of plagues, they have escaped, we suspect,
from the control of the physician;—just as when fires take the name
of conflagrations, you must devote all your efforts to the saving of
what is yet unconsumed, and checking the extension of the flames.
Amongst the consequences of the plague, Dr Hecker notices that
the church acquired treasures and large properties in land, even to a
greater extent than after the Crusades; and that, on the subsidence of
the calamity, many entered the priesthood, or flocked to the
monasteries, who had no other motive than to participate in this
wealth. He adds, also, that,—
“After the cessation of the Black Plague, a greater fecundity in women was
everywhere remarkable—a grand phenomenon, which, from its occurrence after
every destructive pestilence, proves to conviction, if any occurrence can do so, the
prevalence of a higher power in the direction of general organic life. Marriages
were, almost without exception, prolific, and double and treble births were more
frequent than at other times; under which head we should remember the strange
remark, that after the ‘great mortality’ the children were said to have got fewer
teeth than before; at which contemporaries were mightily shocked, and even later
writers have felt surprise.
“If we examine the grounds of this oft-repeated assertion, we shall find that they
were astonished to see children cut twenty, or at most twenty-two teeth, under the
supposition that a greater number had formerly fallen to their share. Some writers
of authority, as, for example, the physician Savonarola, at Ferrara, who probably
looked for twenty-eight teeth in children, published their opinions on this subject.
Others copied from them without seeing for themselves, as often happens in other
matters which are equally evident; and thus the world believed in a miracle of an
imperfection in the human body, which had been caused by the Black Plague.”
That a fresh impetus would be given to population seems to us

quite sufficiently accounted for, without calling into aid any “higher
power in the direction of general organic life.” Men and women
would marry early; and the very fact of their having survived the
plague would, in general, prove that they were healthy subjects, or
had been well and temperately brought up. There would be the same
impetus to population that an extensive emigration would cause, and
an emigration that had carried away most of the sick and the feeble.
The belief that double and treble births were more frequent than at
other times, may perhaps be explained in the same manner as the
belief that there were fewer teeth than before in the human head. No
accurate observations had been at all made upon the subject.
We come next in order to The Dancing Mania—an epidemic of a
quite different character. Not, indeed, as the name might imply, that
the convulsive dance was a very slight affliction—it was felt to be
quite otherwise; but because it belongs to that class of nervous
maladies in which there is great room for mental or psychical
influence. Such disorders spring up in a certain condition of the
body, but the form they assume will depend on social circumstances,
or the ideas current at the time. And thus Dr Hecker finds no
difficulty in arranging the Convulsionnaires of France, or the early
Methodists of England and Wales, in the same category as the
maniacal dancers of Germany. It was in all the cases a physical
tendency of a similar character, brought out under the influence of
different ideas.
Dr Hecker mentions a case which, from the simplicity of the facts,
would form a good introduction to others of a more complicated
character. In the year 1787, at a cotton-manufactory at Hodden
Bridge, in Lancashire, a girl put a mouse into the bosom of another
girl, who had a great dread of mice. It threw her into a fit, and the fit
continued, with the most violent convulsions, for twenty-four hours.
On the following day three other girls were seized in the same way;
on the day after six more. A report was now spread that a strange
disease had been introduced into the factory by a bag of cotton
opened in the house. Others who had not even seen the infected, but
only heard of their convulsions, were seized with the same fits. In
three days, the number of the sufferers had reached to twenty-four.
The symptoms were, a sense of great anxiety, strangulation, and very
strong convulsions, which lasted from one to twenty-four hours, and
of so violent a nature that it required four or five persons to prevent
the patients from tearing their hair, and dashing their heads against
the floor and walls. Dr St Clare was sent for from Preston. Dr St Clare
deserves to have his name remembered. The ingenious man took
with him a portable electrical machine. The electric shock cured all
his patients without an exception. When this was known, and the
belief could no longer hold its ground that the plague had been
brought in by the cotton bag, no fresh person was affected.
If we substitute for the cotton bag a belief in some demoniacal
influence, compelling people to dance against their will, we have the
dancing mania of Germany. Unhappily there was no St Clare at hand,
with his electrical machine, to give a favourable shock to body and
mind at once, and thus disperse the malady before it gathered an
overpowering strength by the very numbers of the infected.
“The effects of the Black Death,” writes Dr Hecker (whose account of the
disorder we cannot do better than give, with some abridgments), “had not yet
subsided, when a strange delusion arose in Germany. It was a convulsion which in
the most extraordinary manner infuriated the human frame, and excited the
astonishment of contemporaries for more than two centuries, since which time it
has never reappeared. It was called the Dance of St John, or of St Vitus, on account
of the Bacchantic leaps by which it was characterised, and which gave to those
affected, whilst performing their wild dance, and screaming and foaming with fury,
all the appearance of persons possessed. It did not remain confined to particular
localities, but was propagated by the sight of the sufferers, like a demoniacal
epidemic, over the whole of Germany and the neighbouring countries to the north-
west, which were already prepared for its reception by the prevailing opinions of
the times.
“So early as the year 1374, assemblages of men and women were seen at Aix-la-
Chapelle, who had come out of Germany, and who, united by one common
delusion, exhibited to the public, both in the streets and in the churches, the
following strange spectacle. They formed circles hand in hand, and, appearing to
have lost all control over their senses, continued dancing, regardless of the
bystanders, for hours together, in wild delirium, until at length they fell to the
ground in a state of exhaustion. They then complained of extreme oppression, and
groaned as if in the agonies of death, until they were swathed in clothes, bound
tightly round their waists, upon which they again recovered, and remained free
from complaint until the next attack. This practice of swathing was resorted to on
account of the tympany which followed these spasmodic ravings; but the
bystanders frequently relieved patients in a less artificial manner, by thumping or
trampling upon the parts affected. While dancing, they neither saw nor heard,
being insensible to external impressions through the senses, but were haunted by
visions, their fancies conjuring up spirits, whose names they shrieked out; and
some of them afterwards asserted that they felt as if they had been immersed in a
stream of blood, which obliged them to leap so high. Others, during the paroxysm,
saw the heavens open, and the Saviour enthroned with the Virgin Mary, according
as the religious notions of the age were strangely and variously reflected in their
imaginations.”
The disease spread itself in two directions. It extended from Aix-

la-Chapelle through the towns of the Netherlands, and also through
the Rhenish towns. In Liege, Utrecht, and many other towns of
Belgium, the dancers appeared with garlands in their hair, and their
waists already girt with a cloth or bandage, that they might receive
immediate relief on the attack of the tympany. It seems that the
crowd around were often more ready to administer relief by kicks
and blows than by drawing this bandage tight. The most opposite
feelings seem to have been excited in the multitude by these
exhibitions. Sometimes an idle and vicious mob would take
advantage of them, and they became the occasion of much riot and
debauchery. More frequently, however, the demoniacal origin of the
disease, of which few men doubted, led to its being regarded with
astonishment and horror. Religious processions were instituted on
its account, masses and hymns were sung, and the whole power of
the priesthood was called in to exorcise the evil spirit. The malady
rose to its greatest height in some of the towns on the Rhine. At
Cologne the number of the possessed amounted to more than five
hundred, whilst at Metz the streets are said to have been filled
(numbering women and children together) with eleven hundred
dancers. Even those idle vagabonds who, for their own purposes,
imitated their convulsive movements, assisted to spread the
disorder; for in these maladies the susceptible are infected quite as
easily by the imitation as by the reality.
The physicians stood aloof. Acknowledged as a demoniacal
possession, they left the treatment of the disease entirely to the
priesthood; and their exorcisms were not without avail. But it was
necessary to this species of remedy that the patients should have
faith in the church and its holy ministers. Without faith there would
certainly, in such a case, be no cure; and, unhappily, the report had
been spread by some irreverend schismatics that the disorder itself
was owing—to what will our readers suppose?—to an imperfect
baptism—to the baptism of children by the hands of unchaste priests.
Where this notion prevailed, the exorcism, we need not say, was
unavailing.
The malady first bore the name of St John’s Dance, afterwards that
of St Vitus’s. This second name it took from the mere circumstance
that St Vitus was the saint appealed to for its cure. A legend had been
framed with a curious disregard—even for a legend—of all history
and chronology, in which St Vitus, who suffered martyrdom, as the
church records, under the Emperor Domitian, is described as
praying, just before he bent his neck to the sword, that he might
protect from the Dancing Mania all those who should solemnise the
day of his commemoration, and fast upon its eve. The prayer was
granted; a voice from heaven was heard saying, “Vitus, thy prayer is
accepted.” He became, of course, the patron saint of those afflicted
with the dancing plague. But the name under which it first appeared,
of St John’s Dance, receives from Dr Hecker an explanation which
points out to us a probable origin of the disease itself, or of the
peculiar form which it assumed.
“The connection,” he says, “which John the Baptist had with the dancing mania
of the fourteenth century, was of a totally different character. He was originally far
from being a protecting saint to those who were attacked, or one who would be
likely to give them relief from a malady considered as the work of the devil. On the
contrary, the manner in which he was worshipped afforded an important and very
evident cause for its development. From the remotest period, perhaps even so far
back as the fourth century, St John’s day was solemnised with all sorts of strange
and rude customs, of which the original mystical meaning was variously disfigured
among different nations by superadded relics of heathenism. Thus the Germans
transferred to the festival of St John’s day an ancient heathen usage—the kindling
of the ‘hodfyr,’ which was forbidden them by St Boniface; and the belief subsists
even to the present day, that people and animals that have leaped through these
flames, or their smoke, are protected for a whole year from fevers and other
diseases, as if by a kind of baptism by fire. Bacchanalian dances, which have
originated from similar causes among all the rude nations of the earth, and the
wild extravagancies of a heated imagination, were the constant accompaniments of
this half-heathen, half-christian festival.”
In a note at a subsequent page Dr Hecker cites some curious

passages to show what in the middle ages took place at “St John’s
fires.” Bones, horns, and other rubbish were heaped together to be
consumed in smoke, while persons of all ages danced round the
flames as if they had been possessed. Others seized burning
flambeaus, and made a circuit of the fields, in the supposition that
they thereby screened them from danger; while others again turned a
cartwheel, to represent the retrograde movement of the sun. The last
circumstance takes back the imagination to the old primitive worship
of the sun; and perhaps the very fires of St John might date their
history from those kindled in honour of Baal or Moloch. Dr Hecker
suggests that mingling with these heathen traditions or customs a
remembrance of the history of St John’s death—that dance which
occasioned his decapitation—might also have had its share in
determining the peculiar manner in which this saint’s day should be
observed. However that may be, as we find that the first dancers in
Aix-la-Chapelle appeared with St John’s name in their mouths, the
conjecture is very probable that the wild revels of St John’s day had
given rise, if not to the disease, yet to the type or form in which it
appeared.
At a subsequent period, indeed, when the disorder had assumed, if
we may so speak, a more settled aspect, the name of St John was no
otherwise associated with it than the name of St Vitus. People danced
upon his festival to obtain a cure. And these periodical dances, while
they relieved the patients, assisted also to perpetuate the malady.
Throughout the whole of June, we are told, prior to the festival of St
John, many men felt a disquietude and restlessness which they were
unable to overcome. They were dejected, timid, and anxious;
wandered about in an unsettled state, being tormented with
twitching pains, which seized them suddenly in different parts; they
eagerly expected the eve of St John’s day, in the confident hope that,
by dancing at the altars of this saint, they would be freed from all
their sufferings. Nor were they disappointed. By dancing and raving
for three hours to the utmost scope of their desires, they obtained
peace for the rest of the year. For a long time, however, we hear of
cases which assumed the most terrific form. Speaking of a period
which embraced the close of the fifteenth century, Dr Hecker says:—
“The St Vitus’s dance attacked people of all stations, especially those who led a
sedentary life, such as shoemakers and tailors; but even the most robust peasants
abandoned their labours in the fields, as if they were possessed by evil spirits; and
thus those affected were seen assembling indiscriminately, from time to time, at
certain appointed places, and, unless prevented by the lookers-on, continuing to
dance without intermission, until their very last breath was expended. Their fury
and extravagance of demeanour so completely deprived them of their senses, that
many of them dashed their brains out against the walls and corners of buildings, or
rushed headlong into rapid rivers, where they found a watery grave. Roaring and
foaming as they were, the bystanders could only succeed in restraining them by
placing benches and chairs in their way, so that, by the high leaps they were
tempted to take, their strength might be exhausted.”
Music, however, was a still better resource. It excited, but it

hastened forward the paroxysm, and doubtless reduced it to some
measure and rhythm. The magistrates even hired musicians for the
purpose of carrying the dancers the more rapidly through the attack,
and directed that athletic men should be sent among them, in order
to complete their exhaustion. A marvellous story is related on the
authority of one Felix Plater: Several powerful men being
commissioned to dance with a girl who had the dancing mania till
she had recovered from her disorder, they successively relieved each
other, and danced on for the space of four weeks! at the end of which
time the patient fell down exhausted, was carried to an hospital, and
there recovered. She had never once undressed, was entirely
regardless of the pain of her lacerated feet, and had merely sat down
occasionally to take some nourishment or to slumber, and even then
“the hopping movement of her body continued.”
Happily, however, this mania grew more rare every year, so that in
the beginning of the seventeenth century we may be said to be losing
sight of it in Germany. Nor shall we follow out its history further in
that country, because the same disorder, under a different form,
made its appearance in Italy, and we must by no means neglect to
notice the dancing mania which was so universally attributed to the
bite of the tarantula. Whatever part the festival of St John the Baptist
performed in Germany, as an exciter of the disease, that part was still
more clearly performed in Italy by the popular belief in the venom of
a spider.
We shall not go back with Dr Hecker into the fears or superstitions
of classical times as to the bite of certain spiders or lizards; we must
keep more strictly to our text; we must start from the period when
men’s minds were still open to pain and alarm on account of the
frequent return of the plague.
“The bite of venomous spiders, or rather the unreasonable fear of its
consequences, excited at such a juncture, though it could not have done so at an
earlier period, a violent nervous disorder, which, like St Vitus’s dance in Germany,
spread by sympathy, increasing in severity as it took a wider range, and still further
extending its ravages from its long continuance. Thus, from the middle of the
fourteenth century, the furies of The Dance brandished their scourge over afflicted
mortals; and music, for which the inhabitants of Italy now probably for the first
time manifested susceptibility and talent, became capable of exciting ecstatic
attacks in those affected, and thus furnished the magical means of exorcising their
melancholy.”
Does the learned doctor insinuate that the Italians owed their
natural taste for music to this invasion of Tarantism?
“At the close of the fifteenth century we find that Tarantism had spread beyond
the boundaries of Apulia, and that the fear of being bitten by venomous spiders
had increased. Nothing short of death itself was expected from the wound which
these insects inflicted; and if those who were bitten escaped with their lives, they
were said to be pining away in a desponding state of lassitude. Many became weak-
sighted or hard of hearing; some lost the power of speech; and all were insensible
to ordinary causes of excitement. Nothing but the flute or the cithern afforded
them relief. At the sound of these instruments they awoke as if by enchantment,
opened their eyes, and moving slowly at first, according to the measure of the
music, were, as the time quickened, gradually hurried on to the most passionate
dance. It was generally observable that country people, who were rude and
ignorant of music, evinced on these occasions an unusual degree of grace, as if
they had been well practised in elegant movements of the body; for it is a
peculiarity in nervous disorders of this kind that the organs of motion are in an
altered condition, and are completely under the control of the overstrained spirits.”
This increased agility and grace of movement is by no means to be

discredited by the reader. It is a symptom which distinguishes one
class of epileptic patients. Some have attributed it to an over-
excitement of the cerebellum. However that may be, there are greater
wonders than this contained in our most sober and trustworthy
books on the disorders of the nervous system. We continue the
account:—
“Cities and villages alike resounded throughout the summer season with the
notes of fifes, clarinets, and Turkish drums; and patients were everywhere to be
met with who looked to dancing as their only remedy. Alexander ab Alexandro,
who gives this account, saw a young man in a remote village who was seized with a
violent attack of Tarantism. He listened with eagerness and a fixed stare to the
sound of a drum, and his graceful movements gradually became more and more
violent, until his dancing was converted into a succession of frantic leaps, which
required the utmost exertion of his whole strength. In the midst of this
overstrained exertion of mind and body the music suddenly ceased, and he
immediately fell powerless to the ground, where he lay senseless and motionless
until its magical effect again aroused him to a renewal of his impassioned
performances.”
We have put the expression “mind and body” in italics, because we

may as well take this opportunity to observe, that although
convulsions of this kind are excited, and assume a certain form on
account of the predominance of some idea, yet, when once called
forth, they are almost entirely mechanical in their nature. Mere
animal excitability—what is called the reflex action, or other
automatic movements quite as little associated with the immediate
operations of “mind”—carry on the rest of the process. And it is some
consolation to think that the appearance of pain and distress which
marks convulsive disorders of all descriptions, is, for the most part,
illusory. The premonitory symptoms may be very distressing, but the
condition of the patient, when the fit is on, is that of insensibility to
pain.
The general conviction was, that by music and dancing the poison
of the tarantula was distributed over the whole body, and expelled
through the skin; but, unfortunately, it was also believed that if the
slightest vestige of it remained behind the disorder would break out
again. Thus there was no confidence excited in a perfect cure. Men
who had danced themselves well one summer watched the next
summer for the returning symptoms, and found in themselves what
they looked for. Thus—
“The number of those affected by it increased beyond belief, for whoever had
actually been, or even fancied that he had been once bitten by a poisonous spider
or scorpion, made his appearance annually whenever the merry notes of the
Tarantella resounded. Inquisitive females joined the throng and caught the disease
—not indeed from the poison of the spider, but from the mental poison which they
eagerly received through the eye; and thus the cure of the Tarantati gradually
became established as a regular festival of the populace.”
It was customary for whole bands of musicians to traverse Italy
during the summer months, and the cure of the disordered was
undertaken on a grand scale. This season of dancing and music was
called “The women’s little carnival,” for it was women more
especially who conducted the arrangements. It was they, too, it
seems, who paid the musicians their fee. The music itself received its
due share of study and attention. There were different kinds of the
Tarantella (as the curative melody was called) suited to every variety
of the ailment.
One very curious circumstance connected with this disease must
not pass unnoticed—the passion excited by certain colours. Amongst
the Germans, those afflicted by St Vitus’s dance were enraged by any
garment of the colour of red. Amongst the Italians, on the contrary,
red colours were generally liked. Some preferred one colour, some
another, but the devotion to the chosen colour was one of the most
extraordinary symptoms which the disease manifested in Italy. The
colour that pleased the patient he was enamoured of; the colour that
displeased excited his utmost fury.
“Some preferred yellow, others were enraptured with green; and eyewitnesses
describe this rage for colours as so extraordinary that they can scarcely find words
with which to express their astonishment. No sooner did the patients obtain a sight
of their favourite colour than they rushed like infuriated animals towards the
object, devoured it with their eager looks, kissed and caressed it in every possible
way, and, gradually resigning themselves to softer sensations, adopted the
languishing expression of enamoured lovers, and embraced the handkerchief, or
whatever article it might be which was presented to them, with the most intense
ardour, while the tears streamed from their eyes as if they were completely
overwhelmed by the inebriating impression on their senses.
“The dancing fits of a certain Capuchin friar in Tarentum excited so much
curiosity that Cardinal Cajetano proceeded to the monastery that he might see with
his own eyes what was going on. As soon as the monk, who was in the midst of his
dance, perceived the spiritual prince clothed in his red garments, he no longer
listened to the tarantella of the musicians, but with strange gestures endeavoured
to approach the cardinal, as if he wished to count the very threads of his scarlet
robe, and to allay his intense longing by its odour. The interference of the
spectators, and his own respect, prevented his touching it, and thus, the irritation
of his senses not being appeased, he fell into a state of such anguish and
disquietude that he presently sunk down in a swoon, from which he did not recover
until the cardinal compassionately gave him his cape. This he immediately seized
in the greatest ecstasy, and pressed, now to his breast, now to his forehead and
cheeks, and then again commenced his dance as if in the frenzy of a love fit.”
Another curious symptom, which was probably connected with

this passion for colour, was an ardent longing for the sea. These over-
susceptible people were attracted irresistibly to the boundless
expanse of the blue ocean, and lost themselves in its contemplation.
Some were carried so far by this vague passionate longing as to cast
themselves into the waves.
The persuasion of the inevitable and fatal consequences of being
bitten by the tarantula was so general that it exercised a dominion
over the strongest minds. Men who in their sober moments
considered the disorder as a species of nervous affection depending
on the imagination, were themselves brought under the influence of
this imagination, and suffered from the disorder at the approach of
the dreaded tarantula. A very striking anecdote of this kind is told of
the Bishop of Foligno. Quite sceptical as to the venom of the insect,
he allowed himself to be bitten by a tarantula. But he had not
measured the strength of his own imagination, however well he had
estimated the real malignancy of the spider. The bishop fell ill, nor
was there any cure for him but the music and the dance. Many
reverend old gentlemen, it is said, to whom this remedy appeared
highly derogatory, only exaggerated their symptoms by delaying to
have recourse to what, after all, was found to be the true and sole
specific.
But even popular errors are not eternal. This of Tarantism
continued, our author tells us, throughout the whole of the
seventeenth century, but gradually declined till it became limited to
single cases. “It may therefore be not unreasonably maintained,” he
concludes, “that the Tarantism of modern times bears nearly the
same relation to the original malady as the St Vitus’s dance which
still exists, and certainly has all along existed, bears, in certain cases,
to the original dancing mania of the dancers of St John.”
In a subsequent chapter, our author informs us that a disease of a
similar character existed in Abyssinia, or still exists, for the authority
he quotes is that of an English surgeon who resided nine years in
Abyssinia, from 1810 to the year 1819. We cannot pretend to say that
we have ever seen the book, which the learned German has, however,
not permitted to escape him—we have never seen the Life and
Adventures of Nathaniel Pearce, written by himself; but, judging by
the extract here given, Nathaniel Pearce must be a person worth
knowing, he writes with so much candour and simplicity. The disease
is called in Abyssinia the Tigretier, because it occurs most frequently
in the Tigrè country. The first remedy resorted to is the introduction
of a learned Dofter, “who reads the Gospel of St John, and drenches
the patient with cold water daily.” If this does not answer, then the
relations hire a band of trumpeters, drummers, and fifers, and buy a
quantity of liquor; all the young men and women of the place
assemble at the patient’s house, and she (for it is generally a woman),
arrayed in all the finery and trinkets that can be borrowed from the
neighbours, is excited by the music to dance, day after day if
necessary, till she drops down from utter exhaustion. The disease is
attended with a great emaciation; and the doctor says “he was almost
alarmed to see one nearly a skeleton move with such strength.” He
then proceeds to recount his own domestic calamity in a strain of the
most commendable candour:—
“I could not have ventured to write this from hearsay, nor could I conceive it
possible until I was obliged to put this remedy in practice upon my own wife, who
was seized with the same disorder. I at first thought that a whip would be of some
service, and one day attempted a few strokes when unnoticed by any person, we
being by ourselves, and I having a strong suspicion that this ailment sprang from
the weak minds of women, who were encouraged in it for the sake of the grandeur,
rich dress, and music which accompany the cure. But how much was I surprised,
the moment I struck a light blow, thinking to do good, to find that she became like
a corpse; and even the joints of her fingers became so stiff that I could not
straighten them. Indeed, I really thought that she was dead, and immediately made
it known to the people in the house that she had fainted, but did not tell them the
cause; upon which they immediately brought music, which I had for many days
denied them, and which soon revived her; and I then left the house to her relations,
to cure her at my expense. One day I went privately with a companion to see my
wife dance, and kept at a short distance, as I was ashamed to go near the crowd. In
looking steadfastly upon her, while dancing or jumping, more like a deer than a
human being, I said that it certainly was not my wife; at which my companion
burst into a fit of laughter, from which he could scarcely refrain all the way home.”
The capability of sustaining the most violent exercise, for a long

time together, and on very little food, is not one of the least
perplexities attendant upon these nervous or epileptic diseases. The
partial suspension of sensation and volition, by sparing the brain,
may have something to do with it. But into scientific perplexities of
this kind we cannot now enter. One plain and homely caution is
derivable from all these histories. Good sense is a great preservative
of health. Do not voluntarily make a fool of yourself, or your folly
may become in turn the master of your reason. Epilepsy has been
brought on by the simulation of epilepsy. We doubt not that a man
might dance to his own shadow, and talk to it, as it danced before
him on the wall, till he drove himself into a complete frenzy. A sect in
America thought fit to introduce certain grimaces, laughing,
weeping, and the like, into their public service. It was not long before
their grimaces, in some of their numbers, became involuntary; the
muscles of the face had escaped the control of the will. A decided
tongue-mania was exhibited a short time amongst the Irvingites.
Happily, in the present state of society, men’s minds are called off
into so many directions, that a predominant idea of this kind has
little chance of establishing itself in that tyrannous manner which we
have seen possible in the middle ages. But it is better not to play with
edged tools. If people will stand round a table, fixing their minds on
one idea—that a certain mysterious influence will pass through their
fingers to move the table—they will lose, for a time, the voluntary
command over their own fingers, which will exert themselves
without any volition or consciousness on their part. They are
entering, in fact, into that state which, in the olden time, was
considered a demoniacal possession; so that, speaking from this
point of view, one may truly say that “Satan does turn the table,” but
it is by entering into the table-turner. When we have been asked
whether there is anything in mesmerism, we have always answered
—a great deal more than you ought, without medical advice, to make
trial of. Nor do we at all admire the performance of the so-called
electro-biologist. Experiments in the interest of science are
permissible; but is it fit that any one should practise the art of
inducing a temporary state of idiocy in persons of weak or
susceptible nerves, for the purpose of collecting a crowd, and passing
round the hat?
The subject of the third treatise of Dr Hecker is the Sweating
Sickness. This third part is more miscellaneous than its predecessors,
and we have no space to do justice to its varied and sometimes
disputable matter. Dr Hecker describes the sweating sickness as a
legacy left us by the civil wars of York and Lancaster. It first
developed itself in Richmond’s army, which had been collected from
abroad, over-fatigued by long marches in a very damp season, and
probably ill supplied with rations. Its rapid extension through the
cities he attributes to the intemperance of the English, to their
overfeeding, and the want of cleanliness in their houses. Gluttony,
and the filth of the rush-covered floors, he detects even amongst the
wealthiest of the land. For a minute description of the disease, and
the Doctor’s investigation into the nature of it, we must refer to the
book itself.
On the physicians, and the manner in which they addressed
themselves to the encounter of this strange calamity, there is a
passage which it may be instructive to peruse:—
“The physicians could do little or nothing for the people in this extremity. They
are nowhere alluded to throughout this epidemic, and even those who might have
come forward to succour their fellow-citizen, had fallen into the errors of Galen,
and their dialectic minds sank under this appalling phenomenon. This holds good
even of the famous Thomas Linacre, subsequently physician in ordinary to two
monarchs, and founder of the College of Physicians in 1518. In the prime of his
youth he had been an eyewitness of the events at Oxford, and survived even the
second and third eruption of the sweating sickness; but in none of his writings do
we find a single word respecting this disease, which is of such permanent
importance. In fact, the restorers of the medical science of ancient Greece, who
were followed by all the most enlightened men in Europe, with the single exception
of Linacre, occupied themselves rather with the ancient terms of art than with
actual observation, and in their critical researches overlooked the important events
that were passing before their eyes. This reminds us of the later Greek physicians,
who for four hundred years paid no attention to the smallpox, because they could
find no description of it in the immortal works of Galen!”
Who shall say, in reading such passages, that the New Philosophy
of Bacon, which reads now like old common-sense, was not sadly
wanted, if the learned physician, while feeling his patient’s pulse,
could see only with the eyes of Galen? In the fourteenth century we
see the physician busied with his astrology, and laboriously fixing the
day when Saturn, Jupiter, and Mars, did battle with the sun over the
great Indian Ocean; in the sixteenth we find him, with quite dialectic
mind, absorbed in the study of his classical authorities; at the
present time we may truly say that there are no inquiries conducted
with a more philosophical spirit, or with greater zeal and energy,
than those which relate to the human frame, its functions and its
diseases. The extreme complexity of the subject renders our progress
slow. And yet progress can hardly be said to have been slow. Let any
one take up that admirable little manual on The Nervous System, by
Dr Herbert Mayo, and compare it with any work a hundred years old:
it is a new science; and that not only from the new facts which a
Robert Bell and a Marshall Hall, and other distinguished men in
France and Germany, have added to our knowledge, but from the
fine spirit of philosophical inquiry which presides over the whole. We
have not only left astrology behind, we have not only left behind the
undue reverence to classical authority, but we have thrown aside that
dislike and depreciation of physiology which the metaphysician had
done his part to encourage, and have entered, as with a fresh eye and
a beating heart, upon the study of the wonders of the human frame.
THE SONG OF METRODORUS.
Παντοίην βιότοιο τάμοις τρίβον. εἰν ἀγορῇ μέυ

κύδεα καὶ πινυταὶ πρήξιες. ἐυ δὲ δόμοις
ἄμπανμ’. ἐν δ’ἀγροῖς Φύσιος χάρις. ἐν δὲ ζαλάσση
κέρδος. ἐπὶ ξείνης, ἢν μὲν ἔχης τι, κλέος.
ν δ’ ἀπορὴς, μόνος οἶδας. ἔχεις γάμον; οἶκος ἄριστος
ἔσσεται. οὐ γαμέεις; ζης ἔτ’ ἐλαφρότερον.
τέκνα πόζος. ἄφροντις ἄπαμς βίος. αἱ νεότητες
ῥωμαλέαι. πολιαὶ δ’ ἔμπαλιν εὐσεβέες.
οὐκ ἄρα τῶν δισσῶν ἑνὸς αἵρεσις, ἢ τὸ γενέζαι
μηδέποτ’, ἢ τὸ ζανειν. πάντα γὰρ ἐσζλὰ βίῳ.
Metrodorus was a rare old blade,

His wine he drank, his prayers he said,
And did his duty duly;
But with grave affairs of Church and State
He never fretted his smooth pate,
For he said, and he said full truly,
If a man about and about will go,
To mend all matters high and low,
He’ll find no rest full surely.
In his chair of ease a thorn will grow,
The gall will in his bladder flow,
Thick seeds of sorrow he will sow,
And make his dearest friend a foe,
And go to the grave prematurely.
One day he sate beside the fire,
With all things square to his desire
—A wintry day, when Boreas blew
Through the piping hills with a halloo—
Just after dinner, when the wine
On the tip of his nose was glowing fine.
A pleasant vapour ’fore him floats,
The logs are blazing brightly,
And in his brain the happy thoughts
Begin to move full lightly.
He never wrote a verse before,
Though now he counted good threescore,
And scarcely knew what poets meant,
When in their high conceited bent
They talked of inspiration.
But now his soul a fancy stirred;
He trilled and chirped like any bird;
His bright imagination
Poured forth a pleasant flowing verse,
Which, if you please, I will rehearse
For gentle meditation.
’Twas Greek of course, but by the skill
Made English, of my classic quill,
As good, or better, if you will,
In this my free translation.
1.
They may rail at this world, and say that the devil
Rules o’er it, usurping the mace of the Lord;
In my soul I detest all such impious cavil,
While I sit as a guest at life’s bountiful board.
I was young; I am old, and my temples are hoary,
On Time’s rocking tide I have gallantly oared;
This wisdom I learned, ’tis the sum of my story,
With blessings God’s earth like a garner is stored.
2.
You blame your condition; by Jove I was never
So placed that I could not with pride be a man;
At rest or afloat on life’s far-sounding river,
Content was my watchword, enjoyment my plan.
Where busy men bustle, to elbow and jostle
What sport! then at home how delightful repose!
What comfort and pleasure your body to measure
At large in the elbow-chair, toasting your toes!
3.
A soldier? how gallant through smoke and through thunder
To ride like the lightning, when Jupiter roars;
A farmer? to gaze on the green leafy wonder
Of April how sweet, and to think on the stores
Of golden-sheaved autumn!—to dash through the billow
Is dear to the merchant who carries his gains;
How sweet to the poet on green grassy pillow,
To lie when spring zephyrs are fanning his brains!
4.
When you find a good wife, Nature urges to marry;
But art thou a bachelor, never complain;
Less sail you display, but less burden you carry,
And over yourself like a king you may reign.
’Tis pleasant to hear children prattling around you,
Thank Heaven you’ve arrows enough for your bow;
But if you love quiet, they’ll only confound you,
So if now you have none—may it ever be so!
5.
Art young? then rejoice in thy youth,—give the pinion
Of passion free play—love and hate like a man;
And gather around thee a mighty dominion
Of venturous thoughts, like the crest-waving van
Of a conquering host. Art old? reputation
And honour shall find thee and pleasures serene,
And a power like to Jove’s, when the fate of the nation
Shall wait on thy word in the hall of the queen.
6.
Blow hot or blow cold, with hearty endeavour
Still witch out a virtue from all that you see;
Use well what you get, giving thanks to the Giver,
And think everything good in its place and degree.
I’ve told you my thoughts, and I think you’re my debtor,
And if you don’t think so, I wish you were dead;
The sooner you rot on a dunghill the better,
You’re not worth the straw that they shake for your bed.

Principles of Functional Analysis Second Edition Martin Schechter

Uploaded by

Copyright:

Available Formats

Principles of Functional Analysis Second Edition Martin Schechter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Principles of Functional Analysis Second Edition Martin Schechter

Uploaded by

Copyright:

Available Formats

Full download ebook at ebookgate.

Principles of Functional Analysis Second

Download more ebook from https://ebookgate.com

Principles of Functional Verification 1st Edition

Hydrology Principles Analysis and Design Second Edition

Obstinate Hebrews Ronald Schechter

The Functional Analysis of English 3rd Edition Thomas

Principles of Insurance Second Edition Loma

Phonological Analysis A Functional Approach Donald A.

PRINCIPLES OF MATHEMATICAL ANALYSIS 3rd Edition Walter

Trace Environmental Quantitative Analysis Principles

American Mathematical Society

American Mathematical Society

Library of Congress Cataloging-in-Publication Data

To my wife, children, and grandchildren.

May they enjoy many happy years.

PREFACE TO THE REVISED EDITION xv

Chapter 1. BASIC NOTIONS 1

§3.7. The open mapping theorem 71

§7.8. Problems 179

Chapter 12. BILINEAR FORMS 265

§15.2. A diﬀerential operator 360

The ﬁrst edition of Principles of Functional Analysis enjoyed a successful

We added sections to Chapters 3, 5, 7, 9, and 13 and inserted a new

We attempted to strengthen those areas in the book that demonstrate

required machinery is already in place. By these means we were able to

The new chapter (Chapter 14) deserves some additional remarks. It

We have included two appendices. The ﬁrst gives the deﬁnitions of

“ ‘Charming’ is a word that seldom comes to the mind of a science

“Professor Schechter has written an elegant introduction to functional

“The book can be warmly recommended to any reader who wants to

The book is intended for a one-year course for beginning graduate or

I have restricted my attention to normed vector spaces and their impor-

This approach results in a few topics being covered in several places in

9, with a complex form given in Chapter 6, and a geometric form in Chapter

This approach has further resulted in a somewhat unique structure for

1.1. A problem from diﬀerential equations

two functions, we shall feel free to impose a further restriction on A and B,

can be made as small as we like by taking m and n large enough. Again I

1.2. An examination of the results

(1) They can be added. If f and g are in C, so is f + g.

(6) For each real number α, αh ∈ C.

for all v, w ∈ X. Then for each u ∈ X there is a unique f ∈ X such that

The uniqueness in Theorem 1.1 is trivial. In fact, suppose there were

As an example, let k(x, t) be a continuous function in the square a ≤

then the operator

1.3. Examples of Banach spaces

and we want to prove

Proof. Let α be any scalar. Then

which is exactly what we want.

Now that we know that Rn is a Banach space, we can apply Theorem

In particular for each i

Set f = (α1 , · · · , αn , · · · ). Is f ∈ l∞ ? Well, let us see. If we ﬁx j and let

But by (1.31), we have

This is precisely what we want.

≤ f 2 + g2 + 2f g = (f + g)2

Fix j in (1.34) and let k → ∞. Then

Since this is true for each n, we have

provided this series converges. By (1.26) we have