Augustin-Louis Cauchy

Painting of Augustin-Louis Cauchy by 1. RoHer (~1840).

 Bruno Belhoste

Augustin-Louis Cauchy
A Biography

Translated by Frank Ragland

With 34 Illustrations

Springer-Verlag
New York Berlin Heidelberg London
Paris Tokyo Hong Kong Barcelona
 Bruno Belhoste
Service d'histoire de l'education
Institut national de recherche pedagogique
Paris 75005
France

Mathematics Subject Classification (1980): 01A70, OlA55, 01A50

Library of Congress Cataloging-in-Publication Data

Belhoste, Bruno.
[Cauchy, 1789-1857, English]
Augustin-Louis Cauchy: a biography/Bruno Belhoste: translated
by Frank Ragland.
p. em.
Translation of: Cauchy, 1789-1857.
Includes bibliographical references.
ISBN-13:978-1-4612-7752-1 (alk. paper)
1. Cauchy, Augustin Louis, Baron, 1789-1857. 2. Mathematicians-
France-Biography, 3. Mathematics-France-History-19th century.
l. Title.
QA29.C36B4513 1991
5\O'.92-dc20
[BJ 89-26329

Printed on acid-free paper.

© 1991 Springer-Verlag New York Inc.
Softcover reprint of the hardcover 1st edition 1991

All rights reserved. This work may not be translated or copied in whole or in part without the
written permission of the publisher (Springer-Verlag New York, Inc., 175 Fifth Avenue, New
York, NY 10010, USA), except for brief excerpts in connection with reviews or scholarly analysis.
Use in connection with any form of information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed
is forbidden.
The use of general descriptive names, trade names, trademarks, etc., in this publication, even if
the former are not especially identified, is not to be taken as a sign that such names, as understood
by the Trade Marks and Merchandise Marks Act, may accordingly be used freely by anyone.

Typeset by Thomson Press (India) Limited, New Delhi, India.

987654321

ISBN-13:978-1-4612-7752-1 e-ISBN-13:978-1-4612-2996-4
DOl: 10.1007/978-1-4612-2996-4
 Foreword

To write a biography about a leading scientific figure is admittedly an

ambitious undertaking for a historian, since, in addition to the usual
difficulties presented by biographical studies, one must now contend with the
strange intricacies of scientific research and thought. This difficulty is
compounded considerably in at least two ways for one who would write a
biography of Augustin-Louis Cauchy (1789-1857). First of all, the vast output
of Cauchy's creative genius can be overwhelming on its own terms-even for
mathematicians. Second, mathematical notation, concepts, and terminology
were far from logical or uniform in Cauchy's day, and thus the historian faces a
double challenge in pursuing the development of scientific ideas and the
relationships between them. An equally difficult (but no less important) task
facing a biographer of Cauchy is that of delineating the curious interplay
between the man, his times, and his scientific endeavors. I think Professor
Belhoste has succeeded admirably in meeting all of these challenges and has
thus written a vivid biography that is both readable and informative.
Professor Belhoste's subject stands out as one of the most brilliant,
versatile, and prolific figures in the annals of science. Cauchy was a man who,
though a creator of our times, was nevertheless very much a creation of his
own age, a particularly dynamic period in Western history that has come to be
known as the Age of Revolutions (1789-1848). Although the revolutions of
that era may have been primarily political upheavals, they were also scientific
in the important sense that it is during this period that science and scientific
scholarship passed from the control of a few enlightened despots and
aristocrats into the hands of the state and its appendages, the universities and
other forerunners of today's research institutions. It can, of course, be debated
whether this professionalization of science was really revolutionary or merely
evolutionary in the sense of a clearly discernable shift in how, where, by whom,
and under whose auspices systematic scientific investigations should be
carried out. However, what Professor Belhoste's work makes admirably clear
is that by the time Cauchy became established in the world of science, these
issues had already been fairly well decided.
v
vi Foreword

Nearly two hundred years have now passed since the young Cauchy set
about his task of clarifying mathematics, extending it, and applying it
(whenever possible) and placing it on a firm theoretical footing. Even as he
reacted to his society's sudden shifts and turns-political, religious, and
intellectual-in ways which may seem to us today as hasty and ill-
considered, he doggedly pursued this youthful vision. In doing so, he made
many fundamental contributions not only to mathematics but to physics and
astronomy as well. The true measure of Cauchy's success in the grand
undertaking of his youth must necessarily be sought in the standards and
methods that he bequeathed to modern science. Ifhe was not wholly successful
in the task he originally set for himself, his lack of success would seem to have
less to do with any shortcomings or defects on his part than with the
stupendous vastness and subtlety of mathematics and its related disciplines.
And this, too, is yet another level on which Professor Belhoste's work achieves
practical importance; for here we are afforded a detailed, rather personalized
picture of how a first-rate mathematician worked at his discipline-his
strivings, his inspirations, his triumphs, his failures, and above all, his conflicts
and his errors. In this respect, then, this study should be of signal interest to
young students of the mathematical sciences, since, all too often, they only
catch a glimpse of Cauchy as a supremely confident and creative genious and
not as a human being endowed with his share of the errors, weaknesses, and
shortcomings that are such an important part of human nature.
In translating this work, I have tried to strike a balance between the flavor
of the original French study and requirements of readability. To this end, in
the body of the text I have given the French titles of Cauchy's works (and for
those of his contemporaries also). Similarly, the original French titles are used
exclusively in the bibliographical notes, and this applies also to the works of
other authors mentioned in the course of the text. In this way, any reader
desiring to consult the originals will have little difficulty in doing so. Similarly,
I have taken care to maintain the French system of citation used by the author.
Finally, I would like to thank Mss. Connie Burgess and Joan Passalacqua
for their patience and steadfast devotion in typing the manuscript.

Frank: Ragland
New York
Winter, 1989
 Preface

Cauchy was the true heir of the great mathematical tradition of the 18th
century, the heir of Euler, Legendre, Lagrange, and Laplace; and yet he was
both a leading standard-bearer and an active creator of an essentially new
approach to mathematics. Indeed, Cauchy and Karl Friedrich Gauss (1777-
1855) may rightly be called the first truly modern mathematicians. The sheer
bulk of his scientific productivity is immense. The Oeuvres Completes, a
publication which covers almost an entire century, from 1882 to 1975, fills
more than 27 large volumes and contains, in addition to five complete
textbooks, nearly 800 research articles and treatises. Cauchy'S work does
contain an element of redundancy. But, on balance, such an enormous
scientific creativity is nothing less than staggering, for it presents research on
all the then-known areas of mathematics: arithmetic, algebra, geometry,
statistics, mechanics, real and complex analysis, and mathematical physics.
Nevertheless, in spite of its vastness and rich multifaceted character, Cauchy's
scientific works possess a definite unifying theme, a secret wholeness. This, at
least, is the essential point of departure of the present study.
There are several ways in which a study of this type might have been
developed. A historian by training and by temperament, I have chosen the
most natural method: the biography. To be sure, there already exists a
biography of Cauchy. In 1868, eleven years after Cauchy's death, CA. Valson,
a mathematician of Catholic persuasion who was preparing the publication of
the Oeuvres Completes, published a biography. 1 Published under the auspices
of the Cauchy family, this two-volume study is not without interest today. This
is mainly so because Valson had at his disposal certain of Cauchy's personal
papers-documents which have now been completely destroyed-as well as
the family's archives and the actual testimony of many of Cauchy's associates.
In spite of this, however, Valson's study fails to meet the standards of rigor and
scholarship demanded by modern historical studies. Replete with irrelevant

1 La Vie et les Travaux du Baron Cauchy (2 volumes, Paris, 1868).

vii
Vlll Preface

moral judgements, and frequently succinct on many aspects of the great

mathematician's life and works, Valson's work would seem to be more
hagiography than history.
In the present work, I have earnestly sought to present a portrait of a
mathematician and the age in which he lived. In this way, I hope to underscore,
to highlight, as it were, the essential coherence of an active, productive career
in the sciences which stretched over nearly half a century. Certainly, Augustin-
Louis Cauchy was a far cry from the romantic heros of that era. His long
and active life lacks the ephemeral sparkle and brilliance of that of his
contemporary Evariste Galois. A devout Catholic and close associate of the
Jesuits as well as a strict royalist, Cauchy quickly became a recognized mathe-
matician who, at a very early age, took a place in the leading scientific
institutions of his day. On the other hand, however, he was a proud man,
a man of passionately held convictions who, whenever occasions arose for
him to defend or to explain those things that he regarded as "the truth,"
consistently refused to allow himself to be swayed by considerations of
personal convenience or self-interest. Thus, for example, in the political sphere
his adherence to the Bourbon cause, was, for better or worse, absolute and
unyielding. Caught up in the political purges of the 1830s, though he himself
had earlier profited from similar procedures in 1816 after the fall of
Bonaparte, Cauchy chose exile over perjury. In fact, from 1830 until his death
in 1857, he steadfastly refused to swear loyalty oaths to any ofthe regimes that
governed France. In a similar way, Cauchy's belief in Catholicism was
uncompromising and apparently untarnished by any doubts whatsoever.
Indeed, throughout his life, Cauchy practiced his faith with all the zeal of a new
convert and with all the feverish passion of a missionary.
Cauchy's view of mathematics sprang from the same deeply felt need for
absolutes, for certainty. If, in politics, this need manifested itself in uncompro-
mising adherence to the Bourbon cause and in religion to strict Roman
Catholic orthodoxy, then in mathematics its expression was reflected in a
demand for rigor and irrefragable proof. To Cauchy, the real work of a scholar,
of a scientist, must necessarily be a quest for truth. 'Truth," he wrote in 1842,
"is a priceless treasure which, whenever we manage to acquire it, cannot bring
us remorse and sorrow; it cannot disquiet and distress our soul. The mere
thought of its heavenly attributes, of its divine beauty suffices to replenish us
for all the sacrifices we may have made in discovering it. Indeed, the joy of
heaven itself is but the full and complete possession of immortal truth." Feeling
uncomfortable in the age in which he lived and often misunderstood by his
contemporaries, Cauchy found a refuge in mathematics. Here, in a world far
removed from the everyday one, his creative genius could thrive, expand, and
reach its full measure of expression.
Still, in a sense, his view of mathematics represents a certain expression of
the concerns of the era in which he lived. Thus it was that Cauchy early
rejected the optimistic rationalism of the Age of the Enlightment in
no uncertain terms even though Laplace, his mentor and protector, was
 Preface ix

ever faithful to the credo of the philosophes. As it specifically concerns

mathematics, it should be noted that analysis had experienced tremendous
growth all during the 18th century. Unfortunately, this growth had come
about at the expense of mathematical rigor; and Cauchy, like the more able of
his associates and contemporaries, set about strengthening the theoretical
foundations of mathematics and restructuring the entire edifice. His position
as a professor of mathematics at the Ecole Poly technique provided him with
an excellent forum for working towards these ends. Certainly, his demand for
rigor in mathematics proved to be exceedingly fruitful, since it opened up
wholly new rich fields of mathematical research.
But Cauchy's creative genius found broad expression not only in his work
on the foundations of real and complex analysis areas to which his name is
inextricably linked, but also in many other fields. Specifically, in this
connection, we should mention his major contributions to the development of
mathematical physics and to theoretical mechanics, two fields which experien-
ced rapid growth during the 19th century. Along these lines we mention,
among others, his two theories of elasticity and his investigations on the theory
of light, research which required that he develop whole new mathematical
techniques such as Fourier transforms, diagonalization of matrices, and the
calculus of residues.
It should be observed that as to his mathematical talents, Cauchy-like
Euler and Gauss-was a universalist in the fullest sense of the term. His few
works on the theory of numbers, algebra, and geometry make us deeply regret
that he did not devote more effort to these areas. It can hardly be doubted that
had he done so, he would have obtained results of signal importance.
Cauchy's creativity, however, bore the stamp of his training as an engineer.
Accordingly, Cauchy, like other great mathematicians of the French School,
always gave priority to questions about applied mathematics which were a
great source of inspiration for him.
Today, two centuries after Cauchy'S birth, his work has been completely
integrated into the austerely beautiful and elegant structure of mathematics, a
structure which is constantly changing. And thus it is that Cauchy's theorems
and theories have been reformulated in newer, more modern terms and the
memory of the scholar himself has been progressively weakened. Only
his name attached to a few outstanding results remains to give testimony
to future generations of the importance of his works. It is the fate of
all mathematicians to see their individual contributions become quickly
absorbed into the great common structure. And while this is hardly the place
to dwell on this simple fact, we should nevertheless pause and consider another
simple fact: mathematics-regardless of how impersonal it may be in its
formal beauty and power, is not something handed down complete and perfect
from heaven. Rather it is the cumulative result of the work (and, quite
frequently, of the sufferings also) of many individual human beings. If,
somehow, this biography of Cauchy should help in establishing this belief,
then I will have attained my goal.
x Preface

This book was the outcome of several years of work, but it could not have
been completed without the help of a number of persons. First of all, lam
deeply grateful to Professor Rene Taton, who directed my initial research with
great care and kindness. I must also express my gratitude to those who were
kind enough to provide me with often rare documents and information. In this
respect, I am particularly indebted to Professors Dugac, Grattan-Guinness,
Ross, Russo, and Yuschkevich. I also recall with special gratitude and
fondness the very fruitful discussion that I had with my good friends Amy
Dahan-Dalmedico and Jesper Liitzen. Nor can I forget the warm reception I
was always accorded at the archives and at the libraries. Special mention must
be given to the Secretariat of the Academy of Sciences where M. Berthon and
Mme Pouret kindly placed at my disposal their profound knowledge of the
archives. Similarly, I was greatly assisted by the cogent advice of Mlle Billoux
at the Central Library of the Ecole Polytechnique. I also want to express my
gratitude to Professor Frank Ragland of the City University of New York who
translated this work into English with great skill and care. Finally, I should
like to thank Jeremy Gray and John Greenberg who carefully read the
manuscript. Their comments enabled me to make a number of changes in the
final text.

Paris, France Bruno Belhoste

Autumn, 1989 INRP, Service d'histoire
de l'education
 Contents

Foreword . . . . . . v

Preface . . . . . . . . vii

Chapter 1 The Formative Years. 1

Chapter 2 Sojourn at Cherbourg 18

Chapter 3 The Waiting Room of the Academy .. 32

Chapter 4 A Man of Science . . . . . . . . . . . . . 43

Chapter 5 Teaching at the Ecole Poly technique .. 61

Chapter 6 From the Theory of Waves to the Theory of Light 87

Chapter 7 From the Theory of Singular Integrals

to the Calculus of Residues . . . . . . . 107

Chapter 8 A Mathematician in the Congregation. . . 132

Chapter 9 Exile in Turin . . . . . . . . . . . . . . . . 143

Chapter 10 The Education of the Duke of Bordeaux 159

Chapter 11 The Legitimist Mathematician. . . . . . 174

Chapter 12 Scientific Works from 1838 to 1848. . . 191

Chapter 13 Practices and Principles in Cauchy's Works 213

xi
xii Contents

Chapter 14 The Final Years 1848-1857 223

Notes . . . . 241

Appendix I: Presentation Notes of Various Works Presented by

Cauchy to the Academie des Sciences (1816-1830). .. 296

Appendix II: Documents on Cauchy's Analysis Course

at the Ecole Poly technique (1816-1819).. . . . . .. 303

Appendix III: Selection of Letters from Cauchy to Various

Persons (1821-1857) . . . . . . . . . . . . . . . .. 321

Appendix IV: Unpublished Documents on Cauchy's Candidacies

to the College de France (1843 and 1850-1851) . . . 338

Manuscript Sources and Bibliography . . . . . . . . . . 359

Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . 370

Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 371
 Chapter 1
The Formative Years

Augustin-Louis Cauchy was born on August 21, 1789 in Paris and was
baptized in the parish church ofSaint-Roch. He was christened Augustin after
the month of his birth and Louis after his father, Louis-Fram;ois. Louis-
Fran~ois Cauchy was the principal commis of the Lieutenant General de
Police of Paris. By an unfortunate coincidence, his firstborn came into the
world at the moment when his career became tragically compromised by the
Paris insurrection. Disturbed by a fear of shortages, and cheered on by a
bourgeoisie that was worried about its incomes and investments, the people of
Paris rioted following the dismissal of the principal minister Necker. But, this
was to be no ordinary riot; it would soon become outright revolution.
A revolutionary council was installed by the citizens of Paris during the
night of July 12-13, with the astronomer Bailly as president. On the next day,
July 14, the insurgents, searching for powder and weapons, attacked the
Bastille, the hated symbol of absolutism. Flesselles, the Provost, was
slaughtered. A few days later, the royal intendants Foullon and Bertier de
Sauvigny fell to the same fate. Thereafter, the royal administration was
powerless in Paris. Louis Thiroux de Crosne, the Lieutenant General de
Police, was deprived of all authority during the first hours of the insurrection
and withdrew discredited from office; on July 15, de Crosne handed his
command of the municipal police over to the new mayor of Paris and on the
23rd fled to England where he remained for the time being. The events also
meant the end of the career of his principal commis, Louis-Fran~ois Cauchy.
It was one offate's strange quirks that Augustin-Louis should be born at the
time his father lost his position and his protector, and the coincidence could
not have gone unnoticed by the family. Indeed, one is inclined to think that this
particular coincidence had an indelible subconscious effect on young Cauchy.
In later life, he worked unceasingly to undo the results of the Great Revolution
of 1789. He regarded it as a fatal disaster and fought against the ideas that had
triggered it. His hatred of the Revolution was so intense and so uncom-
promising that when the events of July 1830 occurred, he simply could not
2 1. The Formative Years

bear to watch the reemergence ofthe Revolution. It is likely that he always felt
a secret guilt vis-a.-vis his father, a guilt for having dared to come into the world
at so unpropitious a time.
A few words about Louis-Fran!;ois are in order. The elder Cauchy was born
on May 27, 1760 in Rouen to Louis-Charles Cauchy, a Rouen businessman (1).
Louis-Charles, who, according to Theodore Lebreton (2), was a master
locksmith and metal worker, had only one sister, Marie-Madeleine, who
married Laurent Larsonnier. Larsonnier was employed at the customs house
at Versailles on the eve of the Great Revolution and became treasurer of the
Chamber of Peers during the Restoration with the support of his brother-in-
law. Louis-Fran!;ois received a good education, a fairly common practice
among the bourgeoisie of that era. Sent to the College de Lisieux in Paris, he
was awarded honors in 1777 by the University of Paris for his performance in
the Concours general, (a celebrated competition between the students of the
Parisian colleges). Mterward, he returned to his hometown, where he worked
as a lawyer near the Parliament of Normandy. Soon, in 1783, he was appointed
Secretary-General of the Intendance (Administration) of Rouen (3), the
intendant Louis Thiroux de Crosne apparently recognized that this young
lawyer was uncommonly able. Thiroux, himself a gifted administrator, had
been intendant at Rouen since 1767. He was a very cultured man, a member of
the Academy of Rouen since 1771. Imbued with modern ideals, he was capable
of recognizing talent when he saw it. Thus, when he was appointed Lieutenant
General de Police of Paris in August of 1785, he took his esteemed secretary
along, naming him his principal commis.
During the final years of the Old Regime, the government made extensive,
elaborate plans for the rebuilding of Paris. At the time, Paris was a city of more
than 600,000, among the largest cities in Europe-second in size only to
London. In spite of its size, however, it was still essentially a medieval town:
overcrowded, dirty, dangerous, and unhealthy. Its narrow, dusty streets, open
drains and ditches, and cemeteries in the heart oftown distressed and shocked
the enlightened minds of polite society, a society that prized order, balance,
and reason. The government, urban designers, and engineers looked forward
to the building of a new city: a city that would be healthy, well lighted, well
policed, and beautiful. At the time, however, the city districts were hemmed in
by tollgates and stations, and traffic swarmed across the bridges. But, new
metropolitan areas and avenues were in the planning.
Louis-Fran!;ois took part in these projects and discussions, which were
soon to be interrupted by the onslaught of the Great Revolution, but which
ultimately would be carried through to fruition under successive regimes
during the 19th century. In 1786, he supervised the removal of the remains
from the Cemetery of the Innocents with Thouret and Fourcroy (4). He was
also in charge of controlling theaters and the book trade and managing state
prisons such as the Bastille (5).
Louis-Fran!;ois was still unmarried, living with his widowed mother,
perhaps at the home of a cousin, Antoine Thibaut Beauvais, a middle-class
 1. The Formative Years 3

Parisian, in the rue Saint-Honore. In October 1787, he married a 20-year-old

Parisienne, Marie-Madeleine Desestre. By his marriage to Marie-Madeleine,
Louis-Franr;ois became part of a family that was well established in the lower
ranks of Paris officialdom; Marie-Madeleine's father, Louis-Jacques Desestre,
was the Elder Dean of Bailiffs of the King's Conseil d'Etat; her uncle, Jean-
Baptiste Desestre, was employed as Inspector of Revenues at the Hotel de Ville
de Paris (Town Hall). Similarly, her mother, Madeleine Paupelin, came from a
solidly bourgeois Parisian family, a family containing lawyers, merchants, and
even some members of the bureaucracy, who were well on their way to
ennoblement.
Marie-Madeleine was a good match for Louis-Franr;ois, who was not well
off; he owned a small family plot at Frettecuisse in the Picardie, and he had an
income of 150 pounds, 5000 pounds in notes and coin, 8000 in other movable
property, and a life annuity valued at 800 pounds. Present worth aside, he now
had an important position and a very bright future indeed. As for his wife, she
had brought a dowry of 40,000 pounds, 15,000 in cash, and an income of 1250
pounds per annum to be paid in perpetuity, as well as a life annuity worth 500
pounds (6). Shortly after their marriage, the young couple bought a few acres
and a spacious country house in Arcueil, in the rue des Carnaux (7).
The Revolution interrupted his climb upward, and the following years were
more difficult for Louis-Franr;ois. He was comprpmised; he had to make
himself forget. According to Valson, he took a position as Chief of the Bureau
of Almshouses and Charity Workshops, a post that he kept until the Reign of
Terror (8). But, finally, when Louis Thiroux returned from abroad and was
arrested and quickly condemned to death on the same day, April 28, 1794,
Louis-Franr;ois began to fear for himself and his family and decided to leave
Paris where, he realized, he stood in danger of being denounced by the
revolutionary authorities. Fleeing the dangerous turmoil that now gripped the
capital, he took his wife and two children, Augustin-Louis and the baby
Alexandre-Laurent, who had been born on March 12, 1792, to their country
house in Arcueil, where they remained until the Reign of Terror had passed.
But, if Arcueil was a haven from denunciation and execution, life there was
hardly easy; sufficient food was a major concern, as the following remarks by
Louis-Franr;ois, which are cited in Valson, attest:
We never have more than a half pound of bread-and sometimes not
even that. This we supplement with the little supply of hard crackers and
rice that we are allotted. Otherwise, we are getting along quite well,
which is the important thing and which goes to show that human beings
can get by with little. I should tell you that for my children's pap I still
have a bit of fine flour, made from wheat that I grew on my own land.
I had three bushels, and I also have a few pounds of potato starch. It
is as white as snow and very good, too, especially for very young
children. It, too, was grown on my own land (9).
During this period, Augustin-Louis contracted smallpox (10); he did not seem
4 1. The Formative Years

to have adjusted easily to these stressful conditions, to the privation and

material uncertainties, and most importantly, to the threats of arrest that
tormented his father. We know very little about these early years, which must
have been crucial in the development of his personality. Although we have no
direct reports left, we may suppose that henceforth he found refuge in thought
and quiet study, a refuge from the agonies and fears of an uncertain period. A
timid, frail boy, withdrawn and pensive, he had no liking for sports and games.
His love for purposeful work, so unusual in a boy his age, was in striking
contrast to the carefree openness of his younger brother Alexandre.
The fall of Robespierre on July 27, 1794 was greeted with relief: Louis-
Fran<;:ois and his family could now return to Paris unafraid. In the autumn of
that year, he became Assistant Director of the Division of Crafts and
Manufacturing in the Commission of Arts and Crafts, which was housed in the
Hotel Mole in the rue Dominique. The following spring, in April 1795, while
the repression was striking telling blows against the last radicals, he moved up
to the post of director in that bureau, replacing Pierron. From the
reestablishment of the ministries on October 2, 1795 until the coup d'etat of 18
Brumaire (November 10, 1799), Louis-Fran<;:ois remained Director of the
Bureau of Crafts and Manufacturing in the 4th Division of the Ministry of the
Interior [this was the new name of the former Division of Crafts and
Manufacturing since the time of the Convention (11)]. Carrying a salary of
5000 to 6000 francs, this position was certainly not minor. According to the
Almanach National, the Bureau had authority over all industry and
manufacturing, useful arts, handicrafts, patents, and certificates of invention,
as well as the administration and supervision of the Conservatoire des Arts et
Metiers and the national manufacturing enterprises at Sevres, Chaillot (the
Savonnerie), Beauvais, Aubuisson, etc.
During the forced leisure at Arcueil, Louis-Fran<;:ois made good use of his
time by undertaking the education of his children, and he continued this for
several years after their return to Paris. Louis-Fran<;:ois was extremely
solicitous of his children's education. He himself had always been a model
student: was it not knowledge, the fine fruit of education, that had enabled him
to rise to the upper level of government under the Old Regime on the eve ofthe
Great Revolution? Socially, he belonged to the bourgeoisie oftalent, a kind of
meritocracy, that had begun, little by little, to occupy important state
positions even before the Revolution. No doubt he envisioned his sons as
following in his footsteps by entering government service.
Louis-Fran<;:ois wrote 'little didactic compositions' for his sons' educations
(12). These writings were usually written in the orderly, balanced style that was
then characteristic of French verse, and they dealt with subjects such as
grammar, history, and ethics. He later introduced the children to Latin and
Greek, requiring that they study the works of the ancient writers. Also he did
not neglect their religious education. Indeed, from an early age, Augustin-
Louis familiarized himself with biblical texts. A sincere Christian, Louis-
Fran<;:ois was particularly devoted to the 'humanities', the term being used
 1. The Formative Years 5

here in its 18th-century sense, and wrote poetry in French and Latin. Indeed,
as we will see, he even acquired a certain reputation as a poet by publishing
verses during the years of the Consulate, the Empire, and the Restoration.
While these verses were mediocre pieces, flattering and praising the current
rulers, it is interesting to note that his three sons, particularly Augustin-Louis,
picked up the habit of writing French or Latin poems that dealt with diverse
topics and were periodically published.
By all indications, Louis-Fran<;ois did not neglect his children's education
in the sciences. A letter dated February 18, 1799 from Louis-Fran<;ois Cauchy
to the Central Bureau of Correspondence of Le Mans, of which he was, as
Director of the Bureau of Manufactures, a nonresident affiliate, shows that he
sought diversion from administrative routines by studying nature:
The sciences are sisters of the arts, and anyone interested in the latter
cannot be unfamiliar with the former ... I myself have always been
especially fond of natural history, and, although conchology (the study
of shells) may not be the most attractive part of it, even it seemed to me to
present, in various respects, a particular charm. From a practical point of
view, it might present useful applications that have not been discerned
heretofore. I think, for example, that conchology can do more than any
other branch of zoology to acquaint us with the way in which sun and
climate affect animal species. Consequently, the thought occurred to
me to form the most complete collection possible of indigenous testa-
ceans in France. I hope, citizens, that some of you will want to show
me everything that the department of Sarthe possesses ... Good health
and fraternity! (13)
Again, Louis-Fran<;ois' intellectual interest stirred similar interests in his
oldest son, because, later on, at Cherbourg, Augustin-Louis enjoyed collecting
plants in his spare time
The coup d'etat of 18 Brumaire overthrew the Directory and established
the Consulate with Bonaparte as First Consul. Louis-Fran<;ois supported the
new regime enthusiastically and praised its virtues in several writings (14). On
January 1, 1800, he was elected to the post of Secretary-General by the newly
created Senate. In this capacity, he was responsible for the transcription and
editing of that body's proceedings. On the same occasion, he became Archivist
and Keeper ofthe Seal ofthe Senate; it was his duty to countersign and release
all Senate dispatches after they had been authorized by its president. He
worked directly under the Chancellor of the Senate, Count Laplace. For
Louis-Fran<;ois, his election to a post at the Senate represented an extra-
ordinary advancement, a promotion that not only doubled his salary to more
than 12,000 francs, but also brought him influence and prestige: he was in daily
contact with the senators, most of whom were men of considerable standing.
Exactly how Louis-Fran<;ois came to receive the nomination to so prestigious
a position is not known. But, it can hardly be doubted that it was through
the good offices of some influential person, such as F ontanes, a habitue of the
6 I. The Formative Years

reactionary and staunchly catholic Elisa salon (15), or Franr;:ois de Neuf-

chateau, a senator and former Minister of the Interior. The Senate sat in the
Palais du Luxembourg; Louis-Franr;:ois Cauchy and his family lived nearby in
the rue de Tournon, and Augustin-Louis frequently came over to work in his
father's office.
By that time, he had developed an interest in mathematics, since, according
to Val son, who studied Cauchy's school notebooks, 'it was not an infrequent
thing to find a paper on a literature assignment suddenly interrupted: A
mathematical idea would have crossed the youngster's mind and so absorbed
him that he would be forced to translate the compelling notion into numbers
and figures' (16). On several occasions, Louis-Franr;:ois presented his son to
Laplace and Lagrange, both of whom were mathematicians ofthe first rank as
well as senators (17). Lagrange seems to have taken some interest in young
Cauchy, for, according to Valson, one day in 1801, Lagrange made the
following remark to several members of the Senate who were meeting in Louis-

Pierre-Simon Laplace (1749- 1827), in the dress of Grand Chancellor of the Senate.
Cauchy was his protege until 1816.
 1. The Formative Years 7

Fran~ois' office, among them Lacepede:

Now you see that little fellow there, don't you? Well, one day he will
replace all of us simple geometers (18).
Even if we should doubt the remark Valson attributed to Lagrange, it is likely
that he did advise Louis-Fran~ois about to his son's education in mathematics:
Do not allow him even to open a mathematics book nor write a single
number before he has completed his studies in literature (19).
On Lagrange's advice, Louis-Fran~ois enrolled his son in the Ecole Centrale
du Pantheon in the fall of 1802. There he was to complete his studies in the
humanities.
The Ecole Centrale du Pantheon was located not far from the Palais du
Luxembourg, on the Montagne Sainte-Genevieve, housed in the buildings of
the Abbey of the Genovefains (20). This institution was perhaps the best of the
3 ecoles centrales that had been established in Paris after 1795. A little more
than 300 students attended it. In the first section, which was open to boys of
more than 12 years of age and which was the section that Augustin-Louis
entered, 2 years of study were devoted to ancient languages, drawing, and
natural history, subjects that were taught by Bachelier, Maherault, and Cuvier
(21). Augustin-Louis attended the Ecole Centrale du Pantheon during a time
when France's educational structure was in a state of transition, between the
creation of the Lycees under the law of May 1, 1802 and the establishment of
the Lycee Napoleon, which replaced the Ecole Centrale du Pantheon, in
September 1804. Students at the Ecole Centrale were not grouped into classes,
but rather could select the courses that interested them, passing freely from one
course to the other. The code of discipline of these institutions was very liberal,
a far cry from the quasi-military code that would soon be applied to the
imperial Lycees.
Augustin-Louis worked particularly hard in ancient languages during the
two years he spent at the Ecole Centrale du Pantheon and showed himself to
be a very bright young scholar. As early as 1803, he took first place in the
competitive examinations in Latin composition, which had been instituted the
preceding year by Napoleon. In 1804, he won second place in the Latin oratory
competition for advanced students and first place in Greek and Latin poetry.
Most important, however, he won the Grand Prize for the Humanities, an
award that was bestowed on the student who ranked highest in the most
competitions (22). He worked unremittingly in preparation for this examin-
ation, even though he was to take first communion on Easter Sunday, 1804
(23).
Encouraged by Louis-Fran~ois, the young scholar was already so fired by
an ambition to always be first that, writing in his resolutions for first
communion, he modestly pledged that:
I shall never flaunt the little learning that I have acquired through the
care and help my father has given me. If I have learned anything, it is only
8 1. The Formative Years

because he took care to teach me. Had he not taken upon himself the
trouble of instructing me, I would be as ignorant as many other children
(24).

He was already reaping the fruits of his increasing reputation: receiving his
awards at the Institut; dinner at the home of the Minister (25). These were
the things that could easily turn the head of a teenager, who, in his mother's
words, 'then had many faults of character' (26).
In spite of his successes in the humanities, Augustin-Louis decided to
prepare himself for entrance into the Ecole Poly technique the following year.
In doing so, he was rejecting a family tradition: Louis-Fran<;:ois had studied
law, and whatever interest he had in natural history seems rather to have come
from a respect for learning in general rather than from a strong inclination
toward the sciences per se. Augustin-Louis' younger brothers, Alexandre and
Eugene, were to follow in their father's footsteps, and both would enjoy
distinguished careers in law (27). Only Augustin-Louis chose to become an
engineer. His decision seems to have been in response to a kind of peculiar
personal bent, which could be discerned in his early liking for mathematics and
which would be confirmed later. Meanwhile, we should not completely
disregard the influence of the family on his decision. It should be noted, first of
all, that the Ecole Poly technique had been created in response to the
requirements of the public sector. By preparing to enter into a major
institution of the state, Augustin-Louis was acting in a way that was wholly
consistent with his father's views and plans.
As for Louis-Fran<;:ois, far from opposing the choice his oldest son had
made, he actively endorsed it, using his influence to advance his son in his
chosen career and providing him with moral and material support when, later
on, Augustin-Louis began to work on his first scientific projects. Moreover, for
the rest of his life, Augustin-Louis retained a deep filial respect for his parents,
particularly his father. Each week he, along with his sisters and brothers,
would go to the Palais du Luxembourg to enjoy a family luncheon, to sit at the
paternal table 'like a young olive tree', and to realize the blessings promised by
the prophet to 'the man who feared the Lord' (28). Later on, father and son
would study Hebrew together; and, on March 11, 1842, they would even
present a report on biblical prosody before the Academie des Inscriptions et
Belles-Lettres (29).
Augustin-Louis' intellectual personality, which is so strongly stamped on all
of his scientific works, was nurtured in an intimate family circle, in close
contact with a very strict and pious mother and a very open and hard-working
father. It was in this circle that he developed his exceptional capacity for hard
work and his curiosity and interest in learning that, as time passed, became an
almost exclusive passion for truth. He also inherited a certain stubbornness and
rigidity of character that his contemporaries frequently mistook for narrow-
mindedness-and this was particularly the case in political matters. Still, if
stubbornness and rigidity were indeed basic features of his character, it is hardly
 1. The Formative Years 9

likely that without them he would have been able to persist in attacking and
solving so many fundamental and difficult research questions, some of which
did not yield to his efforts for long periods of time.
Without overly simplifying, we can explain Cauchy's peculiar mathemat-
ical qualities in terms of his family: an ability to formalize situations and
manipulate abstractions and a conceptual and logical rigor, as well as clarity
and precision in exposition. Such qualities are, of course, of great value in the
theory as well as the practice of law. Not surprisingly, then, one of Cauchy's
major projects was to put on a firm, rigorous footing many mathematical
methods and procedures that had up to then been used without sufficient
theoretical justification. In other words, his aim was to establish clearly
defined guidelines as to what could and could not be done in mathematics.
Could he pay a more worthy homage to his family'S tradition in law than this?
Seen in this light, then, the Cauchy family's propensity for law was not far
removed from the concerns of the mathematician of the family.
Starting in the autumn of 1804, Augustin-Louis attended mathematics
classes given by Dinet, professor at the Lycee Napoleon and an examiner for
admission to the Ecole Poly technique (30). Cauchy made very rapid progress,
and in 1805, he took the competitive entrance examinations at the Ecole
Poly technique. Examined on October 30,1805 by 1.-B. Biot, he was second out
of the 293 applicants and the 125 admitted. According to the rules he now
had to choose the field of public service that he would enter once his
studies at the Ecole had been completed. In satisfaction of this requirement,
then, he presented the following choices, in order of preference: (1) Ponts et
Chaussees (highways and bridges), (2) Genie maritime (maritime engineering),
(3) Mines (mining engineering), (4) Genie militaire (military engineering), (5)
Corps des ingenieurs geographes (topography), (6) Artillerie de terre (land
artillery), and (7) Artillerie de marine (naval artillery). He put at the top of the
list the Ponts et Chaussees service, a select one, to which almost all the
Polytechniciens aspired.
The Ecole Poly technique had opened its doors at the end of 1794. The
school gave the future civil and military engineers a high-level scientific
education. Once they had completed work at the Ecole Poly technique, the
students rounded out their education with more specialized training in an
ecole d' application, such as the Ecole des Ponts et Chaussees or the Ecole de
l' Artillerie et du Genie at Metz, which prepared them for a specific public
service. At the beginning of October 1805, an important reform instituted by
Napoleon went into effect. The Ecole Poly technique was now transferred from
the Palais-Bourbon to the College de Navarre on the Montagne Sainte-
Genevieve. Henceforth, students were to be organized into military corps and
quartered in barracks. For this, they had to pay a fee of 800 francs, provide
their own uniforms and other personal necessities, and pay for their own
books and equipment. During his first year at the Ecole, Augustin-Louis
belonged to the 4th squadron of the 1st company, which was under the
command of Charles-Emile Laplace, the son of the great Laplace.
10 1. The Formative Years

In general, the mathematical sciences occupied the largest portion of the

scheduled class time at the Ecole. Indeed, in 1806, the distribution of time
among the various disciplines during the first and second years was as follows:
for analysis, 29% during the first year and 18% during the second; for
mechanics, 17% and 22%; and for descriptive geometry, 26% and 3%,
respectively; the physical sciences, on the other hand, were allotted consider-
ably less time in the academic program. Indeed, in 1806, 5% was allotted to
physics in the first year and 7% in the second; the figures for chemistry are 9%
for each ofthe first two years (31). The balance ofthe schedule at the Ecole was
devoted to literary studies and to drafting and drawing during the first year
and to studies in applications (fortifications, construction, and mines) during
the second. Basically, at this time, the students were required to read a number
of books and other materials that had been selected by the professors. We can
get a view of the readings and course requirements from the following list for
1805, which has been supplied by A. Fourcy:
Analysis instruction: the Cours d'Analyse Algebrique by Garnier and the
Traite Elementaire de Calcul DifJerentiel et Integral by Lacroix;
Mechanics instruction: the Traite de M ecanique, using the methods of Prony
and edited by Francoeur and the Plan Raisonne du Cours de Prony;
Descriptive geometry instruction: the Geometrie Descriptive by Monge;
Applied analysis instruction: the Feuilles d'Analyse Appliquee a la Geometrie
by Monge and the Application de l'Algebre a la Geometrie by Monge and
Hachette.
These works constituted the core of the required texts for the mathematical
sciences. In addition to these, there were numerous works that were required in
chemistry, physics, and other areas.
The faculty at the Ecole was indeed an illustrious group: Hassenfratz in
physics, Fourcroy in general chemistry, and Guyton in applied chemistry;
Durand in architecture; Neveu in drawing and drafting; Andrieux in grammar
and fine arts; Sganzin in civil engineering problems; and Duhays in
fortifications and topography. In the mathematical sciences, the core
discipline at the Ecole, two professors worked in each area: Lacroix and
Poisson in analysis, Monge and Hachette in applied analysis and descriptive
geometry, and Prony and Labey in mechanics. In April 1807, Poisson and
Labey exchanged professorial chairs. Cauchy studied analysis under La-
croix, whose course closely mirrored his Traite Elementaire de Calcul
DifJerentiel et Integral; descriptive geometry and applied analysis were studied
under Hachette; and mechanics under Prony (32). Much of the student's work
was done under repetiteurs (tutors). Ampere was Cauchy's tutor in analysis
and mechanics, with Teysseyrre and later Bazaine as repetiteurs adjoints
(assistant tutors); Livet and later Paul Binet were his tutors in descriptive
geometry.
As we have seen, Cauchy began his work at the Ecole Poly technique with
great dash, although his successes during the first year were not especially
 1. The Formative Years 11

brilliant. In July 1806, he progressed from the second to the first division, being
ranked thirteenth of the 25 students who had chosen the Ponts et Chaussees as
their public service speciality. Given the military discipline under which the
Ecole functioned and the promiscuity in the barracks,. Augustin-Louis no
doubt experienced some difficulty in adjusting to life at the Ecole. Everything
was in sharp contrast to the warm, easy comfort oflife in the rue de Toumon.
Here, the young man, who had been brought up so carefully and in such a
devout way, now found himself among other young people who were not only
older but also loud, flashy, libertine, and irreligious. Fortunately, the second
year was better (33); and Augustin-Louis asserted himself, so that in October
1807 he was admitted to the Ecole des Ponts et Chaussees. He now ranked first
of the 17 students entering the Ponts et Chaussees and was ranked third in the
entire student body. Even though he showed himself to be a very good
technician per se, he was, nevertheless, not a very exceptional all-around
student of engineering sciences. In fact, his special talent for solving geometric
problems caused some comment among his professors. For example, in July
1806, he solved a problem requiring the determination of the lines of maximal
slope (34). That same month, Hachette published in the Correspondance sur
['Ecole Poly technique, a periodical he edited, an elegant proof by Cauchy of an
important theorem by Monge: 'If a surface whose equation is of degree m is
touched by a cone, the curve of contact of these two surfaces lies on another
surface curve of degree m - l' (35). In the same periodical, Hachette also
published a resume of a report by Cauchy on the problem of constructing a
circle tangent to three given circles (36).
When he left the Ecole Poly technique in October 1807, Cauchy was barely
18 years old. To judge by the comments of the administration, he was of
average height and had light brown hair and gray eyes (37). On several
different occasions, his fragile health had been grounds for various special
permissions. It would appear that Augustin-Louis was the victim of illnesses
that doubtlessly resulted from overwork, but the real source of the problem
seems to have been a kind of nervous tension (38). At the end of 1807, he
enrolled in courses at the Ecole des Ponts et Chaussees. This ecole was housed
in the outbuildings ofthe Palais-Bourbon, occupied by the Ecole Polytechni-
que until its move to the College de Navarre in 1805. Its students were
commuters. They were divided into three classes according to their degree of
knowledge and not to a circle of study. Each year, they submitted their
works-construction projects and theoretical memoirs-for several compe-
titions, which were solemnly judged in the presence of the Minister of Interior.
As a result of these competitions and their practical work during the summer
months, the students obtained marks, called degres, on which depended their
passage from a given class to a more advanced one. The best of the students
could expect to be named engineers-aspirants after two years of study.
There were, at that time, three major professorial chairs at the Ecole des
Ponts et Chaussees: a chair for applied mechanics held by Prony (actually,
Eisenmann substituted for Pro ny, who was in service elsewhere), one for
12 1. The Formative Years

applied stereotomy and construction practices held by the engineer Louis

Bruyere, and one for civil architecture held by the architect Charles-Fran<;ois
Mandar. The students could also take courses in modern languages, first in
English and German only, and then, from 1806 on, in Italian, also (39). Cauchy
studied these languages and learned them sufficiently well to be able to read
scientific treatises in all of them. Courses at the school lasted for four months
each year, from December to March, the major portion of the students'
instructional time being spent on fieldwork in the countryside. During the
warm season, students in the second and third classes were often sent on field
expeditions far away from the capital.
Augustin-Louis proved himself to be a very brilliant student at the Ecole
des Ponts et Chaussees. In the competitions of 1807, he won four first prizes
(40). The library of the Ecole still keeps an elementary study on wheels written
by Cauchy during his first year. Its contents and objectives are indicated by the
title: 'Memoire sur les roues des voitures et les perfectionnements dont elles
sont susceptibles, sur les roues a voussoirs connues sous Ie nom de roues ala
Duboville, les veIociferes, etc'. Written before he was 20 years old, this work is
far removed from the future concerns of the great mathematician (41).
When he left the Ecole Polytechnique, Cauchy lived near his parents, at the
Palais du Luxembourg and in Arcueil. It is likely that he attended some
meetings of the Societe d' Arcueil, a reunion of young scientists studying about
Laplace and Berthollet, in their country houses in Arcueil, in 1808 and 1809. In
the spring of 1808, he was sent on assignment to the works then in progress in
the Ourcq Canal and the Paris Water System. By not sending him far from the
capital at this time, the administration was granting him an exceptional favor.
Cauchy was sent to a vast construction site that was under the direction of
the engineer Pierre Simon Girard, of the Ponts et Chaussees. Girard had been
on Napoleon's expedition to Egypt and was author of several treatises and
scientific works. His most important work was a treatise on the strength of
materials (42). Upon his return from Egypt, Girard had been appointed
director of the Ourcq Canal project by the First Consul. From the Ourcq
Canal, as it then stood, Girard had conceived the idea of constructing an
aqueduct, as well as a navigational canal. Though these twin projects seemed
impossibly contradictory to many, Girard had the Emperor's confidence and
in October 1807 was appointed director of the project with full authority to
proceed with the works. Napoleon attached great importance to the Ourcq
Canal project: this undertaking would modernize and beautify Paris, the city
that had now become the capital of Europe. Not surprisingly, then, Bonaparte
intervened directly on several occasions to ensure that the project was carried
on efficiently.
In 1808, work on the project was started inside Paris. Cauchy was closely
associated with this phase of the construction effort. First, Girard placed him
for four months under the direction of the engineer Pierre-Marie Thomas
Egault, responsible for the general survey of Paris (43). Later on, in December
1808, Cauchy took part in the construction of the Saint-Denis Aqueduct. On
 1. The Formative Years 13

this project, he supervised the unearthing of an ancient Roman highway,

which had been hidden for centuries beneath the cobble stones in the rue Saint-
Denis. At the end ofthis field assignment, Girard sent the Directeur General of
the Ponts et Chaussees a highly complimentary report on the young Cauchy:
The laying-out of the circle-aqueduct inside Paris was the first operation
on which this student worked. He was employed on the leveling work for
this aqueduct. He later worked on the drawing of land over which the
arched Saint-Laurent gallery would pass, from the circle-aqueduct to the
large drain. He prepared various plans under orders from and with the
direct instruction of M. Egault, and he did so with great enthusiasm, as
well as with all the accuracy that could have been desired.
The theoretical knowledge and understanding that he acquired at the
Ecole made it easy for him to carry out these operations, and I believe
that the exactness of these past performances is indicative of the
performance that he will give on similar assignments in the future.
After having assigned him, for nearly four months, to work on
surveying, laying out and leveling, I thought it would be to his advantage
to assign him to one of the large building sites. He was personally
charged with overseeing the construction of a portion of the rue Saint-
Denis Aqueduct and surveying for work done. In this assignment, he
again gave proof of his aptitude and enthusiasm. I should add in this
connection that he has made some useful observations on the procedures
that are ordinarily followed for quantity surveying of materials.
Finally, I directly charged him with the responsibility of raising the
original land of the site and an ancient Roman highway that we found
some two meters beneath the present paving stones. This operation,
which in a way was foreign to our work, has nevertheless inspired in him
the degree of interest that it deserved, and he has achieved a satisfying
result.
I now turn to your question as regards his conduct and use of time
while he is not on duty. It is enough that I tell you that as M. Cauchy's
family resides in Paris he has spent with them the little free time that he
has from his duties. It has seemed to me that his education and principles
have made it unnecessary until now that he be accorded the surveillance
and advice that must ordinarily be given to young people (44).
Cauchy's second year at the school proved to be as remarkable as the first
one had been. During this year, Cauchy was sent on his second field
assignment, again to the Ourcq Canal project, and then later on to the Saint-
Cloud bridge project, the latter under the direction of the engineer Beautemps-
Beaupre. Unfortunately, no record exists of Cauchy's service on the bridge
project (45). However, we know something about his achievements on the
competitive examinations of 1807.
In these competitions, Cauchy took first place in rational mechanics and in
mechanical engineering; he took second place in bridges and street engineering
14 1. The Formative Years

and in wooden bridge engineering (46). Meanwhile, on two other examin-

ations, navigation and stone bridges, Cauchy's studies were lost, and he could
not compete. These studies, rediscovered in Prony's private papers, are kept
today in the library of the Ecole des Ponts et Chaussees (47). The paper on
navigation was divided into 2 distinct parts. The first part contains 11
statements and observations, accompained by various calculations in which
the statements and observations were subjected to a mathematical treatment,
and the second part applied the analysis given in the earlier part to the
particular problems of navigating the river Marne. The most impressive
feature of this paper is the quality of its exposition (48).
Two other papers discuss problems involved in constructing stone bridges.
The first of these studies presents an elementary discussion of construction
problems (49). The second study, which is the more interesting, investigates the
problems connected with constructing bridges with vaulted arches (50). In the
first part, Cauchy examined the relevant theories that had been developed by
various authors, such as La Hire, Couplet, Coulomb, and Bossut. In the
second part, he compares these theories and the results they gave rise to with
experience. The third part of the paper is analytical in its approach. Here,
Cauchy sets forth the general equation for the equilibrium of an arbitrary arch
and then applies his general results to several particular types of arches. Of all
the papers written in 1809, this paper on vaulted arches was by far the most
indicative of the future. Indeed, while he was at Cherbourg in 1811, Cauchy
briefly considered resuming his research on this topic (51).
Given his record and achievements, Cauchy only needed two years at the
school to complete his studies. Thus, in January 1810, he left the institution,
brilliantly ending his apprenticeship in engineering. When he took leave of the
school, he not only had a solid scientific background, but he had also
developed a definite personal, religious, political, and intellectual outlook.
The Cauchy family was very pious, and we have seen the care that
Augustin-Louis took in preparing himselffor First Communion. On the other
hand, the vast majority of the students at the Ecole Poly technique were fairly
indifferent to religion, to faith per se no less than to practices; there, the
dominant attitude among the students was one of liberalism and anti-
clericism. In the meantime, however, the Catholic Renaissance had taken roots
among the urban middle classes, which were still largely devoted to the
philosophical ideas of the preceding century. This movement had slowly and
almost imperceptibly begun to make itself felt at the Ecole.
During the summer of 1805, a few weeks before Augustin-Louis entered the
Ecole Poly technique, a young student at the Ecole des Ponts et Chaussees was
appointed repetiteur adjoint (assistant tutor) in analysis and mechanics,
replacing Matthieu, the incumbent holder of that position. The newly appointed
repetiteur adjoint, a certain Paul-Emile Teysseyrre, was an ardent Catholic
member of the Congregation de la Sainte Vierge.
The Congregation, founded four years earlier, in 1801, by a Jesuit priest
named Father Jean-Baptiste Bourdier-Delpuits, would bring 'young people of
 I. The Formative Years 15

AI,. p+a+1t = JI.",,'l, +Jit, I

3L
J ,:)v·
If,.. a+R.
Jl JL ,
.'L /", o>r I'" "t.
j' "Y"'Y'"!}

Second Memoire sur les Ponts en Pierre. Theorie des Voutes en Berceau, by Cauchy.
Manuscript, 1809, EN PC Library, ms 19. Published by permission of the Ecole
Nationale des Ponts et Chaussees
16 1. The Formative Years

good families' together in prayer and unite them against the menace of the
current faithlessness, irreligion, and secularism (52). Its members met every
two weeks in the rue Saint-Guillaume. They were under a duty to render each
other mutual assistance and charity according to the Congregation's motto:
cor unum et anima una. In the beginning, the membership of this society
numbered only six youths, but it grew rapidly, so that by 1804 it could claim
198 members. Young ecclesiastics flocked to its meetings, as did many young
aristocrats, often scions of the best families, and a considerable number of
young students, such as Hennequin in law, Laennec in medicine, and
Teysseyrre in the sciences. Father Delpuits' aim was to have members of the
Congregation scattered liberally about in positions of importance in society
and in government. He was particularly desirous that his youthful adherents
would be among the intellectual elite.
In a very natural way, then, the Congregation undertook to infiltrate the
Ecole Polytechnique, a veritable hotbed of secular minded scientists and
liberals. The infiltration of the Ecole was Teysseyrre's mission. Teysseyrre was
as busy trying to convert the souls of his young charges as he was trying to
educate their minds. He gathered about himself a number of Catholic youths,
among them Augustin-Louis Cauchy. Teysseyrre was an unusual person with
a strong, ardent personality, who soon converted young Jacques Binet into a
Congregationalist. Jacques Binet, who was Cauchy's personal friend and who
would later become a notable mathematician in his own right, had heretofore
been a complete atheist. Similarly, Teysseyrre exerted a strong influence on the
young Lamennais and left a lasting impression on Cauchy, even though
Teysseyrre would be at the school for only one year. For his own part, Cauchy
became a Congregationalist a few months after he left the Ecole on April 3,
1808.
Little by little, this ostensibly religious society spread its influence and, in
due time, became suspect in the eyes of the authorities. As earlier noted, the
Congregation had begun as no more than a prayer group, a purely religious
organization that no doubt would have been left unmolested by the State had
it remained what it was at its inception. But, that was not to be, and the
Congregation increasingly became a gathering of young people who harbored
royalist notions and who were, therefore, opposed to the Empire. The growing
quarrel and bitterness between the Papacy and Napoleon Bonaparte pushed
the Congregationalists into determined opposition to the Empire. Cauchy's
father, of course, had been a devoted royalist. But once the Great Revolution
had swept away the Old Regime, Louis-Fran<;ois had consistently played a
very prudent hand under the Directorate, and when Bonaparte staged his
stunning coup d'etat of 18 Brumaire, Louis-Fran<;ois had promptly rallied to
Bonaparte's cause. Furthermore, his position as Secretary of the Senate and
the verses he had occasionally written in praise ofthe Emperor now bound the
elder Cauchy more closely than ever to the present regime. However, being
billeted at the Ecole Poly technique had the effect of freeing the younger
Cauchy from any such concerns. Furthermore, the repressive measures taken
 1. The Formative Years 17

by the government against the Papacy and against anyone in France who
dared support it made the Empire, the reign of the Usurper, become more and
more odious to Augustin-Louis. By the time the Congregation managed to
resume its meetings in 1813-this time very discreetly-Cauchy had returned
to Paris from Cherbourg and was once again able to take part in these almost
clandestine activities.
The four years spent at the Ecole Poly technique were undoubtedly of
decisive importance to the intellectual development of young Augustin-Louis
Cauchy. Not only had he mastered a very impressive amount of mathematics,
but he had also acquired a set of political and religious convictions to which he
faithfully adhered for the rest of his life. Unfortunately, we know little about
the influential events that took place during this period. For example, did he
already have plans to embark on a career in science? We do not know for sure.
However, the enthusiasm he showed upon leaving Paris to work in Cherbourg
at the beginning of 1810 rather suggests that even then he had dreams of a
brilliant career in engineering, a career that had been made altogether possible
by his successes at the two engineering institutions.
 Chapter 2

Sojourn at Cherbourg

Having completed the required two years at the Ecole des Ponts et Chaussees,
Augustin-Louis Cauchy was appointed a junior engineer (aspirant-ingenieur)
on January 18, 1810, by Count Mole, the Directeur General of the Ponts et
Chaussees (1). Chosen from among the engineering students who had
completed their program of study, the 15 appointees were given all the duties
and responsibilities offull-fledged engineers, except for the title and the pay of
an engineer. A few weeks after his appointment, in February 1810, Cauchy was
assigned to Cherbourg, where he was to assist the engineers responsible for the
excavation and construction of Port Napoleon (2). Placing a copy of Laplace's
Mecanique Celeste and a copy of Lagrange's Theorie des Fonctions
Analytiques at the bottom of his trunk, he barely had time to pack before
he had to leave on the stagecoach for Cherbourg. Copies of Virgil and the
Imitation of Christ under his arm, he left for his first real position in March (3).
There was the excitement of new experiences and a sense of expectation: the
reading of the two classics that he had brought along; the meditations and
profound thinking to which he could abandon himself with such ease; the
pleasant excitement of discovering the countryside, and the positive, confident
reflections stirred his naturally inquisitive mind. The conversation he had with
his fellow travelers during the long stagecoach ride and the exciting diversity
of France's regional cultures and architectural styles could not but have helped
to relieve the sadness of leaving home. Perhaps he would be away from his
family, from his parents, for a long time. Now, for the first time in his life, he
was leaving the city of Paris for the provinces.
Could he measure up to the confidence that Count Mole had shown in him
by assigning him to Cherbourg, the site of the largest and most important
construction project in all the Empire? A vast engineering effort under the
direction of some of the most outstanding engineers that the Ecole des Ponts et
Chaussees had produced! For more than 30 years, there had been a fierce
desire to construct a first-rate naval facility there. As the inspecteur general of
the Ponts et Chaussees, Joseph-Marie-Fran~ois Cachin, noted in 1803:
18
 2. Sojourn at Cherbourg 19

There has always been a deep-felt need for a Channel port that would
guarantee a place of safety for our naval vessels. But, it was especially
during the disastrous era of the Battle of La Hougue that the
government realized the necessity of establishing a great naval facility for
France on this region of the coast. Since then, the roadstead of
Cherbourg, which is at the edge of the Peninsula known as Cotentin, has
seemed to be the most favorable place for such a facility. The fact is that
its forward location on the sea lanes through the Channel offers
everything we need either for maximum surveillance of the enemy or for
harassing its convoys or, finally, for assembling all the details necessary
for a major military expedition. The Cherbourg roadstead, with its
excellent anchor hold, is equally favorable for the arrival and departure
of vessels, no matter the winds and tides. This location has vast
anchoring and docking space and is susceptible to all kinds of methods
of attack and of protection and defense. Finally, by all military and
maritime accounts, this location has all the advantages that can
influence the fortunes of our naval forces and commercial relations
(4).

On June 23, 1786, in the presence of King Louis XVI, construction began on
an enormous dike enclosing the roadstead from the seashore. This project,
begun some three years before the outbreak of the Revolution, was completed
in 1806. During the Revolution, an arrangement was envisioned whereby
within the Cherbourg roadstead there would be a port of careenage and vessel
refitting separate from the commercial harbor; and, under this arrangement,
there was even the possibility that there would be an arsenal. However, it was
not until the establishment of the Consulate that it was decided on April 15,
1803 to construct a naval base at Cherbourg, an entirely new port that would
be distinct from the old city. The various projects as originally outlined by
Joseph-Marie-Frant;ois Cachin comprised 'a forward port, as well as a port
capable of containing a dozen warships and a proportional number of frigates,
along with the dry docks, no less than all the fortifications necessary to shelter
the port from the enemy' (5). A little later, on September 26, 1804, the First
Consul personally decided that 'the plans be modified so that a huge naval
facility capable of holding thirty front-line vessels that would be kept in a
constant state of readiness' could be built around the inner basin, according to
Cachin's plans. However, this phase of the work was not completed until after
1830.
Cachin began construction of the new harbor, which was first called Port
Bonaparte and then Port Napoleon, at the end of April 1803. There was much
to be done. In fact, it was necessary to dig into rock in order to construct the
forward harbor and refitting docks with sufficient depth to maintain a fleet at
low tide. At any given time, there were 2000 to 3000 laborers working on this
construction site; some were soldiers, but by far the greater number were
prisoners of war. Conditions were so bad at the Cherbourg project that some
20 2. Sojourn at Cherbourg

20,000 prisoners died during the 15 years of work; and, according to Cachin,
about 2000 died on the job (6). Clearly, so vast an undertaking required that
new procedures be developed and original techniques devised. Thus, for
example, two steam machines were installed to work continuously at draining
off the seepage (7).
By the time Augustin-Louis was assigned to Cherbourg, work on the
forward port had already been carried forward to an advanced state. Cauchy
had barely arrived in Cherbourg before he became involved in the direction of
the construction site; and, it would seem that he found this responsibility to be
in no way disagreeable. In fact, on June 8,1810, he wrote to his father that 'the
project at Port Napoleon is more and more important and my job assignments
are very instructive' (8). For his own part, Louis-Fram;ois, from afar, remained
ever watchful of his son's future and, on July 9, 1810, asked Count Mole to
allow Augustin-Louis to remain at Cherbourg since '[he] found the work
there to be very instructive' (9). On December 13, 1810, an imperial decree
appointed Augustin-Louis a second-class engineer-ordinary attached to the
project at the port of Cherbourg.
Throughout his sojourn at Cherbourg, Augustin-Louis worked on almost
all phases of the project: the excavation of the basins for the military port by
blasting through schistose rock (10); the construction of the two piers that
formed the entrance to the port and the construction of the encircling dikes
joining the one on the north to Fort du Hornet; the draining of the water that
seeped into the harbor across the cofferdam that temporarily barred its way;
the construction of a row of signal posts along the shore; and the building of a
lock for the dike enclosing the roadstead. These were all considerable
undertakings for the young engineer. But, in addition to these, he devoted time
to various other construction projects, such as shelters for the prisoners;
barracks for the troops on the site, forges, and covered buildings for the granite
piers on which the future arsenal would be located. Cauchy described his work
during those days at Cherbourg in a letter cited by Valson. Dated July 3,1811,
the letter reads:

I get up at 4 o'clock each morning and I am busy from then on. My usual
work load has been increased over the past month by the unexpected
arrival of Spanish prisoners of war. We were told of their coming only a
week before they arrived, and within those few days' time, we had to
build lodgings and bunks for 1200 men. The buildings in which these
prisoners are housed consist oftwo structures, each containing 19 rooms
for a total of 38. When the prisoners arrived, 12 rooms had already been
prepared, and we had to build another 12, the walls of which had not
even been completely raised. We had no roof tiles for these buildings, and
I had to visit the local quarries to try and get some. We had no kitchens,
and I had to put together some temporary stoves. At last, two days ago
all the Spaniards were set up in lodgings. They have bunk beds and straw
mattresses and are given food. They count themselves as being quite
 2. Sojourn at Cherbourg 21

well-off. I had no sooner finished the job oflodging the prisoners of war
than I had to get busy on another assignment. Today, I drew the plans
for forges that I am to have built in granite. I am also constructing two
lighthouses, one on each of the two piers that are located at the entrance
to the harbor.
I do not get tired of working; on the contrary, it invigorates me and I
am in perfect health ... (11 ).

Augustin-Louis' performance and enthusiasm for his work certainly

pleased his superiors. His real scientific investigations-investigations that
were begun at Cherbourg and as we will see, represent his first breakthroughs
in this area -had the effect of raising him in the esteem of the other engineers
on the Cherbourg project. He seems to have been presented to the Emperor,
with a very flattering introduction, when Napoleon and the Empress Marie-
Louise visited the Cherbourg construction site in May 1811. Be that as it may,
it is certain that after Cauchy had left the project Cachin addressed his former
subordinate in particularly warm terms when he informed him of the grand
opening of the forward port by the Empress on August 1, 1813.
So it is done, my young friend. The ocean has taken possession of our
works ... and, just think ofit - you were not even here! Your absence
made me feel rather out of sorts; however, hopes for your happiness lifted
my spirits. I hope that soon, when I get a chance to see you, it will be
possible to tell you of my wishes that you will never stop being a
member of a group that regards you as one of its most distinguished
members, a group among which you will always find true friends. I
myself am happy to be among that group and to offer you, no matter the
circumstances, testimony of my unalterable devotion (12).
At Cherbourg, Augustin-Louis not only had to supervise teams of men,
make decisions, endure the hardships of the construction site, and spend long
hours outside in all kinds of weather, he also had to learn how to move about
in polite society, how to form and cement the types of social relations his new
position demanded. Certainly, being barely 20 years old and unmarried, he
would have to establish a good, solid reputation for himself. But this would
not be easy for this rather frail young man, a youngster who, though he might
have the perfect drawing-room politeness and polish he had learned in the
salons ofthe Palais du Luxembourg, was nevertheless formal and too cold. On
the recommendation of his parents, he had been received at the home of
Marie-Charles-Cesar Fay, Count of Latour-Maubourg, then the leading
citizen of Cherbourg. The count had been appointed to the Senate by
Napoleon, with whom he had allied himself upon returning from his stay
abroad as an 'emigre'. Returned to France and in the Emperor's good graces,
Latour-Maubourg was now commander of the Cherbourg Military District.
Thus, without any effort on his own part, Augustin-Louis was introduced into
the small, tightly knit circle of notables in the port city. This privilege, an honor
22 2. Sojourn at Cherbourg

that many others might have been very happy to have, soon began to appear to
Augustin-Louis as tedious. A friend of Lafayette, the Count was a liberal and a
'philosophe', a man who was indifferent to religion and whose company
offered no comfort to the young Congregationalist. Cauchy's interest soon
flowed toward people who shared his piety, his studiousness, and his general
serious-mindedness:
The homes I visit in Cherbourg, and which I naturally should visit, are
those of my superiors: the homes of M. Cachin, of M. Franqueville, and
of MM. Duparc and Vallot. And I am well received at each of them. I
cannot get out of seeing M. L ... , to whom I am so obliged. At these
homes, particularly that of M. Franqueville, one meets people from all
the various groups in the city. Among them are some in which religion is
honored and respected; there are others, however, in which the only
thoughts are of amusement. I am very closely connected with some of the
former and the time I spend there is not wasted. I visit some four or five of
these homes. They are very closely connected to each other and to those
of MM. Cachin, Franqueville, and Duparc. I also associate with a few
other persons whose acquaintance should prove useful, such as M.D .... ,
a cleric now returned from England; M.G.... , the headmaster of the high
school; and M.V.... , who is very able in mathematics (13).
Quite a few people were antagonized by Cauchy's austere behavior and his
cool exposure of his religious sentiments. Moreover, the slowness he showed in
calling on those whose ideas he judged to be different from his own was
especially offensive. No young man should be so pensive and rigid. So, the
gossiping began. Finally, echoes of the criticisms that were leveled against him
reached his parents, probably through the offices of Count Latour-Maubourg,
and Augustin-Louis had to defend his conduct:

So they're claiming that my devotion is causing me to become proud,

arrogant, and self-infatuated. Exactly who is making these claims? Not
people who have much religion themselves; for the people that I have
talked with have all urged me to follow my usual code of conduct, and
everything that has been reported to me (about these people) tells me
that they do not blame me. However, a few days ago, a certain person in
the ... Society told me in a friendly way that religion often makes young
people self-infatuated. I talked with her for a while on the subject and
showed her that I was not self-infatuated. Now, as for those people who
have no religion themselves, I have decided never to discuss it at all with
them; and I will only reply (to them) when they attack me on the matter.
Thus, when I arrived in Cherbourg, M ... decided one day to tell me,
while we were talking about religious obligations, that I would soon get
over all such notions. Without getting upset, I told him that when a man
was engaging in wickedness and wrongdoing, he would do well to get
over it quickly; and then I asked him if he found anything in my conduct
 2. Sojourn at Cherbourg 23

that was wicked or harmful. Another person who was present overheard
what was being said and, being of my mind on the matter, took my part.
After making a bit of uneasy small talk, the gentleman ended his remarks
with a great deal of politeness and said nothing more to me about
religion. I am now left alone about religion, and nobody mentions it to
me anymore except to encourage me to continue with what I believe in.
There are a few philosophes who say that religion has made me self-
infatuated. But, for my own part, I am truly happy that in a country
where one hears so much gossip and where certain people keep busy
slandering and picking from dawn to dusk, they have not found other
grounds upon which to reproach me (14).
Augustin-Louis was basically a solitary person, a man who was more
attracted by nature's spectacle than by the doings of human beings. Whenever
his work left him time to do so, he would devote himself to such thoughts,
thoughts that would later prove to be the source of his most abstract
mathematical discoveries. But, right now, of course, Cauchy's meditations on
nature were quite far removed from any kind of learned scientific investig-
ations, rather they were indicative of his own personal style. In a letter to his
parents, he stated:
From time to time I get a moment to relax, and I use it to make little
walks around Cherbourg. There are some very scenic spots hereabouts,
one of the most picturesque being the Quincampoix Valley. There is a
small river that twists its way through this valley and on whose banks are
several factories. The river winds its way across a magnificent prairie,
which is filled with flowering apple trees. The clear waters of this little
stream can only be seen through a dense growth of green bushes and
shrubs, which seem to form a kind of cradle about the river. On one side,
the meadow is bordered by great patches of fern and plants of every kind;
on the other, by immense rocks and cliffs whose peaks rise sharply into
the air or hang suspended over the dale.
Cauchy, of course, was not content to observe the sights ofthe surrounding
countryside in a passive way. He wanted to understand, to explain, to classify.
So, since his father had taught him much about plants during his childhood, he
began to collect various plants. Changing its tone, Cauchy'S letter describing
the countryside around Cherbourg continued:
Among the plants that are found in abundance in this part of the
country, I have noticed foxglove, white orchids with tiger stripes, orchids
with red petals, various thick-leaved plants, and a great deal offems, called
polyp ode, which cover all the walls of the city like a hood. There is a
treelike fern with shiny, varnished leaves that split up into many smaller
ones; there is another fern that looks much like the one I just described,
except that it is not treelike; there is still another species that resembles
the two preceding ones except that it is rather short; then another kind of
24 2. Sojourn at Cherbourg

fern grows in a rampant, random way; and, finally, there is the type a few
leaves of which I sent to you in an earlier letter. I must also tell you that
while walking I have come across some uncommon insects, such as the
'cardinale' [red dragon fly?] and the blue stag beetle. So, you see, I hardly
need leave the city in order to have all the delights of the countryside (15).
When he described to his parents how the sea looked during a storm,
perhaps thinking of the awful tempests that had twice damaged the great dike
at Cherbourg-once on November 2,1810, and then shortly thereafter on the
night of November 10-11, 1810-he sought to give an analytical explanation
of how the billows swelled and rolled as they approached the shore rather than
to present a strictly visual portrayal of the motion.
Vernet's pictures can give some idea of how it looks when it is calm,
but cannot possibly describe it when it is disturbed. They cannot show
how the wave, after having dashed against the rocks, angrily withdraws,
receding only to come again to the point whence it had departed, more
furious and more terrible than before. They cannot show how, at the
meeting of swells that have been hurled back by the rocks with those
rolling in from the open sea, each particular wave rises up and then
smashes against the shore, where it leaves a long trail of foam (16).
This way of looking at things, essentially the scientist's way of viewing nature,
was presented more fully in a lecture that Cauchy gave before the Cherbourg
Academic Society on November 14, 1811 (17). In this address, he used bold
strokes to paint the portrait of a thinker, a scholar-scientist, who is trying to
penetrate the secrets of nature:
He fixes his gaze on the earth on which he stands and maps out a design
of it with the same ease that he would use in mapping out a design for his
own garden. With a sure and steady hand, he sketches the courses of the
rivers and the contours of the oceans. He measures the heights of the
mountains and the depths of the abysses. He notes the plants that cover
the globe and the beasts that dwell among them, from the moss that
grows on stones to the Cedars of Lebanon, and from the delicate little
shellfish scampering along beneath the seas to the gigantic elephant
possessed of a tremendous strength. Even the bowels of the earth seem
no longer to posseses anything that can be kept hidden from him. He
questions. Armed with mining tools, he sets forth to discover the. hitherto
unheard-of minerals that lie within the bowels of the earth. He lifts his
gaze from the land to the sky. His thoughts soar and embrace the general
system of the universe. He marvels so as thus to articulate the secrets of
nature. From the spot where he is placed in the great vastness of the
world, he measures distances, though he himself cannot run their course.
He plots the paths of the stars that surround him and studies their
influence on the seasons and on the climes and on the tides. On his scales,
 2. Sojourn at Cherbourg 25

he weights the moon and the planets and then tells the comets the days
on which they shall once more visit them.
Clearly, as the little of his address suggests, he was trying to give a precise
definition of the 'view of things', which 'measures' and 'marvels so as thus to
articulate the secrets of nature' in order to determine their proper limits:

If we observe that all our intelligence and all our skills and devices are
contained within certain bounds that cannot be cast aside, we can
convince ourselves without too much trouble that our knowledge is as
limited as our senses.

Although the bare words of Cauchy's address might ostensibly suggest that he
was giving an impersonal detached description of the 'general' scholar-
scientist, we suspect that he, a heretofore almost unknown young engineer,
had completely dedicated himself to science and was, in fact, describing himself
to his audience.
Whatever may have been his interest in the natural sciences, he spent his
spare time at his desk studying the exact sciences. As earlier noted, Cauchy had
taken Laplace's Mecanique Celeste and Lagrange's Theorie des Fonctions
Analytiques to Cherbourg with him. Other works were av.ailable to him either
through his father who sent them from Paris or by borrowing them from
various libraries in Cherbourg. At first, he had planned to 'make a coherent
study of all branches of mathematics, starting with arithmetic and proceeding
to astronomy, clearing up obscure points as well as possible, working on
simplifying proofs, and trying to discover some new propositions' (18). Soon he
became involved in his own research. On Lagrange's advice (19), he first
attacked a problem in pure geometry that had been posed two years earlier by
Poinsot.
In a paper presented to the Institut on July 24, 1809, Poinsot had
established the existence of three new nonconvex regular polyhedra: two
dodecahedra and one icosahedron. These were now added to the five regular
convex polyhedra that had been known since antiquity and to Kepler's star-
shaped dodecahedron. But Poinsot had been unable to show that no other
regular polyhedra existed. 'This', he wrote, 'is a question that deserves to be
investigated; and it seems to me that it is one that will not be easy to solve in a
fully rigorous way'. The Institut received Cauchy's study, 'Recherches sur les
polyedres', on February 11, 1811 (20). Employing a development that
generalized the methods Poinsot had used in constructing all the star-shaped
regular polygons and his three nonconvex regular polyhedra, Cauchy dealt
with Poinsot's problem in the first part of his study. The second part was
devoted to the Euler Formula
S+F=A+2 (2.1)
on the number of faces F, edges A, and vertices S of a polyhedron. Cauchy
26 2. Sojourn at Cherbourg

pointed out the equivalence of the Euler formula Eq. (2.1), to the relation
S + (F - 1) = A + 1 (2.2)
between the number of sides A and vertices S of the network of F - 1 polygons,
which can be obtained by cutting out one face of the polyhedron and
projecting the other faces on the cut face. Thus, from a proof of Eq. (2.2), he
deduced a proof of the Euler formula Eq. (2.1). Then, Cauchy generalized the
Euler formula to the case of a network of polyhedra. He obtained the relation
S+F=A+P+l (2.3)
on the number of polyhedra P, faces F, edges A, and vertices S of the network;
the Euler formula is given by the special case P = 1 in Eq. (2.3). The
commission whose responsibility it was to evaluate the study was composed of
Legendre and Malus, and its report on Cauchy's research was very favorable.
'The proofs', observed Malus' report on May 6, 1811, 'are rigorous and
developed in an especially elegant way' (21).
Acting on the advice of Legendre and Malus, Cauchy undertook further
studies on polygons and polyhedra, and on January 20, 1812, he presented
another paper to the Institut (22). In the first part of it, he established eight
theorems on the variations of the angles of rectilinear and spherical convex
polygons. These basic results and the Euler formula, Eq. (2.1), were used in the
second part of the study to give a proof by reductio ad absurdum oftheorem 11
in Book 9 of Euclid's Elements. According to this theorem -a result that had
not theretofore been proved-two convex polyhedra whose faces are equal
and are similarly placed are necessarily equal, either by symmetry or by
superposition; consequently, a polyhedron with rigid faces is completely rigid
(23).
This last result made a strong impression. At first, Cauchy had a
somewhat difficult time of it in trying to convince Malus of the validity of his
proof by reductio ad absurdum.

If M. Malus seemed not satisfied with the proof I sent to you [wrote
Cauchy to his father who was acting as intermediary] it probably has to
do with the fact that you did not advise him of what I had taken care to
tell you; namely, that my proof rests on several lemmas that are easy to
prove. It does not, therefore, surprise me in the least that M. Malus has
concluded that I assumed things that could not be assumed. But, that is
not the question: ifl had the time, I should have sent you the proofs of the
lemmas I used. Today, I will reduce the question down to the matter of
knowing whether or not my proof is acceptable, assuming the lemmas
are established. As to the form of proof that I used, I think it would
be not only difficult to change it, but downright impossible. The reason is
that until now a geometric argument has not been given, except in terms
of reductio ad absurdum, of the theorem that in 2-dimensional geometry
is analogous to the one in 3-dimensional geometry that I dealt with: I
 2. Sojourn at Cherbourg 27

./
0/,,- It.I.

} I )
.'1 l ':. .. -IlJl
c"~4.b...--,,~ ,·,..7_ .... ~t. _."4'.....

Title page of the manuscript of the paper on the polygons and polyhedrons. January
20, 1812. Published by permission of the Ecole Poly technique.
28 2. Sojourn at Cherbourg

mean the theorem by which it is proved that two triangles are equal if
their three sides are equal. If one should establish this latter theorem
without using either trigonometry or reductio ad absurdum, I would
agree that my proof ought not be admitted. It thus seems impossible
to banish the reductio ad absurdum proof from geometry; and this is
particularly true in the present case. In fact, in order to prove that
under certain conditions only one polyhedron can be constructed, it is
necessary to see that after the first figure has been constructed subject
to the given conditions then one cannot construct a second figure
without encountering a contradiction. I insist on this argument because
the type of proof I gave seems to me to be inherent in the nature of
the theorem in question. Moreover, it is precisely what M. Legendre
used in establishing several particular cases of the same theorem (24).
Malus died on February 24, 1812, and thus did not participate in the
commission that evaluated the study. This commission, composed of Legen-
dre, Carnot, and Biot, gave a very glowing report on February 17, 1812.
Written by Legendre, the report concluded:
We wanted to give only an idea of M. Cauchy's proof, but have
reproduced the argument almost completely. We have thus furnished
further evidence of the brilliance with which this young geometer came
to grips with a problem that had resisted even the efforts of the masters of
the art, a problem whose solution was utterly essential if the theory of
solids was to be perfected (25).
Cauchy's first research on polyhedra demanded more mathematical virtuosity
than mathematical knowledge per se. Strangely enough, he seems not to have
attached much importance to mathematical works of this kind, works that
were destined to make him famous. At his lecture on November 14, 1811, he
declared:
What can I say about the exact sciences? Most of them seem to have
already been carried forth to their highest stage of development.
Arithmetic, geometry, algebra, and higher mathematics are sciences that
can rightly be regarded as having been completed, as it were; and nothing
more remains to be done with them except to find new areas of useful
applications (26).
This opinion was widely held in France's scientific community at that time.
Most decidedly, the future seemed to belong to the engineering sciences (27).
Such pessimism could hardly have fired Cauchy with a desire to persevere
in the mathematical sciences. However, his research on polyhedra had given
him a reputation in Paris. On February 22, he was nominated a corresponding
member of the Societe Philomatique. Founded in Paris in 1788, the Societe
Philomatique was patterned after the old Academie des Sciences, with regular
members and correspondants. The society published the Bulletin, which
 2. Sojourn at Cherbourg 29

contained scientific information and summary papers of new works communi-

cated by its members. On March 21, Poisson, who was the mathematical
editor ofthe Bulletin, presented Cauchy's paper on polyhedra to the society. In
a report written by Biot and himself, Poisson made a recommendation
favoring Cauchy's nomination to the Societe Philomatique (28). But, Cauchy
aimed far higher. His object was the Institut itself. After all, had not Legendre
himself declared, following the research on polyhedra, that Cauchy should get
a seat in the First Class of the Institut (29)? To reach such a goal, however, he
needed another spectacular result, something like his proof of Euclid's
theorem. According to a letter Louis-Fran!;ois wrote to his son, the young man
should undertake investigations into Fermat's theorem on polygonal num-
bers. Addressing his son, Louis-Fran!;ois wrote:
Your last paper on polyhedra made a deep impression on the Academie.
If you prove one of Fermat's theorems, the way will be wide open for you.
The moment is favorable for you. Do not let it slip by (30).
So, Cauchy began work in this area, work which would last for several more
years.
The third research paper Cauchy sent from Cherbourg dealt with a quite
different area of mathematics, the theory of the directrixes of conic sections.
Unfortunately, this work was a failure in the sense that it was neither published
nor read to a learned body. The reason for this seems to have been that this
paper contained no new results at all; and we only know of its existence by way
of a letter written by Augustin-Louis. Cited by Valson, this letter is undated,
but was no doubt written sometime during the summer of 1812:
It would seem that all my theorems have the good fortune of being found,
on the first reading, to be either false or already known. But, might it not
be with the result on the directrix of an ellipse as it was with the theorem
on polyhedra? Though I lost my argument the first time, is it not possible
that if another look be given to the matter, the theorem will indeed be
found to be a new result? In any event, do not think that I attach great
importance to that theorem. However, as I have gone through all the
books on second-degree curves and have found nothing in them, I at
least ask that you should find out from M. Poisson the title of the work
where the theorem I discussed can be found (31).
Augustin-Louis was already contemplating his return to Paris when he wrote
this letter, for he added:
Moreover, I have some other, more interesting, theorems to work on;
and I already have the material for one or several papers that I will write
when I get back to Paris.
He had to return to Paris if he was to continue his research under favorable
conditions and become situated in a position in the educational establishment
as a professor, examiner, or academician. Such a position would leave him
30 2. Sojourn at Cherbourg

enough time for his mathematical pursuits. Besides, everything was happening
in Paris; Paris was not only the political capital of the Empire, it was the
scientific and intellectual capital of Europe.
From his father's connections with leading people in the capital and his
growing reputation as a mathematician, Cauchy could well expect to get a
position as an engineer-until something better came along-or as a teacher
somewhere. But, matters did not wait for the changes he was contemplating.
Badly overworked, he fell ill. He was no doubt physically exhausted by the
harshness of the climate and the working conditions at the construction sites.
But, even more than that, he had developed an acute case of strained nerves.
The intensive intellectual activity, the long evenings at his desk piled high with
books and papers, and the hours he spent tutoring youths who wanted to go
on to the various schools of special study (32) served to undermine his morale,
as well as his physical health. That he should have become exhausted at this
time is all the less surprising when we consider the importance of the research
work that he undertook while at Cherbourg, a place where conditions
approached isolation, and, in the meanwhile, still managed to conduct his
engineering work at the highest level. His research, of course, included not only
those investigations that had been presented before the various learned bodies
of the time, but also studies that he had undertaken on permutations and, no
doubt, some ofthe material that, in the years to come, would go to make up his
treatises on algebra and analysis.
We have no information about the exact nature of Cauchy's illness, and
Valson's biography is unusually vague on the subject, mentioning only that
'Cauchy's health had never been particularly robust', and when his mother
came to Cherbourg after he had taken ill, she 'found him in a weakened and
depressed condition, which threatened to cut short this very promising career'
(33). One point, however, is a letter dated January 12, 1814, in which Louis-
Franc;ois declared to Baron Costaz, the Directeur General of the Ponts et
Chaussees, that his son's health was 'profoundly altered by the heavy burdens
he had to bear during the three years he spent at Cherbourg' (34). Both
remarks are somewhat misleading, for the illness seems to have been as much
emotional as organic, with psychological overtones that were as pronounced
as the physical ones. All available information on Cauchy's personality tends
to support this hypothesis. His cold, aloof, and inflexible personality was poor
cover for his deep sensitivity. At once easily hurt and hotheaded, he was a
loner and a person who pushed himself excessively. In his personal life, as in his
work, he was both volatile and stubborn. Moreover, the view that this illness
had psychological underpinnings is also supported by certain features of the
malady itself. An ill-defined sickness, it lasted, off and on, for about a year. This
long period of convalescence was periodically interrupted by a few weeks
during which Cauchy was able to work. Numerous other facts point to the
conclusion that throughout his life Augustin-Louis Cauchy suffered from a
nervous condition.
 2. Sojourn at Cherbourg 31

During the summer of 1812, his condition gradually grew worse. His
parents were greatly disturbed by his health. Towards the middle of
September, they received a letter from Cauchy's father confessor, informing
them that their son was planning to be married. Madame Cauchy, by now
very anxious, immediately left Paris with her second son, Alexandre. She
arrived in Cherbourg on September 24th and, finding her son in very poor
health, took him back to the rue de Tournon (35).
Altogether, Cauchy had stayed in Cherbourg for close to three years. During
his time there, he had learned to work as an engineer and had performed
his assigned tasks excellently. But, his particular disposition, interests, and
abilities had, by now, inclined him in another direction. Burdened with poor
health and always more attracted by abstract questions than by concrete
problems, he seems to have become increasingly disinterested in the goings-on
at the construction site. At the same time, however, he was putting more and
more of himself into his mathematical studies and had won his first laurels. He
used his spare time to broaden his knowledge of pure science. A good deal of
fruitful thoughts and ideas probably took root at this time, although it would
still be a number of years before they would be fully developed and written in
treatises and papers for posterity. During the years in Cherbourg, he also made
his first scientific discoveries. They were sufficiently important to attract the
attention of learned society back in Paris. Henceforward, the die was cast: his
whole life, to use one of his favorite expressions, would be devoted to 'the
search for truth'.
 Chapter 3

The Waiting Room of the Academy

At home in Paris, back with his family, Augustin-Louis recovered his health.
What he seemed to have needed most was the affection, closeness, and
understanding that were typical of the Cauchy family. During this period, he
worked contentedly on his new papers. First, he perfected his two-part memoir
on symmetric functions, which he presented to the First Class of the Institut on
November 30,1812; the memoir was published in two articles in the beginning
of 1815 (1).
In this important study, Cauchy created the calculus of substitutions, a
topic that he would pursue in much more detail during 1845-1846 and would,
in the years to come, playa major role in the development of group theory.
Interestingly enough, in 1815, Cauchy seems to have had no real interest in the
problem of the solvability of equations by radicals, although this avenue of
investigation was the mathematical framework in which mathematicians had
regarded permutations. Cauchy began the investigations that were to become
the two-part study while he was in Cherbourg. At that time, he was working on
number theory, especially on Fermat's theorem on polygonal numbers; and
this research was to continue until 1815. In connection with his work in
number theory, he made a careful study of Gauss' Disquisitiones Arithmeticae,
which had been published in 1801 and translated into French in 1807. He
seems to have mastered very quickly the methodology and sense of Gauss; and
grasping the importance of the theory offorms that Gauss had used in proving
Fermat's theorem on triangular numbers, he managed to simplify it and to
generalize some of the results, particularly the ones on discriminants.
These investigations on number theory, he declared in the first article, led
him to work on the 'theory of combinations' and from there to prove a
theorem that was more general than the results obtained by Lagrange and by
Ruffini on the number of values of a function of n given quantities: namely, if
the number of distinct values assumed by a function of n quantities is less than
the largest prime factor p of n, it is less than or equal to 2.
32
 3. The Waiting Room of the Academy 33

JOURNAL
DE L'ECOLE POLYTECHNIQUE .

MEl\10IRE
SUR Ie Nombre des Valeurs fju'U/lC FOIlClioll ptllt tlcfjuerh', lorsfju 'Oil Y
pel7lulI& de lomlS les 1I1anj(r~s possiblts Its qlltlluirts qll'dle ul:ftrllle;

PAR A. L. CA UCHY. INGE ' IEUR DES PO . TS ET CH.WSSEES.

MM. LAGRANGE et VANDERMONDE sont. je crois. Ics premiers qui

aicnt consider': les fonctions de plusieuJ's variables l'elalivement 'au
nombre de va[eurs qu'elles peuvent obtenir. lorsqu'on substitue ces
variables it. la place les unes des autres. IIs ont donne plusieurs thco-
remes intcressans relatifs it. ce sujet. dans de.LIX Il1cll1oires impril11es ell
J 77 I • l'un it. Berlin. ['alltre it Paris. Depuis ce temps. quelques gco-

metres haliens se sont occupes avec slIcces de cette ll1atii:re, et parti-.

cuiierell1ent M. Ruffini, qui a consignc: Ie resultat de ses recherches
dans Ie tome XII des Memoires de la Societe italienne. et dans sa
Thcorie des eqll8tiolls numeriques. Une des :onscquences les pillS j'e-
marquables des travaux de ces divers gcomCtres, est qu'avec un lIombre
donne de lettres on lie peut pas tOlljours former une fonctioll qui ait
un nombre detcrmin6 de valeurs. Les caracteres ral' lesquels cctte
xm'~~ A

The calculus of substitutions, created by Cauchy in 1812. 'Memoire sur Ie nombre des
valeurs qu'une fonction peut acquerir, lorsqu'on y permute de to utes les manieres
possibles les quantites qU'elle renferme', Journal de I'Ecole Poly technique, 10, 1815,
pp. 1- 28. Published by permission of the Ecole Poly technique.
34 3. The Waiting Room of the Academy

In the proof of this result he introduced new concepts, notations, and

methods, directly inspired by Gauss' Disquisitiones. He denoted

as the substitution transforming the permutation Ar of n letters into the

permutation As of the same letters and
(lX,f3)
as the transposition of two letters IX and 13 in a given permutation. He defined
the product

to be the substitution

and used a method of proof that amounts to decomposing the group Sn into
disjoint classes according to one of its subgroups. In the first article, he also
gave a simple proof of Lagrange's theorem - the order of any subgroup of Sn is
a divisor of n! -and determined all the subgroups of a cyclic group (2).
In the second article, Cauchy developed the abstract theory of deter-
minants. First, he gave another result on substitutions: the decomposition of a
substitution into disjoint cycles and the distinction between even and odd
substitutions according to the number of disjoint cycles and, consequently, to
the number of transpositions the given substitution contains. He also proved
the existence of the alternating subgroup An of Sn. Then Cauchy introduced
the notion of the 'symmetric alternating function S( ± K)'. This function,
deduced from a given function K(a 1 , ... , an) of n given quantities, can be
written in modern notation as

aeGn

(G n is the substitution group on {1, ... n}).

Cauchy showed that

i<j
[S(±a~a~, ... ,a:-l) is a Vandermonde determinant] and that

S( ± a~a~, ... , a:) = a1a Z"'" an

i<j
TI
(ai - aj). (3.1)

He used expression (3.1) to define the determinants as follows: replacing

 3. The Waiting Room of the Academy 35

each exponent by a second subscript of the same value in the expansion of

S( ± a~ a~, ... , a~), he obtained thefunction S(± au, a 2,2"'" an,n), which is the
determinant (of nth order), denoted Dm of the 'symmetric system' (i.e., the
square matrix):

r'
al,2 al,n
a2,l a2,2 a2,n

an,l an,2 an,n

This symmetric system can be written more simply (al,n)' For instance, since
S(± a~a~) = a l az(a 2 - a l ), we have D2 = a U a 2,2 - al,2a2,l by replacing a l
by a l ,l,a2 by a 2,l' a 1 2 by a l ,2, and a/ by a 2,l'
Thereafter, Cauchy gave the principal properties of determinants: Dn is null if
two rows or columns of its symmetric system are equal; Dn changes by a sign if
two rows or columns are interchanged; the determinant Dn of a symmetric
system (al,n) is also the determinant of its conjugate (i.e., its transpose) (an, 1 ).
Cauchy proved the rule for the expansion of a determinant according to rows
and columns. The most important result, however, concerns the product of the
determinants: if(ml,n) is the resulting system of two component systems (al,n)
and (IXl,n) [i.e., if(ml,n) is the product of the two matrices (al,n) and (IXl,n)], then
the determinant Mn of (ml,n) is equal to the product of the determinants Dn
of (al,n) and bn of (IXl,n) (3).
The level of rigor and abstraction in his two-part memoir on symmetric
functions of November 30, 1812, is certainly the first full-fledged example of
Cauchy's mathematical style.
On February 1, 1813, the sick leave that had been granted to him came to an
end. However, Cauchy had no desire to return to Cherbourg and resume his
duties as an engineer. Desperate, he asked Prony, the director of the Ecole des
Ponts et Chaussees, for a position of associate professor. On February 3, the
Ecole's committee denied his request. Meanwhile, expressing a desire to see
him [Cauchy] devote himself more completely to scientific research, the
board 'urged' the central administration to assign him to an engineering
position in Paris (4). Given this supportive recommendation and, no doubt,
some other well-placed pressure from influential quarters, Cauchy was
reassigned from the Ministry of the Marine to the Ministry of the Interior
on March 17, 1813, and was quickly appointed to an engineering position
at the Ourcq Canal (5).
He was thus returning to the construction project where he had worked so
diligently and so successfully as a student engineer. Pierre-Simon Girard, who
was still director of the Ourcq Canal and Paris Waterworks and who had a
deep appreciation for young Cauchy'S abilities, might well have been
responsible for having his former protege assigned once more to the Ourcq
Canal project. In any event, Augustin-Louis was to replace the engineer
Charles-Jean Lehot, who, because of engineering malfunctionings of the inner
36 3. The Waiting Room of the Academy

aqueduct, now found himself in Girard's bad graces. In spite of all his
protestations and requests, Lehot was to be punished, the administration
having decided to assign him far away from the capital, in the department of
the Puy-de-Dome (6). Before he could leave for Clermont-Ferrand, however,
Lehot was to remain in Paris for a while. During this time, he was to complete
a report on the works that had been under his supervision. In the meantime,
Augustin-Louis had resumed his work as an engineer and was able to help
Lehot in writing the required report (7).
So, Cauchy was once more working on an engineering project. But, even
though he was still in Paris, his thoughts were elsewhere, far removed from the
engineering routines of the Ourcq Canal project. Cauchy now contemplated
resigning from the engineering service. Because of his father and his own
abilities, he remained with the project, hopeful and expectant. On April 10,
1813, Lagrange died. The death of this eminent mathematician left two places
vacant, one on the geometry staff at the Bureau des Longitudes and the other
on the geometry staff at the Institut.
At the Bureau des Longitudes, Cauchy aspired to the position of librarian,
which should have been vacant as a consequence of staff changes: Pro ny,
supernumerary geometer at the Bureau since 1802, should have filled the
vacancy created by Lagrange's passing; Prony's appointment to the Lagrange
vacancy would mean that a supernumerary position would have to be filled,
and the Bureau unanimously chose his librarian, Claude-Louise Mathieu, for
that position; thus, a new librarian would have to be appointed. Cauchy won
the majority of the votes for this position. On May 4, 1813, Lehot, still in
disfavor because of his past performance, wrote to Count Mole:
I have just learned that Monsieur Cauchy is about to be appointed to an
important post at the Bureau des Longitudes; and, should that position
materialize, he has decided to quit his job on the Ourcq Canal project.
Sir, I dare hope that you would be so kind as to have me placed in that
position (8).
Unfortunately, for both Lehot and Cauchy, the Minister of the Interior chose
to disregard the Bureau's elections; the Bureau could only submit nominations
for the Lagrange vacancy; its decisions were not binding. Accordingly, three
candidates were nominated: Legendre, Poisson, and Prony. On May 26, 1813,
Legendre was appointed to the position at the Bureau, and Cauchy lost all
hopes of getting appointed to the librarianship since Mathieu would now
remain in that position (9).
Cauchy was no longer contented with the First Class of the Institut.
Nevertheless, with Poisson's help, Cauchy carefully laid the groundwork for
his own candidacy. On April 12, 1813, Poisson submitted a report to the
commission that was responsible for evaluating Cauchy's two-part study of
November 30, 1812, on symmetric functions. Although Poisson's report was
superficial, it was flattering to Cauchy. After examining the historical
development of symmetric functions and giving a short account of the results
 3. The Waiting Room of the Academy 37

contained in Cauchy's two studies, Poisson's report gave the grounds on

which the commission's approval was based, stating:
The questions that are examined in his [Cauchy's] latest studies are,
without doubt, of less importance than his studies on polyhedra.
Nevertheless, the present work gives new proof of the insights and
abilities that he demonstrated in the study just mentioned (10).
Cauchy was able to insert this favorable report in the brochure which he had
published for the elections and which contained the reports on all the studies
he had submitted to the Institut (11). Moreover, in May 1813, the Journal de
l'Ecole Poly technique published a brief article in which Cauchy gave several
theorems on congruences he had taken from his studies on polygonal
numbers. In particular, there was a theorem that had already been proved by
Lagrange in 1770 (12).
Finally, a week before the geometry section was to submit its list of
candidates, Cauchy presented a new study on the theory of equations to the
First Class. Thus he artfully caught the eyes of the members ofthe Institut (13).
Cauchy expected to give verbally a concise account of the contents of this
paper at the First Class' meeting of May 17. But other readings filled up the
meeting. In this account, which he published a few days later, Cauchy intended
to pay his respects to Lagrange, 'the great geometer whose passing has
saddened learned circles everywhere in Europe as much as it has saddened us',
and he presented his own work as a response to a problem that Lagrange had
posed in his Traite de la Resolution des Equations Numeriques de tous les
Degres.
Lagrange had published that work in 1798, and in it he examined Descartes'
rule on the number of positive roots of an algebraic equation. This rule had
been proved by De Gua in 1748, and Lagrange's work proceeded to pose-but
not solve-the more general problem of the a priori determination of the
number of real roots, whether negative or positive, of an equation. While it
may no doubt be that Lagrange's research was the source of Cauchy's inquiries
on the subject, as Cauchy himself declared in his account, it is at best only
indirectly so, for it was Poisson who had suggested this research problem to
Cauchy. Moreover, nowhere else in the memoir Cauchy published on the
subject two years later is Lagrange mentioned (14). The homage Cauchy paid
to Lagrange thus appears to have been rendered with an eye on the coming
election to the seat that Lagrange had left vacant.
Only eight days after the meeting, on May 24, 1813, Poisson, in a move that
was unusual at the Institut but which can be explained in light of the coming
election to the geometry section, gave his flattering report on Cauchy's papers.
While emphasizing the young engineer's strengths, he pointed out certain
methodological shortcomings and called on Cauchy 'to be more attentive [to
them] if he would improve his work'. On the same day, the geometry section
presented its list of candidates to the Institut: Cauchy was in second place,
behind his young friend Jacques Binet, but ahead of Duvillard, Poinsot,
38 3. The Waiting Room of the Academy

Joseph-Louis Lagrange (1736- 1813), the best mathematician at the

end of the eighteenth century.

Puissant, Ampere, and Parse val. Nevertheless, on May 31,1813, Poinsot won
by 23 votes over the obscure Duvillard on the third ballot. Cauchy had been
eliminated on the first ballot, having received only 2 votes as opposed to 19 for
Poinsot, 17 for Duvillard, and 12 for Binet (15).
There was now no chance that Cauchy would be able to give up his position
as an engineer, at least for the moment. He was no doubt deeply hurt by these
sharp checks on his ambitions, and being only partially recovered from his
illnesses, he requested a six-months' leave of absence without pay to regain his
health. This sick leave was granted on June 5, 1813 (16). However, he had to
wait until mid-July to be replaced by the engineer Denoe! before enjoying his
leave (17).
Cauchy took advantage of this new sick leave to resume his research on the
problem of the a priori determination of the nature and number of roots of
 3. The Waiting Room of the Academy 39

an algebraic equation. The method he had propounded in May 1813 had,

in effect, turned out to be difficult and applicable only to equations with
simple roots. Thus, on October 18, 1813, he submitted a new paper to the
First Class of the Institut, stating:
Two months ago, I had the honor of submitting to the Class a method by
which I could determine a priori the real roots of a given equation. The
Class decided to honor it with their approval. However, although that
method is capable of yielding a solution to the problem when applied to
equations of any given degree, one must take account of the number of
equal roots and of the multitude of special situations that are prejudicial
to the results obtained. The method I propose today does not have these
drawbacks. It allows us to determine the number or roots, positive and
negative, in a general way and that is independent of any considerations
of equal roots and of other special cases that might arise.
Cauchy first developed his ideas in geometric terms with no 'mathematical
formulation', and then proceeded to develop an 'application to analysis of the
preceding discussion'. Again, Poisson, assisted by Arago, evaluated this
research. As the penciled-in marginal notes confirm, Poisson examined the
manuscript very carefully. It seems that he was quite dissatisfied with the
study, for in his comments, he demanded clarifications and concrete details.
Cauchy wrote several supplementary notes, probably during the following
months; and, at last, on November 22,1813, he submitted a new study on this
subject to the First Class of the Institut (18).
Cauchy had now spent many months working on this problem, and
although Poisson's report was favorable, the undertaking was only partially
successful. To be sure, he had shown for the first time that it is always
possible-at least in theory-to determine a priori the nature and number of
roots of an algebraic equation. In practice, however, this method led to long,
complicated calculations and was accordingly almost unusable.
Thus, the year 1813, which Cauchy had spent researching equations, clearly
seems, on balance, to have been less fruitful than 1812 had been or 1814 would
be. Overall, this decline in Cauchy's output can perhaps be explained in terms
of his poor health. In any case, over the course of his life, Cauchy would come
back several times to tackle the problem of the a priori determination of the
nature and number of roots of an equation. But, by then, as we will see, he
would have the resources of analysis at his command, singular integrals and,
later on, the calculus of residues.
In January 1814, when Augustin-Louis' sick-leave was due to expire, Louis-
Franc;:ois wrote to Baron Costaz asking that an additional three months' leave
be granted his son whose health had been 'profoundly undermined by his work
load' during the three years of service at Cherbourg (19). Following this
supplementary leave, it seems that Augustin-Louis resumed his duties at the
Ponts et Chaussees in March 1814, although the relevant records are silent on
40 3. The Waiting Room of the Academy

this point (20). In any case, the events of 1814 and 1815 brought work on the
Ourcq Canal project to a stop, and Cauchy thus had the time he needed to
devote himself to research. After an unfruitful year in 1813, the next two years
would be intensely creative.
On March 7, 1814, as the Emperor was battling the allies, who were now
only a few kilometers outside Paris, Cauchy submitted another paper to the
First Class of the Institut. Dealing with the theory of errors (21), this study was,
as he put it, a 'work undertaken on command'. Obviously mindful of placing
himself under the patronage of a man such as Laplace, who had great influence
not only with the Institut but with civil officials also, the young engineer
acknowledged that he had been 'guided to this research topic by M. Laplace'
and was 'duty-bound to comply with Laplace's requests'. Again, a commission
was charged with the responsibility of evaluating the paper. This commission,
which was composed of Laplace and Poisson, made no report on the
study.
But, Cauchy was already preparing another study, suggested by the reading
of Laplace's works, which would have quite a different scope. This long study,
entitled 'Sur les integrales definies', set forth some entirely new methods for
calculating numerous definite integrals. Truly, it was the starting point of all of
Cauchy's work in analysis, especially in complex analysis. The study was
submitted to the First Class of the Institut on August 11, 1814. A commission
composed of Lacroix and Legendre, the latter acting as reporter, was charged
with the responsibility of evaluating this study, and a favorable report was
issued. Cauchy was now established as one of the most gifted young
mathematicians of his day.
Still, two years had now passed since his return from Cherbourg, and he had
not been able to quit his position at the Ponts et Chaussees. New opportunities
for entering the Institut occurred by the end of 1814: the deaths of Charles
Bossut on January 14 and of Pierre Leveque on 16 October 16, 1814, created
two vacancies in the First Class of the Institut, as well as one vacancy as
Examiner of Naval Students. According to a letter Laplace wrote to Charles-
Louis Huguet de Semonville, an important official in the chamber of Peers
(22), Cauchy was canvasing for the latter position. In this letter, Laplace
declared that:
M. Cauchy [Louis-Fram;ois] has just informed me of the death of M.
Leveque, my colleague at the Institut and an examiner of naval students.
He would like his son to fill this position as examiner. I can assure you
that no one is more qualified to fill this important position than that
young man.
Alluding to Augustin-Louis' candidacy for the vacancy that Bossut's death
had created on the geometry section of the First Class, Laplace added:
The younger M. Cauchy is a very distinguished geometer whose many
excellent papers dealing with different topics in science prove to the
 3. The Waiting Room of the Academy 41

Institut that he possesses an unusual ability in mathematics. Of all those

who aspire to the present vacancy at the Institut, he is the one who seems
to me to have the greatest right to it. I believe that he would be a worthy
replacement for the geometers in this illustrious body. I take great
pleasure in recommending him (23).
The elections were to be held on November 24, 1814, and Augustin-Louis
carefully prepared for it, writing up a printed prospectus on his works (24).
Above all, his father, taking advantage of his position in the Chamber of Peers,
used his influence to 'swing the election' in his son's favor. Following are
portions of a letter from Cuvier to Charles-Henri Dambray, President of the
Chamber of Peers and Keeper of the Seal, which, like the letter from Laplace
cited above, provides clear proof of the unwarranted interference that now
took place:
Sir, young M. Cauchy, about whom your excellency has requested me to
write, is really one of our most distinguished geometers. All the great
masters of science have attested to his ability, and it cannot be doubted
that the section will find that it has an excellent addition in him.
Cuvier, noticeably sensitive to pressure from higher authority, thought it wise
to add that:
The interest that your excellency shows in this adds immensely to my
own desire to see him [take his place] among us, a desire that is
unanimously shared by my colleagues, even those who might wish to see
some of his rivals advanced ahead of him in the present competition (25).
Ampere, Cauchy's principal rival this time, thus had good ground for worrying
and wrote:
I have a very well supported competitor in the person of one of my
[former] students from the Ecole Poly technique, now an engineer from
the Ecole des Ponts et Chaussees. He is the son of the Secretary of the
Chamber of Peers, etc... Unless I submit papers each month from now
until the nominations are over, I could be easily defeated (26).
Nevertheless, on November 28, 1814, Ampere was easily elected on the first
ballot. Seven candidates had competed: Ampere, Binet, Cauchy, Duvillard,
Francoeur, Parseval, and Puissant. Of the 52 votes, Ampere won 28, Cauchy
10, Binet 7, and Duvillard 6. Compared with the results of the May 1813
election, Cauchy had improved his score noticeably (27).
A few days later, on December 10, 1814, a vacancy was announced at the
Societe Philomatique de Paris, with which Cauchy had been a correspondent
since 1812. On December 17 of that year, a committee composed of A. H. de
Bonnard, Bosc, Collet-Descotils, and Poisson selected four candidates for the
vacancy, Cauchy, Lepelletier, Thillaye, and Henri. On December 31, Cauchy
was elected a member of the Society (28). This was surely a consolation
42 3. The Waiting Room of the Academy

because, instead of having become a member of the Institut, where his age was
a handicap in the view of many members, Cauchy would now take his place on
a learned society that was then regarded as a 'waiting room' for future
members of the First Class of the Institut. Cauchy made a new and final
attempt to get into the First Class when Napoleon Bonaparte, having escaped
from Elba and reentered Paris (March 20, 1815) gave up his place in the
mechanics section. On May 1, 1815, the section submitted a list of seven
candidates for this vacancy to the First Class: Breguet, Hachette, Gengembre,
Cagniard-Latour, Molard, Cauchy, and Binet. The names of three other
candidates were added to this initial list: Girard, Lenoir, and Janvier. The
election took place on May 8, 1815, and of the 53 votes cast, Molard, the
winning candidate, received 28; Gengembre was next with 9; Breguet followed
with 8; Hachette received 5; Janvier ranked next with 2, and Cagniard-Latour
was last with only 1. Cauchy received not a single vote in this competition (29).
At this time, Augustin-Louis was only 26 years old. He was now looked
upon as a gifted, young mathematician with a bright future ahead of him, a
scientist following in the footsteps of his brilliant mentors, Lagrange, Laplace,
and Poisson. Yet, his job situation was bleak indeed; in spite of his efforts and
the strong support that he had received, he still had not been able to secure a
position in the learned community. Indeed, such positions were now becoming
increasingly difficult to come by, in education, at the Bureau des Longitudes
and the Institut. All doors remained closed. In all fairness, however, the Ponts
et Chaussees director had been very cooperative in giving him an assignment
on the Ourcq Canal project and granting him many leaves of absence, which
allowed him to work on his research in mathematics. But, even at that, had
Cauchy not had an influential, well-to-do, understanding father who helped
him materially and morally, it is doubtful that after his return from Cherbourg,
he could have undertaken the research that opened (albeit with the aid of some
politicking) the way to a brilliant scientific career.
It is difficult to use a particular case to generalize about the whole learned
community in France during the last years of the First Empire. Yet, it does
seem that after the great flowering of scientific institutions that had taken place
between 1795 and 1800, a blossoming that allowed many young men, such as
Poisson, Arago, Biot, and Ampere, to advance very rapidly to important
positions, the revival slackened off in proportion to the number of very young
people who happened to have 'made it'. For the next generation, the
generation of Cauchy, Fresnel, and Navier, success was more difficult to come
by because the avenues leading to positions were narrower and more
congested. It is perhaps possible to identify this as one of the causes for the
relative decline of science in France a few years later.
 Chapter 4
A Man of Science

On July 18, 1815, after the defeat at Waterloo and the Emperor's second
abdication on June 22, and thanks largely to the intrigues of the regicide
Fouche, Louis XVIII reentered Paris on July 18, 1815, amidst the rejoicing of
the royalists. A new era had now dawned in France. It would be a new era for
Augustin-Louis Cauchy also; an age more to his liking, more compatible with
his interest and ambitions. Since 1812, he had been nurturing a deep desire to
embark on a career in science. He had produced studies of merit and had the
support of influential people, but he had not been able to realize his hopes.
Things would be different now. His father, Louis-Franc;ois-ever the loyal
supporter of whoever happened to be holding the reins of power-had
managed to preserve his position in the Chamber of Peers, the new name of the
Senate after the Hundred Days. Thus, Louis-Franc;ois had also salvaged his
political influence. Augustin-Louis could count on him using this influence to
advance his career (1). Moreover, in this period of reaction, which followed
Napoleon's fall, Augustin-Louis could count on the goodwill and support of
powerful royalists whose friendship he had been nurturing since 1808 when he
joined the Congregation. Henceforth, Augustin-Louis' political views and
leanings would be ultra.
For Augustin-Louis, the Restoration Era would be a time of rebirth after
the somber years of the Great Revolution and First Empire. Thus, in a letter
dated September 3, 1815, the young man expressed his views on the current
situation:

You have no doubt heard that some 40 hours of prayers have been
offered here, prayers in repentance for the excesses and crimes commit-
ted during the Revolution. This is as it should be: nobody should stop
praying, there is so much to set right in France. Besides, it is rather a
consolation to see that the prayers offered in the matter of the elections
appear to have borne good fruit; it seems one can generally be satisfied
with the choice of deputies (2).
43
44 4. A Man of Science

In such a frame of mind, he could identify his own interests with those of
France and promptly accept political interventions and influence peddling,
backstairs maneuvers that were solicited by him or by those close to him.
At the Institut, no less than at the Ecole Poly technique, there was now a
move afoot to replace those scientists who had become politically compro-
mised by their overly close association with the now discredited Revolution
and Bonapartist regime. For Cauchy, then, it was a question of taking
advantage of what his fellow royalists saw as offerings of atonement.
It was at the Ecole Poly technique that Cauchy was first able to assess the
effect that the new order would have on his career. On November 2, 1815,
when he first presented himself as a candidate seeking the chair in mechanics
that Poisson had vacated, the Conseil de Perfectionnement (Improvements
Committee) at the Ecole had preferred Jacques Binet to him. Binet, of course,
was already a repetiteur (tutor) in analysis and, like Cauchy, a member of the
Congregation. But, the matter did not rest here; for, a few days later, the
Governor of the Ecole appointed Cauchy assistant professor of analysis,
responsible for the first division (that is, the second year) course.
The two courses in analysis at the Ecole were taught by Ampere and
Poinsot, the latter having occupied the chair in analysis since 1809 when he
replaced Labey. Holding the rank of associate professor, Poinsot was lax
about his duties. He continually used his poor health as an excuse for not
teaching his courses himself. During 1812-1813 and 1813-1814, the Governor
of the Ecole had appointed Antoine Reynaud, a repetiteur, as Poinsot's
substitute. Reynaud was such a mediocre teacher that at the beginning of the
November 1815 term the students sent a delegation to Poinsot 'imploring him
to please be kind enough not to get sick anymore' (3). Poinsot, however, was
unwilling to resume his teaching duties until he had wrung from the
administration a change in the schedule when his class sessions would be
given.
Poinsot's hostility to teaching his assigned classes in the normal way sprang
from his displeasure with his professorial rank. Being only an associate
professor, he was required to split his salary with Labey, the full professor
whom he had replaced and who was still incumbent.
This was the early 19th century, and of course, there were no retirement
funds at the Ecole. The oldest professors, men who were frequently ill and who
could no longer teach, preferred to let themselves be 'substituted for', rather
than resign outright. By this device, the incumbent full professor took part of
the salary while his replacement assumed the full teaching load. Poinsot, of
course, was not the only member of the faculty at the Ecole who was caught in
this salary-splitting trap: Francois Arago in applied analysis and Alexis Petit
in physics were in the same situation. The Conseil de Perfectionnement had
made several unsuccessful attempts to come to grips with this problem. On
November 2, on the advice of a special commission that was headed by
Laplace, the Conseil decided to retire those professors who no longer taught,
naming them professors emeriti. The chairs were affected to the extent that the
 4. A Man of Science 45

actual teachers would obtain the full professorships. This measure would not
affect pay, of course: salary splitting would continue. However, it did offer
Poinsot, Ampere, and Petit titles of full professor along with the hopes and
chances of getting their chair's full salary once the respective emeritus was
dead.
On November 15, 1815 several days after this measure was effected, Poinsot
declared to the Governor that his health would not permit him to teach at the
scheduled time; and, having this time been unable to get a change of schedule
(as he had done before), he asked to be replaced. Instead of asking that
Reynaud (whose work had been unsatisfactory) be named as the replacement,
Poinsot proposed the repetiteur Lefebure de Fourcy. Upon receiving
Poinsot's letter, the Governor of the Ecole sought the advice of Durivau, the
Directeur des Etudes (Dean of Studies), who considered Cauchy a better
choice (4). Durivau's advice was accepted. Thus, the Governor took the early
opportunity of fulfilling a secret promise to admit Cauchy into the faculty of
the Ecole. The position carried a salary of only 1500 francs, a mere fourth of
that received by a full professor and the same amount repetiteurs were paid.
Yet, Cauchy accepted the offer-no doubt because the Governor, now
exasperated by Poinsot's whims, gave him grounds to expect that a definitive
appointment would be soon forthcoming (5). On December 2, 1815, the
Governor informed the Conseil de Perfectionnement of his decision and gave
notice that Cauchy would teach the first division (second year) course for the
entire year. Moreover, the Conseil decided, much to Poinsot's displeasure, to
suspend his (Poinsot's) nomination to full professor, which had been advanced
a month earlier. Thus, the way was now open for Cauchy to displace Poinsot
once and for all.
By appointing a young engineer to the faculty of the Ecole (6)-and to the
most prestigious of all the professorial chairs at that-the Governor had
broken a well-established tradition, which had also worked against Cauchy, and
in favor of Jacques Binet, on November 2. The rule was to appoint professors
from among the repetiteurs on staff at the Ecole. To be sure, the importance of
the works Cauchy had written could easily have justified such an exception to
the tradition. But, the recommendations of several of the most illustrious men
on the faculty-particularly of Laplace, Cauchy's protector-would doubt-
lessly have not sufficed to gain him the appointment had not political events
worked in his favor. The forced resignations of Monge, Guyton-Morveau,
Hassenfratz, and Lacroix had provided an elegant opportunity to staff the
Ecole with politically acceptable scientists. The replacement of Poinsot by
Cauchy was obviously part of this great purging enterprise, an undertaking
that would culminate several months later in the reform of the Ecole
Poly technique (7).
So, Cauchy was able to satisfy his academic ambitions during the period of
reaction that followed the Second Restoration. He had already been presented
to the Institut three times-in 1813, 1814, and 1815-and each time his
ambitions had been checked. Still, he could hope that in 1816 he would be
46 4. A Man of Science

elected on his own merit. Thus, it was during the course of the second semester
of 1815 when he presented two outstanding studies, works that undisputably
made him equal to the greatest.
Dealing with the theory of waves, the first work was presented anonymously
to the Institut on October 2,1815. This study, which competed for the Grand
Prix in mathematics, is one of Cauchy's most carefully written works. On
December 26, 1815, it took the prize of 3000 francs (8). But it was another
study, presented to the Institut on November 13, 1815, that made him famous
(9). In this work, he gave the general proof of Fermat's conjecture on polygonal
numbers. 'Every number', wrote Fermat in a letter of 1636 to Father Mersenne,
'is the sum of three cubes, of four squares, and so on, indefinitely'. Lagrange
had proved the conjecture for the squares in 1770 and Legendre for the cubes
in 1798. Gauss gave a new proof of the theorem for cubes and squares in his
Disquisitiones Arithmeticae. His demonstration was based on the theory of
binary forms. Cauchy, who had read Disquisitiones, had been wrestling with
the general proof of the conjecture since 1812. Cauchy's proof used the general
proposition that it is always possible to decompose an integer into four
squares, the positive roots of which add up to a sum that is equal to a given
whole number of the same parity (10). The proof of Fermat's theorem on
polygonal numbers made him a veritable sensation in the world of mathema-
tics. The announcement of his proof may have supported his appointment to
the Ecole Polytechnique a few days later. Following Cauchy's discovery, the
First Class of the Institut proposed a new research topic for its Grand Prix de
Mathematiques: Fermat's last theorem (11). In spite of the efforts of the
contestants, among them Gabriel Lame and Sophie Germain, nobody took
the prize.
Cauchy's studies on the theory of waves and on Fermat's theorem opened
the Institut's doors to him. He needed only to wait for the next election, but
even this was not necessary, because, as in the situation at the Ecole
Polytechnique, Cauchy took advantage of the purge that had been undertaken
by the new regime. On March 21, 1816, a royal ordinance reorganized the
Institut. The Academie des Sciences was reestablished under its former name.
At the same time, several members of the First Class were replaced by
newcomers; in mechanics, Carnot and Monge were removed for political
reasons, and Louis XVIII appointed Cauchy and Breguet to their places.
These purges were regarded by the Academie, and by learned society in
general, as contemptible affronts. Carnot and especially Monge were respected
scientists, first-rate minds. Cauchy nevertheless accepted his appointment
without hesitating. Judgment has been harsh on him for his insensitive
attitude. 'In the corridors and in the drawing rooms', wrote Bertrand,
'invectives and slanders are accepted, no doubt in good faith, by savants
worthy of respect as well as by other important persons'. This politically
imposed appointment to the Academie caused considerable harm to young
Augustin-Louis, who promptly proceeded to make enemies for himself:
'Cauchy', Bertrand added, 'found few defenders. He has seen more than one
 4. A Man of Science 47

Gaspard Monge (1746-1818). Cauchy replaced him at the

Academie des Sciences in 1816.

friend who, though naturally tolerant and decent, turned away and refused to
call him "brother'" (12).
A few days after this nomination, Cauchy's first year ofteaching at the Ecole
Poly technique came to a premature end, with the disbanding of the student
body (13). The royalists regarded the Poly technicians with dislike and
mistrust, for these students had given proof of their liberal and Bonapartist
way of thinking. Since the Hundred Days, their lack of discipline had been on
the increase. An incident within the Ecole brought the royalists' anger to a
head: the repetiteur Lefebure de Fourcy, a dyed-in-the-wool conservative, was
subjected to a boisterous display of disrespect by the students in the second
division, and the students in the first division joined their comrades. Among
the disruptive youths was Auguste Comte, who had an intense dislike for the
arrogant repetiteur. On April 12, 1815, the student assembly demonstrated
against the administration for having punished the students for their actions.
Here the matter might have rested with no further ado. But the authorities
seized on the incident as a pretext for attacking the republican, antireligious
mood that prevailed at the Ecole. On April 13, the student body was disbanded
48 4. A Man of Science

by royal ordinance, and on the next day they were sent home. The teaching
faculty was now put on half-pay until further notice and a five-member
commission, headed by Laplace, was appointed to prepare a reorganization
plan for the institution (14).
On September 4, 1816, a new ordinance that profoundly modified the
bylaws of the Ecole was issued. The principal step taken was the demilitariz-
ation of the institution's structure. This reorganization however, was merely a
pretext for getting rid of a certain number of professors. The purge that had
begun in November 1815 was now to be completed. A campaign led by the
reactionaries got under way. One of the participants in this movement was
Lamennais, who published an anonymous pamphlet entitled Quelques Reflex-
ions sur l'Ecole Poly technique (15) Published in June 1816, this little brochure
was designed to put pressure on Laplace's commission, and its most obvious
result was the dismissal of Andrieux, the professor of fine arts. Moreover, this
pressure may not have been unrelated to Cauchy's appointment as full
professor in analysis and mechanics. Thus, the liberal Poinsot was set aside to
the advantage of the extremely conservative Cauchy, who, it will be recalled,
had already been rewarded six months earlier with a political appointment to
the Academie des Sciences.
In the climate of reaction that prevailed in France during those days, the
effects of the political atmosphere cannot be overlooked. Yet, there are
nonpolitical reasons that might have justified such an appointment. As we
have seen, Poinsot had not taught his classes since 1812; moreover, the
commission decided to combine the analysis and mechanics courses under a
single teacher. This would mean considerably more work for the full professors
responsible for these courses. It would require teachers who were completely
available-something that Poinsot certainly was not. In selecting Ampere and
Cauchy, the commission had certainly chosen two scholars of undisputed
intellectual qualities (even though both would turn out to be rather poor
teachers), who consistently taught their assigned courses during all the years
they were on the faculty of the Ecole. The situations of the two professors,
Cauchy and Ampere, were quite different: Ampere had taught analysis since
1809, when he replaced Lacroix; Cauchy, on the other hand, had only taught
analysis for one year-and then only as a replacement. Thus, it is clear that
Cauchy benefited from the special favor of the authorities.
Cauchy was now 27 years old and a recognized scientist. He was also a
teacher, and preparation for his courses at the Ecole absorbed his attention
almost completely. In subsequent years, he also took on teaching jobs at the
College de France and at the Faculte des Sciences. He could not become a full
professor at these institutions since there were no vacancies and had to content
himself with being a substitute or a replacement for professors who were on
leave.
The first opportunity had to do with the chair in mathematical physics at
the College de France. The incumbent professor in that chair, J.-B. Biot, was
preparing for a geodesy expedition to Scotland and the Shetland Islands. On
 4. A Man of Science 49

The Montagne Sainte-Genevieve in 1840. Lithograph. In the foreground, the Ecole

Poly technique, in the middle ground, left, the College Royal Henri IV, formerly Ecole
Centrale du Pantheon, which the young Cauchy attended in 1802- 1804, and, in the
background, the Pantheon. Published by permission of the Ecole Polytechnique.

November 10, 1816, he proposed to the faculty council that he should be

replaced by Cauchy if his planned expedition materialized (16). In May
1817, Biot left Paris, but it is not known exactly when Cauchy began teaching
at the College. In his 1817 course at the College de France, Cauchy presented
for the very first time the integration methods that he had discovered in 1814,
but that to date had remained unpublished (17).
Biot returned to France and shortly afterward seems to have resumed
teaching. Following his defeat in the election to the position of Secretary of the
Academie des Sciences, he temporarily withdrew from active participation in
scientific life. In 1824, he left with his son on a year-long trip to Italy and Sicily.
Before leaving, he asked that Cauchy teach his courses during his absence (18).
A year later, when he returned to France, he wrote a letter to the administrator
of the College proposing that Cauchy become his regular substitute at 2000
francs, to be deducted from his annual salary starting January 1, 1826 (19).
Thus, Cauchy taught courses at the College de France, first as a replacement
and then as a regular substitute, from 1824 to 1830. There he developed
mathematical methods that were applicable to the physical sciences. In
particular, he presented methods and techniques that could be used to solve
certain classes of linear differential equations with constant coefficients, both
ordinary and partial differential equations. Moreover, in 1830, he delivered
lectures on the theory of light (20).
50 4. A Man of Science

Aside from this, Cauchy was appointed substitute professor, replacing

Poisson on the Faculte des Sciences de Paris (21). Poisson had given up his
teaching position on the Faculte des Sciences once he became a member of the
Royal Committee on Public Instruction in 1820. At that time, Ampere had
agreed to teach the course in mechanics in Poisson's stead (22). But, the task
proved to be too much of a burden, and during the academic year 1820-1821,
he interrupted his teaching for a while. Cauchy was quite willing to act
temporarily as Ampere's replacement (23). At the beginning of the October
1821 term, Ampere made up his mind to retire and suggested to Poisson that
he be replaced by Cauchy. This appointment to the Universite marked an
important point in Cauchy's scientific career.
Since 1816, his time had been monopolized by preparation for the courses
he taught at the Ecole Poly technique. Now that the first part of his course on
analysis, entitled Analyse Algebrique, had been published, the course he taught
at the Faculte presented him with a chance to focus attention once again on the
mechanics of deformable bodies, a subject that he had not worked on since
1815. At the Faculte, as at the Ecole Poly technique, a large number of his
lectures were devoted to the mechanics of solids, approaching the topic from
the standpoint ofthe theory oflinear moments. However, in other lectures, he
dealt with fluid mechanics and the methods of solving equations [for sound as
well as the wave equation and the heat equation]. It is in this context that
Cauchy gave a full exposition of his general theory of elasticity in 1821 and
1822 (24). On December 1823, he was made an associate professor at the
Faculte and remained on staff until 1830 (25).
Established in 1815 and 1816 by virtue of his political connections,
Cauchy's reputation in scientific circles continued to develop along the lines
that it started out on. At the outset of his career, he had the firm support of
Laplace and Poisson and considered himself as being their disciple. But very
slowly, as the years went by, he began to become more independent. His own
ability was now confirmed, and the positions in scientific institutions that he
had obtained between 1815 and 1816 secured for him a greater measure of
independence from those who had been his patrons. The ever-increasing
degree of mathematical rigor that Cauchy advocated led him to criticize
certain works of Laplace and Poisson.
In this regard, the introduction to Analyse Algebrique is significant. Even
though he mentions in 'recognition' the names of Laplace and Poisson, as
regards the principle of 'the generality of algebra', which they (Laplace and
Poisson) subscribed to, he wrote:

As to the methods, I have sought to endow them with all the rigor that is
required in geometry, doing so in such a way that no recourse to reasons
based on the generality of algebra is needed. Such reasons, although
quite commonly admitted, especially in going from convergent series
and from real expressions to imaginary ones, can be only regarded, it
seems to me, as inductions that can sometimes be suitable for guessing
 4. A Man of Science 51

the truth. But, they agree very poorly with the vaunted exactitude
claimed by the mathematical sciences. It should even be noted that
they [reasons based on the generality principle] attribute an indefinite
(i.e., infinite) realm of applicability to algebraic formulas when, in truth,
most of these formulas hold only under certain conditions and for
certain quantities satisfying those conditions.

These remarks were aimed directly at Laplace and Poisson, the former having
based his theory of generating functions on the consideration of generally
divergent series, while the latter had advanced a method of computing definite
integrals by intuitively passing from the real line to the complex domain (26).
Elsewhere in this same introduction, Cauchy let it be known that he was
opposed to the philosophical conclusions Laplace had drawn from his theory
on the probability of testimonies (27).
Relations between Cauchy and Laplace and Poisson deteriorated as the
years passed. During these years, of course, Laplace had all but retired from
any active role in science, and his influence, though still considerable, had
declined accordingly. But, the unkind remarks that he made around 1825
before the Conseil de Perfectionnement of the Ecole about Cauchy raise
serious questions about his former protege as a teacher, if not as a
mathematician. More decidedly, these remarks reveal the lack of sympathy
that now separated the two men.
F or his own part, Poisson was still very much active in the world of science
and continued to play an important role in the different scientific institutions
of Paris. Moreover, competition was very sharp between him and Cauchy.
They tended to do research in the same broad areas: theory of definite
integrals, theory of elasticity, and linear partial differential equations. Indeed,
there was sometimes a veritable race between the two. But their competition
was always kept confined to scientific matters, so that, on the whole, the two
men managed to get along. An examination of their scholarly activities tends
to confirm this impression: Poisson was a member of 19 of the 42 evaluation
commissions with Cauchy as reporter between 1818 and 1823. No other
member of the Academie collaborated so closely and so frequently with
Cauchy during this period as did Poisson (28).
Relations between Cauchy and Poisson deteriorated during 1824 to 1825.
This was triggered by the evaluation report on a study dealing with the
solutions of differential equations that Brisson presented to the Academie on
November 17, 1823 (29). The evaluation commission, consisting of Cauchy,
Laplace, Fourier, and Poisson, with the first acting as reporter, was unable to
reach an understanding on the report that Cauchy drew up: Fourier and
Cauchy being very favorably disposed to Brisson's work, while Poisson was
adamantly opposed to it. Very prudently, old Laplace kept his distance
from all the brouhaha; and, when in a letter of March 10, 1825, Brisson
demanded that Cauchy's report be read, Laplace decided to withdraw
52 4. A Man of Science

Simeon-Denis Poisson (1781- 1840), the French rival of Cauchy in analysis and physical
mathematics. Photograph by J. L. Charmet, permission of the Academie des Sciences.

from the commission, claiming the pressing need to attend to other

business.
Up until this time, the dispute was confined to the commission, but it
became public on March 21, 1825, when Cauchy read his report to the
Academie, and under Poisson's instigation, the Academie refused to adopt it.
It was a stunning humiliation for Cauchy insofar as it was highly unusual for
the Academie des Sciences to reject a report's conclusions. Thanks to Fourier,
however, Cauchy was able to get his revenge two weeks later. Following a
letter from Brisson, the Academie decided at its meeting of April 4, 1824, that a
new commission should be appointed to evaluate Brisson's study-and
Poisson was to be excluded from it, although Cauchy would still be reporter.
The second report was adopted by the Academie on June 13, 1825.
 4. A Man of Science 53

In the wake of the Brisson affair, Cauchy and Poisson seem to have avoided
being on the same commissions. Nevertheless, in 1827 and 1829 similar
situations arose. However, in these cases no report was submitted to the
Academie. Scholarly collaboration between the two suddenly burst forth
again in 1830. Over a period of eight months, they worked together on six
commissions, something that had not been seen since 1823. No reports were
presented, however, for Cauchy left France in September 1830. Although
nothing is known about the reasons why the two scholars so suddenly changed
their attitudes, it is doubtful that there was a real reconciliation.
The scientific rivalry between Cauchy and Poisson had, in fact, been very
much a reality for the three years preceding the former's departure from
France in the fall of 1830. The issue on which it centered was the molecular
theory of elasticity, a research area that was then in full bloom. On several
points, Cauchy sallied forth to express criticisms or issue an objection to
something that Poisson would have communicated to the Academie. Thus, a
veritable argument broke out between them in Apri11829 about some points
on the theory of fluid flow. Again, in June 1830, Cauchy made some biting
remarks to the Academie on a question of priority between himself and
Poisson relative to fluid flow and the theory of light (30).
It is difficult to determine the precise nature of Cauchy's relations with
other members ofthe Academie des Sciences. He kept away from all the cabals
and shifting groups and maintained a distant coolness but perfect politeness
toward his colleagues. In general, he avoided getting involved in the debates
and squabbles that ripple across the quiet of academia from time to time.
Nevertheless, the nature of the relationships he maintained with the other
academicians seems to have been determined, in large part, by political and
religious considerations. Besides, no one had forgotten the conditions under
which he had come to be on the Academie in the first place. Cauchy stayed on
good terms with those of his fellows who were conservative in their political
learnings and staunchly Catholic in their religious life.
He was especially close to Ampere, who had been his repetiteur in analysis
at the Ecole Poly technique during 1805-1806, and they almost always stood
together in the face of criticisms from the various committees. They obviously
saw a lot of each other at the Ecole, where, since 1816, they occupied the twin
chairs of analysis and mechanics. Ampere, being less interested than Cauchy in
analysis and continuing his research on physics, tended to let his colleague
take the initiative in preparing the courses they taught. The two men
collaborated quite regularly at the Academie. From 1816 to 1830, Cauchy was
the reporter on 96 evaluation commissions at the Academie; Ampere was on
26 of these, more than any other member of that body. Ampere thought highly
of Cauchy. According to Valson, he used to attend his colleague's classes,
sitting right alongside the ordinary students. As for Cauchy, he publicly
thanked Ampere in the introductions and forewords to his textbooks for his
wise advice and observations (34). Later, on several occasions, he pointed to
Ampere as the exemplar of a genial, deeply Christian scholar and scientist. It
54 4. A Man of Science

should be noted, however, that at no time did Cauchy ever show any great
interest in electrodynamics, Ampere's field of research.
On the other hand, Cauchy was interested in optics during the decade of the
1820's, and that was Fresnel's research area. Unfortunately, we know almost
nothing about his relationship with the founder of the wave theory of light.
They were the same age and had the same intellectual background, as well as
the same political and religious learnings. But, Fresnel died prematurely in
1827, just at a time when Cauchy began to work on his own theory of light.
Cauchy's relationship with colleagues of a liberal persuasion seems to have
been determined by political considerations. He could hardly have been held
in high regard by Arago and Prony. Cauchy and his old teacher, Pro ny, had a
mutual dislike dating from the Ecole Poly technique. Of course, Cauchy was
frankly detested by Poinsot, whom he ousted from a chair at the Ecole (32) and
who reproached him for having plagiarized his theory of couples and renamed
it the theory of linear moments (33).
In spite of all this, however, it would be incorrect to regard the personal
relationships between the academicians as consisting only of opinionated
quarrels. For example, we know nothing of how Fourier and Cauchy got
along. In spite of the differences in their political, philosophical, and religious
learnings, nothing indicates that relations between these two men were bad.
Outside the Academie, Cauchy was bound to a certain number of fellow
scholar-scientists whose work he esteemed. In this respect, particular mention
should be made of Jacques Binet, with whom he had become close during the
time they were both students at the Ecole Poly technique. Binet shared
Cauchy's extreme conservative notions and, also like Cauchy, was a member
ofthe Congregation. In 1816, Binet became Inspecteur des Etudes at the Ecole
Poly technique and in that capacity was able to spend time regularly with his
old friend (34).
Relations between Cauchy and the engineer Barnabe Brisson were
altogether different between 1823 and 1828. Cauchy attached considerable
importance to the works that Brisson submitted to the Academie des Sciences,
although the engineer was totally opposed to him at first, as Brisson had been
one of Monge's favorite disciples and was also Monge's nephew by marriage.
But, Brisson was also the brother-in-law of Biot, with whom Cauchy was on
good terms.
One of the most difficult challenges facing an established scientist is that of
finding just the right tone in dealing with young, as yet unknown, scholars who
ask for advice and protection. The real challenge lies in criticizing without
discouraging, protecting without smothering, listening without taking over
another's work. In his youth, Cauchy had the wise, strong support of
established mathematicians, such as Lagrange, Laplace, and Poisson. While it
is true that he was never head of a school, it is also true that he lacked the moral
authority of Laplace, just as he lacked the intellectual generosity and openness
of Fourier. But, his reputation as a mean-spirited, callous scholar (35), a
reputation that was eagerly hawked about in liberal publications and
 4. A Man of Science 55

confirmed by several unfortunate slights and misdeeds that have since been
shown to have been pinned on him, was largely unjustified. As a matter offact,
throughout his long career, particularly during the 1820's, he showed a sincere,
but admittedly clumsy, concern for the aspiring young who sought him out.
At the Academie Cauchy shouldered a good deal of the work on the
commissions that were responsible for evaluating submitted studies. Indeed,
between 1816 and 1830, he evaluated 32 papers outside the commissions. Most
of these were of no importance, and he merely gave the Academie verbal
reports on them (36). Acting for the commissions, he actually gave the
Academie reports on 43 papers, one verbal and 42 written. Another 18, which
were submitted in 1829 and 1830, probably could not be evaluated until after
Cauchy left the country in 1830 (37).
Among the young mathematicians whose studies Cauchy examined were
Poncelet, Libri, Lame, Clapeyron, Rouche, Woisard, Abel, Ostrogradski,
Sturm, Galois, Liouville, and Duhamel. Several of these youths were
disappointed with Cauchy's attitude.
Poncelet's work on projective geometry was criticized for its lack of rigor.
Years later, Poncelet could still recall with anger and bitterness how one day in
June 1820 Cauchy had literally 'sent him packing'.
Fearing, and with good reasons, that twelve years of work and ceaseless
meditation would not clarify a deceptive problem and perhaps even
make me the subject of ridicule in the eyes of my superiors, of my friends,
and of everybody interested in geometry, though inditTerent and a bit
indulgent, I managed to approach my too-rigid judge at his residence at
No.7 rue Serpente. I caught him just as he was leaving for Saint-Sulpice.
During this very short and very rapid walk, I quickly perceived that I
had in no way earned his regards or his respect as a scientist, and that it
might even be impossible to get him to understand me. Humble
petitioner that I was, I thus restricted myself to respectfully informing
him that the objectionable points and difficulties that he believed he saw
in the adaptation of the principle of continuity to geometry were
essentially results of the insufficient attention that had heretofore been
accorded to the law of signs, a law that had absorbed my attention since
1813, when I was in Russia, and especially since my return to France in
1814. I explained that the mathematical discussion of this law could have
preceded my communication with the Academie, had the esteemed men
on that body not dissuaded me from doing so. However, without
allowing me to say anything else, he abruptly walked otT, referring me to
the forthcoming publication of his Le{:ons a['Ecole Poly technique, where,
according to him, 'the question would be very properly explored'. (38)
A short while later, Abel also complained about Cauchy's behavior in this
regard. He had sent his study 'Propriete generale d'une classe tres etendue de
fonctions transcend antes' to Cauchy several days before submitting it to the
Academie on October 30,1826. But, Cauchy refused to pay any attention to it.
56 4. A Man of Science

Jean Victor Poncelet (1788- 1867), the inventor of projective geometry. Photograph by
J. L. Charmet, permission of the Academie des Sciences.

Cauchy is mad and there is nothing that can be done about him,
although, right now, he is the only one who knows how mathematics
should be done.
Abel wrote to his friend Holmboe on October 24. A little further on in this
letter, Abel elaborated:
I have completed a big paper on a certain class of functions to present
to the Institut. That will take place on Monday. I showed it to Cauchy,
but he hardly took the time to glance at it. Without boasting, I dare say it
is good, and I am curious to see what judgment will be given at the
Institut (39).
 4. A Man of Science 57

Niels Henrik Abel (1802- 1829).

Unfortunately, Abel was not to have that satisfaction. He wrote:

I waited every day for the decision on the paper I handed in at the
Institut. But, the sluggish men there did not finish with it. Legendre and
Cauchy were judges, Cauchy being 'reporter', Legendre merely going
along (40).

When Abel's untimely death occurred on April 6, 1829, Cauchy still had not
given a report on the 1826 paper, in spite of several protests from Legendre.
The report he finally did give, on June 29, 1829, was hasty, nasty, and
superficial, unworthy of both his own brilliance and the real importance of the
study he had judged (41).
Galois received the very first time around the kind of attention that Cauchy
ordinarily did not give. In fact, on May 25, 1829, Cauchy agreed to present
Galois' first research papers to the Academie himself, even though Galois was
a very young man and completely unknown. It was a step he had taken only
once before, on June 27, 1825, when he presented Frizon's study 'Sommation
des puissances sembi abies des racines d'une equation et calcul des fractions
continues'. On January 18, 1830, Cauchy remained at home indisposed.
58 4. A Man of Science

Evariste Galois (1811 - 1832).

However, he sent a letter to the Academie declaring his intentions to present an

evaluation of Galois' work the following week. Intentions aside, Cauchy did
not make the report on January 25 nor at any time later. Moreover, he made
no further reference to Galois, either in his own works or in his subsequent
statements to the Academie.
We can assume, as does R. Taton (42), that the decision not to present the
report was made in agreement with Galois in light of Galois' correction of the
paper and his participation in the Grand Prix de Mathematiques. However,
the fact remains that in December 1831, Galois complained bitterly that 'parts
[of his paper of February 1820] had been sent in 1829, but no report on it has
followed, and I have found it impossible to retrieve the manuscripts'. These
remarks were obviously aimed at Cauchy, and they raise the possibility that he
never returned to Galois the paper that had been submitted to him (43).
Not every young mathematician who sought Cauchy's help suffered
because of his indifference or lack of understanding as did Poncelet, Abel, and
 4. A Man of Science 59

EXERCICES
DE

MATHEMATIQUES,
••
PAR M. AUGUSTIN-LOUIS CAUCHY.

II(GillUWII .Illi CBI!P DBS pOlin BT CHAussbs. PIIOF!SSBUB A L'1!COLB 1I0YAJ..1I POLYTBCnNJQV.t,
PI\OP£SSBUn ADIOINT A loA FACOLTt DBS SCIllNeBS, 1III1IDBB DB L'ACAoillli' DBS SCIBNClIS,
CUi;VALIER DE LA LI!GION D'DONIi.BUR.

ISSIII§JE"1

SECONDE ANNEE•
• fila I iVi>;I

10.

A PARIS,
cnEZ DE BURE FRERBS, LIDRAffiES DU nOI ET DE LA BIBLIOTHEQU£ DU MH ,
nUB SERPENTB, 1\.' 7.

Exercices de Mathematiques, 2, 1827. Collection of papers by Cauchy, published by

his father-in-law, de Bure, between 1826 and 1830. Published by permission of the
Ecole Poly technique.
60 4. A Man of Science

Galois. Ostrogradski, for example, had been taking courses under Cauchy
since his arrival in Paris in 1824 and in. 1825 had fondly referred to him as his
'brilliant teacher'. As for Cauchy, he praised Ostrogradski for his contri-
butions to the development of the calculus of residues and cited him in several
of his works. It further seems that on several occasions Cauchy came to
Ostrogradski's aid during the latter's stay in Paris by getting him out of
debtor's prison, where the young Russian had been thrown for not paying his
rent (44).
The Restoration era was certainly the most fruitful period of Cauchy's
career. The number of research papers that he presented during this period,
some one hundred in all, including textbooks, articles in scientific journals,
and extracts, was considerable (45). Starting in 1826, he published a review
entitled Exercices de Mathematiques, which he himself edited. In fact, he was
the only contributor to this periodical, which appeared in regular installments
until 1830. By editing and publishing his own work, he was able to present the
results of his immense productivity more quickly and more thoroughly to the
public. The appearance of Exercices seems to be connected with his retirement,
for some unexplained reason, from the Societe Philomatique in 1825. Until
then, he had been able to get his summaries, extracts, and papers published
quickly in the Societe's Bulletin. The fact that Cauchy was connected to a
family of publishers, the de Bure family, provided him with the means of
publishing a periodical that was devoted solely to his own research studies, an
unheard-of privilege in the history of mathematics.
During this period, Cauchy's creative work was dominated by three crucial
themes: teaching, with emphasis on the foundations of classical mechanics and
especially on analysis; mathematical physics, with special interest in the theory
of elasticity and its application to the theory of light; and finally, higher
analysis, with emphasis on the development of the theory of functions and the
calculus of residues. Each of these themes will be explored in the following
three chapters. Here, we see that, as with his dual career as a professor and
member of the Academie, different research interests took shape, each one
acting as a source of inspiration for the other. Thus, in his immensely creative
mind, there was an interplay of problems, methods, and results, which between
1821 and 1825 culminated in his great textbooks, the creation of the general
theory of elasticity, and the development of complex integration and the
calculus of residues.
 Chapter 5
Teaching at the Ecole Polytechnique

In the preceding chapter, we saw how Cauchy became a member ofthe faculty
of the Ecole Poly technique, first, in November 1815, by replacing Poinsot as
professor of analysis and then, in September 1816, by obtaining an appoint-
ment as a full professor of analysis and mechanics. After the closing of the
Ecole in May, classes did not resume until January 1817. However, the Conseil
d'Instruction (Curriculum Committee) of the Ecole began to hold a series of
sessions on November 15, 1816, in order to organize the instructional program
for the academic year 1816-1817. The courses in analysis and mechanics,
which would henceforth be taught by the same professor, were of particular
concern.
The first piece of business facing the Conseil d'Instruction had to do with
the selection of repetiteurs who would work under Cauchy and Ampere.
Cauchy proposed that Coriolis, a young engineer from the Ponts et Chaussees,
serve as tutor for his courses. Born in 1792, Corio lis had entered the Ecole
Poly technique in 1808 and was only a few years younger than Cauchy. Cauchy
and Coriolis had never worked together, either at the Ecole or in the
engineering services, but they shared the same political and religious views.
Endorsing Cauchy's choice, the Conseil d'Instruction selected Coriolis on
November 28, 1816, preferring him to Paul Binet, brother of the Inspecteur
des Etudes (Dean of Studies) and to Destainville. Coriolis served in this
capacity until 1830, when he temporarily replaced Cauchy. It is beyond doubt
that he exercised a certain amount of influence on Cauchy's teaching (1).
Meanwhile, the basic issue facing the Conseil d'Instruction at its meeting of
November 15 was that of defining and restructuring the Ecole's academic
programs. On his own authority, Cauchy requested a change in the
organization of the course in analysis and mechanics:

M. Cauchy asks for a change in the structure of the analysis course and
the mechanics course. He describes the inconvenience of not being able
to present the differential and integral calculus completely and
61
62 5. Teaching at the Ecole Polytechnique

thoroughly during the first year, but rather to be forced to veer off and
devote time to statics and dynamics. As a result, it is necessary to go back
and teach differential and integral calculus again during the second year,
as well as dynamics. He notes that changes in the structure of the
curriculum had been requested several times; but such changes could not
be effected because the courses in analysis and mechanics were divided
between four professors-a situation that does not exist today. This fact
should allow a better arrangement.

His position was supported by his colleague Ampere and by the Inspecteur des
Etudes, Jacques Binet (2). Two meetings later, on DeceVlber 11, 1816, Ampere
submitted an instructional plan for the first year analysis course, a plan that
had been developed and written by Cauchy. This plan can be found in
Ampere's papers (3).
In line with what he had proposed at the November 15 session, Cauchy
planned that the entire course in analysis would be given during the first year
and that mechanics would be restricted to the second year.
He also proposed some important changes in the Ecole's analysis program.
As was customary, the course would begin with a section on algebraic analysis
(4). This initial section introduced three innovations. The first was an
instructional unit entitled 'Imaginary Expressions'. It was to be taught before
DeMoivre's theorem and the imaginary exponential were introduced. This
unit was followed by one on the difference between continuous and
discontinuous functions, a topic that was totally neglected in the traditional
program. Finally, there was a unit devoted to the rules governing the
convergence of series. The plan gives few details about these innovations, so it
is impossible to state exactly which specific topics would have been covered in
these instructional units. However, it can be assumed that at this stage Cauchy
already had some of the important results that were to appear in 1821 in
Analyse Algebrique. Moreover, examination of his study Sur les Integrales
Definies, which had been presented to the Academie on August 22, 1814,
suggests that by this time he had begun developing the concepts of limit and
continuity in the form that they would have in 1821 (5).
The second section of the plan was a course on the calculus of finite
differences. Cauchy meant to define the finite differences and integrals of first
and higher orders, to develop the analogy between powers and differences, first
brought out by Leibniz, and to introduce 'the simplest notions about
integration of some finite difference equations'. The interpolation formula
would end this section. The addition of a complete course on the calculus of
finite differences was an innovation at the Ecole. That was probably useful for
introducing infinitesimal calculus, as well as being of interest in its own right.
The third section, entitled 'Differential and Integral Calculus', was modeled on
the second section: Cauchy drew parallels between the two calculi, contrary to
custom. The fourth section of the plan dealt with 'The Application ofIntegral
Calculus to Geometry', a subject treated in the second year.
 5. Teaching at the Ecole Polytechnique 63

Cauchy had also written a report in which he gave the reasons for the
proposed changes. However, in spite of the fact that this position was
supported by Ampere, who said that the plan left 'nothing to be desired',
Cauchy's program was not adopted by the commission that was charged with
the task of proposing new programs to the Conseil de Perfectionnement
(Improvements Committee). Statics was reintroduced into the first-year
curriculum and only a few insignificant changes were made in the analysis
course. Although the commission gave no reasons to justify its rejection of
Cauchy's and Ampere's proposals, it is not hard to figure out what they were.
The views of the commission and of Cauchy and Ampere were diametrically
opposed. In the latters' opinion, understanding, assimilating, and using the
principles of mechanics required such a thorough knowledge of analysis that it
was necessary to devote the entire first year to analysis and to restrict the
mechanics to the second year. On the other hand, in the commission's view, the
analysis was only a tool-albeit an indispensable one-for mastering
concrete problems in construction, ballistics, engineering, design, etc. The
Ecole had been founded not for the sake of mathematics and mathematicians,
but for the training of engineers and the development of the engineering
sciences. The professors should introduce analysis in as quick and convenient
a way as possible and present instruction in mechanics and its application
parallel to instruction in analysis during the first year.
The Ecole Polytechnique reopened on January 17, 1817, in the presence of
the Duc d'Angouleme, the King's nephew. Cauchy, in agreement with
Ampere, taught the second-division (i.e., first-year) course. Later, Cauchy and
Ampere took turns teaching the second division. The one who taught the
second division also taught the same students in first division (i.e., the second
year) the next academic year. Thus, Cauchy taught the second-division course
in 1817, 1818-1819, 1820-1821, 1822-1823, 1824-1825, 1826-1827, and
1828-1829, and the first-division course in 1817-1818, 1819-1820, 1821-1822,
1823-1824, 1825-1826, 1827-1828, and 1829-1830. At first, he was busy
developing and refining his lectures. As a result, by 1821 he had considerably
slowed the pace at which he submitted papers to the Academie des Sciences.
Eventually, he devoted himself more and more to his personal research. The
curriculum's registers (registres d'instruction), kept by Jacques Binet, Inspec-
teur des Etudes, give valuable information about the evolution of his teaching
(6); from the beginning, the originality of his lectures was evident. Cauchy did
not follow the instructional program and gave free play to his inspiration
instead, changing his teaching year after year, at least until 1823.
In his second division courses of 1817 and, especially, 1818-1819, he
devoted a great deal of attention to algebraic analysis: the principal theorems
about means of several quantities; the explanation of the method oflimits and
the definition of continuous function [as far back as 1817, he stated the
intermediate-value theorem for such functions, which he used in his first proof
of the fundamental theorem of algebra (7)]; a thorough study of ordinary
functions with imaginary values of the variable; rules of convergence applied
64 5. Teaching at the Ecole Polytechnique

to the binomial series expansion and to the expansions of eX, cos x, sin x, and
log (1 + x). All these matters are summarized without any change in the
Analyse Algebrique of 1821. The section on differential and integral calculus
was much more succinct. The examination of the curriculum's registers does
not allow us to specify its standard of rigor. In his lectures, Cauchy probably
used the method of limits (the derivative defined as the limit of a difference
quotient and the integral as the limit of finite sums). He probably avoided
series expansions whose convergence was not demonstrated (as far back as
1817, Cauchy put off the study of Taylor-series expansions to the end of the
second-division course, after the lectures on integral calculus).
In his first division course of 1817-1818, he explained his method of
integration by passing from real to imaginary variables and dwelt on Euler's
method of polygonal approximation for solving differential equations. There
is no indication that he had deduced an existence theorem for solutions of a
differential equation (Cauchy's problem) from Euler's method by this time.
Cauchy gave also the first general method for solving first-order partial
differential equations, now called Cauchy's method of characteristics. He
presented a paper on this matter to the Academie on December 21, 1818 (8).
His mechanics teaching was less innovative. Cauchy expounded the
principles of mechanics in the tradition of Euler, insisting upon the notions of
force and torque and relegating the Lagrangian principle of virtual velocities
to the end of the course. In the second-division course of 1820-1821, he
developed the general theory of resultants and linear moments, which would
constitute the framework of his further research in mechanics.
Incontestably, Cauchy taught with zeal at the Ecole Polytechnique. His
best students took advantage of his teaching. Charles Combes, who had been
the major (first passed) of the class of 1818, wrote about Cauchy in 1857:
We all found that this professor was extremely energetic, good natured,
and tireless. I often heard him repeat and review, for several hours on
end, whole lessons that we had not understood clearly; we would then
become impressed by the elegant clarity of his analysis, an analysis dry
and tedious. Indeed, M. Cauchy had the genius of Euler, Lagrange,
Laplace, Gauss and Jacobi, and his love for teaching, which bordered on
pure zeal, brought with it a kindness, a simplicity, and warmth of heart
that he retained until the end of his life (9).
However, the originality of his lectures with respect to the official program
soon provoked unfavorable reaction within the Ecole. The first evidence
consists of some 'Remarks on the lack of progress in the 2nd division analysis
course', extracted from the minutes of the March 4, 1819, meeting of the
Conseil d'lnstruction (10). In these 'Remarks', Arago, the professor of applied
analysis, complained about the poor training (in analysis) of some of the
students in the second division. He attributed this lack of preparedness to the
fact that 'the analysis course is behind schedule' and 'does not keep pace with
the course in applied analysis'. In particular, he criticized the fact that the
 5. Teaching at the Ecole Polytechnique 65

lectures on the applications of differential calculus to geometry had not yet

been presented. Cauchy's position was that the delays (that Arago spoke of)
were a result of the enormous amount of material that had to be covered, and
he proposed that the material on statics be abridged, so that the material on
analysis could be completed. That, he alleged, was what had been done in
previous years. However, neither of these explanations nor the support he
received from Jacques Binet, the Inspecteur de Etudes, sufficed to calm down
Cauchy's critics at this meeting, and the assembly proceeded to examine
Cauchy's teaching methods. Adopting Arago's position, Petit, the professor of
physics, expressed concern about the delays in starting the material on
differential calculus. Specifically, the record shows that:
He [Petit] asked that this material [on differential calculus] be
presented without certain notions from algebra, which mainly had to do
with series and which, he alleged, the students would never have occa-
sion to use in the [engineering] services. Moreover, he insisted that the
method of infinitesimals be used. This method, he claimed, was one that,
in his opinion, the students seemed to be largely unfamiliar with and, at
the same time, was so useful that it should be thoroughly known (11).
Finally, the director of the Ecole repeated the criticisms that had been leveled
and warned Cauchy that:
It is the opinion of many persons that instruction in pure mathematics is
being carried too far at the Ecole and that such an uncalled-for
extravagance is prejudicial to the other branches (12).
He also asked Cauchy to adhere strictly to the official syllabus and 'to devote
time to questions at the end of class so as to introduce the students to
numerical applications'. This warning did nothing to produce a fundamental
change.
On June 15, 1820, the Conseil d'Instruction directed Cauchy and Ampere to
revise their courses (13). The students would benefit by these revisions,
particularly in those areas that the professors did not cover in class. In this
way, it would be possible to control, at least a posteriori, the content of the
courses in question. Cauchy took this opportunity to inform the Conseil
d'Instruction that 'its desires relative to the analysis course would soon be
fulfilled by the publication of a work now being printed, the first volume of this
work soon to appear'. During a discussion of the principle of continuity,
around this same time, he referred Poncelet to the forthcoming publication of
his lectures at the Ecole Poly technique (14). These two facts lead us to believe
that by the spring of 1820 Cauchy had already written what was to become
known as his Analyse Algebrique. Printing delays prevented the work's release
in time for the opening of school in the fall of 1820, and perhaps Cauchy took
advantage of this delay to add a number of new notes at the end of the work
during the 1820-1821 academic year. Ampere eagerly awaited the publication
of the material that Cauchy had prepared 'in order to help with any work that
66 5. Teaching at the Ecole Polytechnique

needed to be done on it [the course material], as well as to be sure that their

instructional approaches followed the same plant'.
The first part of the Cours d'Analyse, entitled Analyse Algebrique, was
finally published by de Bure Publications in June 1821 (15). Certainly, the
publication of this work constitutes a landmark in the history of analysis. The
task the author had set before himself was indeed ambitious. His aim, as he
explained in the introduction, was to endow proof in analysis with precisely
the same level of rigor that was used in geometry since Euclid's time (16). Such
a goal, of course, demanded that he undertake a considerable investigation of
the very foundations of analysis.
In the opening discussions, Cauchy quickly introduced the concepts of
number (i.e., the absolute measure of magnitude), real quantities (i.e., numbers
preceded by a + or - sign), and algebraic operations ( +, -, x, and -;-). But,
most important of all, he laid the cornerstones for the new structure by
giving the definition ofthe limit of a variable quantity-that is, of a quantity
that 'assumes values successively, each value differing from the other'. He
said:
When the successively attributed values of one variable indefinitely
approach a fixed value in such a way that they finally differ from it by as
little as desired, then that fixed value is called the limit of all the others
(17).

The limit concept allowed him to define infinitesimals:

As the successive numerical values of the same variable decrease

indefinitely, so as to become less than any preassigned given number, this
variable becomes what is called an infinitesimal or an infinitely small
quantity. A variable of this type has zero as limit (18).
Likewise, an infinitely large quantity is taken as a variable whose numerical
values continually increase in such a way as to become greater than any
preassigned number.
In the preliminaries, Cauchy also introduced the arithmetical notion of
mean of several quantities: the quantity x is a mean ofthe quantities a, a' a", ...
if it satisfies the inequalities
. f(a, a,,,
In ,a ,... ) -..::
~ ~
X -..:: sup (a, a,,,
,a ,... ) .

A mean of a, a', a", ... is denoted by Cauchy M(a, a', a", .. .). Cauchy showed
that

a+a'+a"+··· )
( a a' a"
b+b' +b"+ ... M b'b"b"""
if the quantities b, b', b", ... all have the same sign;
(B+B'+B"+ '''JJA + A' + A" + ... = M(.qA, J{/f,~, ...)
 5. Teaching at the Ecole Polytechnique 67

COURS D'ANALYSE
DE

L'ECOLE ROY ALE POLYTECHNIQUE;

PAR M. AUGUSTIN-Louts CAUCHY,

Inl[enieur del Ponti etehan.s'; • • , Profe6Scnr d'AnnlYI. a 1'£CoTe polylechnique,
M.embre de 1"Academie de. sciences, Che"alier de la Ugion d'hoDueur.

I:' PARTlE. ANA.LYSE ALGi:BIUQUE.

DE L'UIPRUIERIE ROYALE.

Chez DEBt;JlE frere, Libraires du Roi ct de la Bibliolhequc du Roi..

rU t) ::ierpente. n, · 7.

1821.

Title page of the Analyse Algebrique, 1821. Published by permission

of the Ecole Poly technique.
68 5. Teaching at the Ecole Poly technique

if A, A', A", B, B', B" all are (positive) numbers;

exa + ex' a' + ex" a" + . .. = (ex + ex' + ex" + ... )M (a, a', a", ... )
if the quantities ex, ex', ex", ... all have the same sign.
Cauchy skillfully used these means, instead of inequalities, as powerful tools
to prove many theorems of his analysis course: the mean-value theorem,
theorems on the convergence of series, the proof of the existence of the definite
integral of a continuous function and of a solution of a differential equation.
Chapter I was devoted to real functions. Cauchy defined a function in a
rather general way as a variable that can be expressed by means of one or
several other variables, which he called 'independent variables'. If the
function is multivalued, Cauchy denoted itf((x)). He then examined several
different types of functions: explicit, implicit, simple, algebraic, and trigo-
nometric.
In Chapter II, he reexamined infinitesimals. Using the notion of limit, he
compared infinitesimal quantities in terms of orders of magnitude and then
introduced the important notion of continuity on an interval. This concept, he
asserted, 'should be ranked among the matters that are closely connected with
the investigation of infinitesimals'.
Cauchy gave the following definition of the continuity of a real function of
one variable:
Let f(x) be a function of the variable x and suppose that, for each value
of x between two given bounds, this function constantly takes one finite
value. If, from a value of x between those bounds, one attributes to the
variable x an infinitely small increment ex, the function itself will receive
as an increment the difference
f(x + ex) - f(x),
which will depend at the same time on the new variable and on the value
of x. This being granted, the function f(x) will be a continuous function
of the variable x between the two assigned bounds if, for each value
of x between those bounds, the numerical value of the difference
f(x + ex) - f(x) decreases indefinitely with ex. In other words, the
function f(x) remains continuous with respect to x between the
given bounds, if, between these bounds, an infinitely small increment
in the variable always produces an infinitely small increment in the
function itself (19).
The continuity of the function was established at each point of the interval.
Cauchy used the same infinitesimal ex, however, and did not distinguish
between uniform continuity and simple continuity. This confusion is one of the
most serious flaws in Cauchy's development of the course. Thus, when he
extended the notion of continuity to functions of several variables, he gave a
theorem that is only true for a function uniformly continuous with respect to
each variable:
 5. Teaching at the Ecole Poly technique 69

34 o RS l)'''~ALYSE. 3.3
nmh.erlque de celie variable, Ie J1ol!J1lome fim't plI1" port a z entre 16$ limilas donmics I $; I entre Ct!$
itre COll$lammcllt de. -mime si;{lIC que SOIt p,onnier limiles, fU~ ffccroissement inJinimelll petil do La !la-
ttrme. ritlb/c produil (ott/oun: till fll't'roi$srmCllt ill..fillimtntt
petit de fa fiMClioll ~"c-memt.
S. 2:' Dc fa cOfltinuili dct FDncti"II~ . 0 .. dil CllcOre que I. fonction / ( .• ) ..I, d,,,,.
Ie voisinRge d'ullc ,"alcur pW'licuficre l\ttcibuce ;l
Parmi Ics objets qui sc- r.tttachcnt 11 In. considc~
In ,':triable oX, looeti~1I contillue de ccue v:1ria.ble.
ration des infinimcnt petits, on doit pincer les no- tontcs les riJi~ ~u'clre est conlinuc cntt''C deux limites
tion5 rdnti\'es R la conlinuilc ou it [a discominuitc de .c meme tl-cs-r3pprochccs. qui rcnfcrmcot fa
I

des {anctions. E::tal1linons dtuhord sous cc point de valcul'" dunt il ~·asit.

,,' llC Ics foncttons d'une sculc ,'al'iablc.
Enfiu, 10l"Sqll'U il 0 fonctioo f (.t') ec sc cI'efre eon-
Soil [(.r) unc fonction de III yariable :It , ct tinue dUllS Ie ,'oisinagc J'unc "Rieur p:U1iculicre de
tiOpposoos ~Iue 1 pour chuque ' Ialeur de x intc.,ne. la "AI·inhlc :C, 011 dit 'i,l'cllc tle\·jimt :llor d,:fcoll-
diairc entre deux limifc~ donnees J ccltc rOllctiou fj'll«C, cl qu'il Y u pour cetlc \lalcur l)articuliere
od;nette constamment une ,,,leur unique: ct finie. solutiou tllJ confinu;le-.
Si, en parlant d'une "nleur de :r complisc entre CC$ O'npre. cCS e.plio",io"" il se..... fa tile de J·ecolI·
limite , on ttltl'ihue ~ lu vllriable ~ un aCClooisscmcnt nlllh"C ('lure queUes limitcJ> unc fonction donnce
inJinimcnt petit a.} In f011clion cllc- rucme l'C(:eYI'l de fa. variable or C5t continue par ..apport ;\ cede
.:rour Ilceroisscment I" difference ,·:triable, Ainsi . par cxcmplc I III fOlletion ~jlJ, x,
f(:e-+-«. )-f(;r;) , 1l(lme.tI.ant 1>our cha.'lue ,·"lour pal'ticulierc de Ill..
vanllilic .c Ulle \'u.fcUI' unique ct finie 1 sem ront-inue
qui dcpendra ell meme lemps de I. nouvclle :.•.
entre deux {jrnitcs ~uelcoJJques de cett,c "'\nri"'hfe.
riable c<. cl de la volel.t de x. Ccl. pose, I. foncboD
a,tendu que lit "-8_Ic.ur llumcnquc de "in. (7"")~ et
/(:1:) sera, ~ntre Jes deux limites ft is:n~5 it lsa. v~.
par sui Ie celie de I~ difference
liable :e, ronetion co"li,jue de ceUe finable I $1.
pou~ cbo.que valeur de x intcrmediaire entre ccs ..5in.(;r-t-a.) -lIin,.%'=:1 5in_( ~ ct.) cos, (.e ....... {- «.).
limite" I" valeur nunuSricl uc de In tlifference decroi~n t indefiniment n,'cc. celie de ct.. qucllc que
f(x+c<.)-f(z) soil d'oillcurs I. vnleur fonie que roo nltribuc a z.
deerol! ind,,6nimeh! ."C. celie de «.. En "'a"!reS
termes, /" fonetion /(s) restera coniimi. par ,.ap-
En g~no ... I, si ron envisage sous Ie ..ppor. d. 1&
c-OntJHuite les ollze fO"C00I15 simples que nous IlYons
c

Definition of a continuous function, Analyse Algebrique, 1821, pp. 34- 35. Published by
permission of the Ecole Poly technique.

If the variables x, y, Z"" have the fixed and determined quantities X, Y,

Z, .. , as limits and ifthe function f (x, y, Z" , • ) is continuous with respect
to each variable x, y, Z, . , , in the neighborhood of the system of values X,
Y, Z" " , f (x, y, z, . .. ) will have the limit f (X, Y, Z" . ,) (20),
Finally, he stated the intermediate-value theorem (21) and sought to determine
the value of certain functions at singular points by continuous extension,
The notions Cauchy presented in the first part of Analyse Algehrique were
not new (22), Following D'Alembert, several mathematicians had attempted
to base analysis on the limit concept. Some, notable are Lhuillier and Carnot,
had obtained results that anticipated Cauchy's. Furthermore, the method of
limits was the approach used in teaching calculus at the Ecole when Cauchy
was a student there, In the meanwhile, in 1811, the Conseil d'Instruction had
renounced this method, which 'to be truthful, lacked rigor and was not
amenable to applications' and reinstated the former method of infinitesimals
as the approach for teaching calculus, Mathematicians did not ignore the
properties of continuity of the functions commonly studied; but, in spite of the
works of Arbogast, they continued to use Euler's definition according to which
70 5. Teaching at the Ecole Polytechnique

a continuous function is a function that is defined by a single analytic

expression (23).
While Analyse Algebrique made use of then current notions on limits,
continuity, and infinitesimals, it broke with the traditional texts treating such
concepts. On the one hand, the definitions given by Cauchy in Analyse Algebrique
are much clearer, more precise, and more general than those that had been
used up to then. On the other hand, the relations between the basic concepts
are more neatly exhibited. By defining infinitesimals as variables that converge
to zero, Cauchy, as he explained in the foreword to Calcul Infinitesimal,
reconciled the rigor of taking limits (which was given there for the first time in
terms ofinequalities) with the simplicity of the results obtained by considering
infinitesimals directly (which he generally used in his course because of the
requirements of the official program of the Ecole) (24).
The set of basic notions thus given formed a logical, coherent basis on which
it would be possible to erect the structure of the course in analysis.
Chapter V was devoted to five functional equations:
fjJ(x + y) = fjJ(x) + fjJ(y)
fjJ(x + y) = fjJ(x) x fjJ(y)
fjJ(xy) = fjJ(x) + fjJ(y)
fjJ(xy) = fjJ(x) x fjJ(y)
fjJ(y + x) + fjJ(y - x) = 2fjJ(x) fjJ(y)(fjJ continuous).
Cauchy used the second of these for studying the binomial series in the next
chapter.
In this chapter (Chapter VI), Cauchy took up the theory of series. He
defined a series as a sequence of partial sums,
sn = Uo + Ul + ... + Un-I·
The sum of the series is equal to

when this limit exists. In that case, the series is said to converge; otherwise, it
diverges. If the series diverges, then it has no sum. This is the reason that
Cauchy devoted the main part of the chapter to establishing certain tests for
the convergence of series. The most general of these tests, known today as
'Cauchy's test', was presented in the following terms:
In order that the series u o, u 1 , U2' U3' ••• Un , un + 1 , ••• , etc... shall be
convergent, (... ) it is necessary that, for increasing values of n, (... ) the
sums of the quantities Un' Un + 1 , Un + 2 , etc .... taken from the first one, in
any desired number, end up by constantly assuming numerical values
that are less than any assignable limit. Conversely, whenever these
different conditions are fulfilled, the convergence of the series is
guaranteed (25).
 5. Teaching at the Ecole Polytechnique 71

Cauchy stated, but did not prove, the converse. In the same section, Cauchy
stated that the sum of a series of continuous functions is continuous. In 1826,
Abel was the first to note the insufficiency of this statement by giving the
famous counterexample

f(x) = f sin(2n + 1)x,

n=O 2n + 1
which is discontinuous at 1t (26).
Cauchy proved the rules governing absolutely convergent series, 'Cauchy's
rule' and 'D'Alembert's rule', and the ones regarding sums and products,
etc. He first gave a rule for determining the radius of convergence of power
series, and he showed the uniqueness of the power series expansion of a
continuous function. Then, he investigated the series expansions of some
important functions in their circle of convergence: Newton's binomial series,
the exponential, the logarithm of (1 + x). But, it was not until 1831, with the
theorem of Turin, that he succeeded in determining the context of the complex
function theory, the conditions under which a function shall have an infinite
series expansion.
Chapters VII through X were devoted to the theory of imaginary
expressions. Cauchy took an imaginary expression as a symbolic expression
that has no meaning in itself but is equivalent to two real quantities. In the
same way, he regarded an imaginary equation as "the representation of two
equations connecting real quantities." In order to obtain the product of two
imaginary expressions, he used the ordinary rules of multiplication in algebra
by treating F-T as a real quantity with square - 1. Cauchy examined the
trigonometric form of imaginary expressions, as well as the integer powers of
such expressions. He also investigated fractional and irrational powers of
imaginary expressions, paying particular attention to the nth roots of unity.
Finally, Cauchy defined imaginary functions of the form tfJ(x) + x(x) F-T,
where tfJ(x) and X(x) are real (complex functions of a real variable). He also
extended the definition of eZ , sin z, and cos z to complex z by means of
convergent 'imaginary series' with the general term Pn + qn F-T and defined
the multivalued functions log ((z», arcsin ((z», arccos ((z», etc. (27).
Analyse Algebrique had a considerable influence on young mathematicians.
'It should be read,' wrote Abel in 1826, 'by any analyst who likes rigor in
mathematical investigations' (28). Nevertheless, it was not well-thought of at
the Ecole and was never used by the students there for the following reason. A
few months before its publication, a serious incident took place in Cauchy's
second-division course. On April 12, 1821, at the end of a lecture that had gone
into overtime, the incident occurred just at the moment the class monitor
turned his back. Cauchy was booed and hissed by five or six students whose
identities were never determined by the administration.
Baron Bouchu, the Governor of the Ecole, took the matter seriously. He
submitted a report on it to the Minister of the Interior (29), and he ordered the
72 5. Teaching at the Ecole Poly technique

class leaders to apologize for the class' misconduct to Cauchy, who refused to
continue teaching the class following this affront. In his report, Baron Bouchu
blamed the students and the professor equally, raising questions about
Cauchy's method of teaching. Cauchy, he felt, respected neither the official
program nor the official schedule:
It cannot be denied that M. Cauchy, by his own stubborness and by
extravagantly extending his class sessions [beyond the allotted time],
unwisely pushed his students to the thoughtless insult that took place ...
I can no longer hide the fact that for the past five years, he has been given
many warnings to simplify his teaching methods so as to bring them into
line with the official program.
The Minister of the Interior, however, did not find these explanations
convincing. The whistling and booing, he believed, was a manifestation of the
students' hostility toward a teacher who did not bother to hide his extremely
royalist views. Following the assassination of the Duc de Berry (the Dauphin)
in February 1820, the government had once again curtailed civil liberties. The
liberal opposition, always very popular among the students, as well as in the
army where the Charbonnerie movement had taken roots, became radicalized.
Cauchy, of course, was not highly regarded by the students at the Ecole
Poly technique, because they had not forgotten that at the Academie he had
replaced Monge, the scholar who had founded the Ecole and whose memory
was still held sacred by the students. The affair was discussed by the Conseil
d'Instruction on April 17:
The Inspecteur [des Etudes] recalls that at the meeting before last, M.
Cauchy stated that, after 53 [teaching] lessons, he still needed 7
additional ones to complete the analysis course. The former number was
not enough for him, and a total of 66 was required instead of the 50
contemplated by the schedule. This situation meant 'slow-down' of some
five weeks in the combined progress of the course in analysis and the
course in mechanics ... M. Cauchy stated that he would be able to give
his reasons for an increase in the number oflessons in the analysis course
at the end of the year. He stated that the amount of material for the first-
year course had been increased and that he had not gone beyond what
was required by the curriculum. He recalled that for five years it has not
been possible, either for him or for M. Ampere, his colleague, to keep within
the prescribed number of lessons for the first-year course; while, for the
second-year course, the limits outlined were always adhered to (30).
No doubt, Baron Bouchu was anxious to protect the Ecole from the
government's wrath, and accordingly, on April 21, he submitted a second,
much more severe report concerning Cauchy (31). In this document, the Baron
explained that the students had complained to their parents about the analysis
course, but had not dared to register their grievances with the authorities, for
fear of being labeled agitators and troublemakers.
 5. Teaching at the Ecole Polytechnique 73

These considerations explain, on one hand, why the students were

displeased; and, on the other, it is easy to see that they should have felt
discontent. In fact, whatever their attitude towards the course, they
cannot lose sight of the fact that the material [in the course] is essential, if
they are to pass the examinations, which are exclusively based on the
syllabus. They are given copies of the syllabus, and it outlines the
required rules of instruction, just as it outlines the duties of the professor.
These course syllabi are the fruit of many years of experience, and they
prescribe, for M. Cauchy's course, a total of 85 lectures, 50 for analysis
and 35 for mechanics. The neglect of the material on mechanics as a
result of an excess of lectures on analysis must not be allowed, given
that the students must still answer all the questions dealing with the
other part of the course even if it has not been entirely taught. All of
the students willingly acknowledged M. Cauchy's enthusiasm and
appreciated his outstanding abilities. However, though he [Cauchy] is
not the only professor here who possesses superior abilities and excellent
principles, he is the only one who perseveres in disregarding the official
syllabus...
The incident of April 1821 had some important consequences. It clearly
showed that even if Cauchy was able to interest and inspire the most gifted
students, he was unable to adapt his course to the level of the majority of the
students who came to the Ecole with only the most rudimentary knowledge of
mathematics and who were required to master a good number of topics and
concepts in two years of study. Testimonies of contemporaries corroborate
this impression. In his speech at Cauchy's funeral, Charles Dupin observed
that:
Once our illustrious confrere had been appointed to the Ecole Polytech-
nique, he was not content to follow the programs and lectures that the
eminent professors who preceded him had devised. He restructured, so
to speak, the material on algebraic and infinitesimal analysis along lines
and by methods that he deemed appropriate. He was found to be-and
why should it not be said here? For it is the only reproach that seems
possible to make, and it is a reproach which is, as it were, a praise. He was
found to be too learned, too brilliant, for the students who came to him in
large numbers expecting to learn the practical material required for the
public services and not his thoughts and erudition at the Institut. These
practical applications demanded above all else that his methods and
means be clear, simple, common, and quick first, as military and civil
requirements often are (32).
This is confirmed by a comparison that J. Bertrand drew between Poinsot and
Cauchy:
When Cauchy replaced Poinsot, the students were divided [in their
opinion]. 'Poinsot did not teach us anything', remarked the students
74 5. Teaching at the Ecole Polytechnique

who liked the new course; 'Cauchy will disenchant them with science
forever', said Poinsot who never bothered to hide his views... Poinsot, it
is true, did not teach many things in any given lecture; but what he did
present was presented very well indeed! Cauchy, on the other hand, was
forever going beyond bounds, and only a few, very gifted students could
understand him. This elite indeed found him praiseworthy (33).
Cauchy's lectures were too ambitious to be presented in the time allotted for
his courses in the official schedule. Consequently, he was forced to sacrifice the
applications and exercises that the Conseil contemplated in its official
program. The writing and editing of Analyse Algebrique, which took an
inordinate amount of time, had further upset the balance between algebraic
analysis and the remainder of the course. Applications was the area suffering
the most. This lack of balance created a situation that was unacceptable to the
students and to the Conseils at the Ecole.
Thus, in November 1821, when he resumed teaching the second-division
course, Cauchy had to reduce the number oflectures that preceded differential
calculus (34). Abandoning publication ofthe remainder of the Cours d'Analyse,
only the first part (Analyse Algebrique) of which had appeared in 1821, Cauchy
began to write up summaries of his lectures. He had been ordered to do so by
the Conseil d'Instruction:
In order to satisfy the repeated requests of the Conseil d'Instruction, the
professors of analysis and mechanics [Cauchy and Ampere] are now
busy writing summaries of the most difficult points in their courses. We
regret that it took so long for the work in the first-division classes to get
started and that it has not been carried out quickly. However, these
summaries should be very useful to the students and will help them a
great deal (35).
These summaries (in a volume entitled Resume des Le{:ons Donnees al'Ecole
Royale Poly technique sur Ie Calcul Infinitesimal) were published together on
August 11, 1823, by Bure Publications (36).
The first part of Calcul Infinitesimal of 1823 was devoted to differential
calculus. After a brief exposition in the first two lectures of the basic concepts of
analysis, which he had developed in the initial sections of Analyse Algebrique,
Cauchy introduced the notion of the derivative of a function of one variable
y = f(x) in the third lecture. The term derivative and the accompanying
notation f'(x) were borrowed from Lagrange (37). Lagrange, however,
developed a theory of differentiable functions that was always assumed to be
analytic, while Cauchy used the method of limits: iff(x) is continuous, then its
derivative is the limit of the difference quotient:
Ay =f(x + i) - f(x)
Ax
as i tends to O.
 5. Teaching at the Ecole Polytechnique 75

RESUME DES LEGONS

DONNEES

A L'ECOLE ROYALE POLYTECHNIQUE,

SUR

LE CALCUL ' INFINITESIMAL,

PAR M, AUGUSTIN-LoUIS CAUCHY,
Jog~oieur de. Ponts-et-Chau"ees, Professeo, d'Analyse 'a l'Ecole royale Polylcchnique,
Membre de l'AcaMmie des Science., Chevalier de la Legion d'nonneur.

TOME PREMIER.

A PARIS,
DE L'IMPRIMERIE ROYALE,

Chez DEBURE, fceces, Libraires du Roi et de la Bibliolhe<l\le du Roi,

rue Serpente) n," 7.

Title page of the Caicull,y,nitesimal, 1823. Only the Tome Premier has been published.
Published by permission of the Ecole Poly technique.
76 5. Teaching at the Ecole Poly technique

TROIS1EME .LEetON.
D.i.ri"ees ties fpnc,jollS d'mlt StJI!t Variaole.
, ';11 i ;; i

LORSQUE la f0-l.1ction y=/{«) re\t, ' i01l!1.IlUe .entre de ux limites

donnees de la variable ~. , et que I'on assigne"a cetle variable une valeur
comprise entre les deux limites dont il s'agi!. un aceroissement infini-
ment peti·t, attribuc A In va;ial>le, produit un aceroissement iFifiniment
petit de ·Ia foncdon el[e-m~me. 'P.ar conscquent. si I'on pose a[ors t. :<=i,
les deux 1ermes d-ll rapport (lUX difflrellus
~_ f(x+i)-flxo)

seront des quantitcs infiniment petites. Mais. tandi~ .que ces deux termes
s'approcheront inJefiniment et s~ml.ltnn6mellt de la Jimite zero. Ie rapport
lui-m~me poun'a converger vers une autre limite. soit positive. soit ne-
gative. Celte Jimite • Jw:tqu!elle .ex·isle • a ·u.ne valeur de.te,rl,l;ltnee. !Hour
ehaque valeur particuliere de x: mnis elle vade avee x. Ainsi. par
exem.pJe.• .s.i J'o.n prend /{x)=x"', III desi.gnallt lIl1 nombre _l:n~i~. Ie
rapport entre les differences infinimen( petites sera
, ("'.;-il~~"ft
I
= mx-'"-' -+- rn(m-I)
1 • .1
xm- 'J -I- . : .-I-i m-,

~t -il .q4il<l pftwJ.i:I'nite:ia q \lf1ntite-/1I~""' .• e'es(·a,dtr~ , line l}o~,\'~lleJo,ne~i~w

de la variable x. II en.sera de meme en genera!; seulement, la forme de
la fonetion l;Iouvelle qui servira de Ii ;ite au rapport f(x +- i ~ - fIx)
In fonetion ·pmposee J f( x). PQlIr jnpiquer
. dc.!pend~a .de la ·forme .de =
e~ue dc!pendan"e , on donn~ a la nouvelle fonction Ie nom de finelioll
il/,;br.ee, et on la dcsigne, a l'!\ide d'un Rccent, -pllr ~a notation
l ou f' (x). I

-Pitns l;ne.chercne .des .deliiv.ces des Ionctipns .d:une-se\lle v..llcii\ble ,\. ,

if est utile de dj~~llgtI6r les fonctions que 1'0n nomme simples, et que
14(0'" d, M. C.U(-!y. 8

Third lesson of the Calcul Infinitesimal, 1823. Definition of the derivative of a

continuous function. Published by permission of the Ecole Poly technique.

Cauchy had no doubts about a continuous function being differentiable.

The notion of the differential of a function of a single variable was linked in a
natural way to the notion of the derivative: the differential dy of y = f(x) is not
an infinitesimal but the finite expression l' (x)dx, wherein dx is an arbitrary
constant (38).
 5. Teaching at the Ecole Poly technique 77

In the fifth lecture, Cauchy defined the differential of a complex function of a

real variable u + v.j"=l by the formula du + dV.j"=l.
After having applied the notion of the derivative to maxima and minima
problems and to the determination of expressions involving indeterminate
forms, he presented the mean-value theorem in the seventh lecture: iffandJ'
are continuous between x and x + i, then
f(x + i) - f(x) = iJ'(x + (}i) (5.1)
He constantly used this theorem in the remainder of the lectures. But,
although he made use of the very first 'epsilonizations' of the concept of the
limit in this proof, Cauchy lacked the basic notion of uniform continuity, a
notion that was absolutely necessary if his argument was to be completely
rigorous (J' must be uniformly continuous between x and x + i). In the
subsequent lectures, Cauchy generalized his definitions of the derivative and
differential to the case offunctions of several variables. But, the proofs for these
generalizations were far from rigorous. Moreover, his generalizations did not
question whether a function with partial derivatives with respect to each of its
variables is indeed differentiable. He simply took this for granted.
The second part of the Calcul lrifinitesimal was devoted to the integral
calculus. Cauchy began by giving a new definition of the integral of a
continuous functionf(x) in the twenty-first lecture: by transforming Euler's
method of approximation into an existence theorem, he defined the integral

f:/(X)dX

[the f:o sign had been proposed by FOUrier]

as the limit, when the interval [xo, X] is indefinitely subdivided, of the sums

or

where
X1,X 2, ... ,Xn - 1 lie between Xo and X.
But his proof of existence, like his proof of the mean-value theorem, fell
short because of a lack of the notion of uniform continuity. In the following
lecture, Cauchy derived from the definition of the definite integral a proof of
the integral mean-value theorem:

(xo ~ ~ < X, cP ;;:: 0). (5.2)

78 5. Teaching at the Ecole Poly technique

Cauchy generalized his definition of the definite integral. First, he introduced

the notion of imaginary integral:

I x u(x) + v(x) j=1 dx = IX u(x) dx + IX v(x) dx j=1.

~ ~ ~
(5.3)

In the twenty-fourth lesson, he defined improper definite integrals and

investigated the case where the function to be integrated is infinite at some
points of the interval of integration by using the notions of principal value of
an integral and of a singular integral (39).
Later, in the twenty-fifth lecture, Cauchy proved the fundamental theorem
of the integral calculus, which reduces the problem of evaluating a definite
integral to that of calculating the primitive of the function to be integrated. In
order to prove this theorem, he showed that the indefinite integral

y(x) = f:of(e)de
is the solution of the differential equation :~ = f(x), satisfying the initial
condition y(x o) = O. This procedure initiated the study of the ordinary
differential equations, which was restricted to the first-division course.
The concluding material on the integral calculus dealt with MacLaurin's
and Taylor's formulas. By using the equation

dd fX
(x - z)m f(z)dz = m fX
(x - z)m-l f(z)dz
x Xo Xo

Cauchy proved, in the thirty-fifth and thirty-sixth lectures, the MacLaurin's

and Taylor's formulas with integral remainder
x x2 xn - 1
F(x) = F(O) + TF'(O) + 2! F"(O) + ... + (n _l)!F(n-l)(O)

+ fX(X-Z t - 1 F(n)(z)dz
o (n - 1)!
and
h h2 hn - 1
f(x + h) = f(x) + If'(x) + 2! f"(x) + ... + (n -1)! j<n-l)(u)

+ fh (h
o (n-1)!
-
zt- 1 j<n)(x + z)dz,

where F (resp. f) and all of its derivatives up to order n were implicitly

supposed to be continuous in the neighborhood of 0 (resp. x).
From the MacLaurin's and Taylor's formulas with an integral remainder,
Cauchy derived the MacLaurin's and Taylor's formula's with the Lagrange
remainders:

xn F(n) ((}x) and hnf(n)(x + (}h) (0 ~ 8 < 1)

n! n!
 5. Teaching at the Ecole Poly technique 79

and with the Cauchy's remainders:

x(x - (}xt- 1 F(n)((}x)

(n - 1)!

and
(h_(}h)n-l
(n _ 1)! hpn)(x + 8h) (0 ~ (} < 1).

Afterward, in the thirty-seventh lecture, Cauchy obtained the MacLaurin and

Taylor series expansion of indefinitely differential functions F in the neighbor-
hood of 0 and f in the neighborhood of x:
X x2 xn
F(x) = F(O) + -F'(O) + -F"(O) + ... + -F(n)(o) + '"
1 2! n!

and
n
+ h) = f(x) + -h f'(x) + -h h
2
f(x f"(x) + ... + _pn) (x) + ...
1 2! n!
by assuming that the remainder of the MacLaurin's and Taylor's formulas
tends to 0 as n tends to infinity. This condition of convergence is obviously
necessary, but it is not sufficient. In the thirty-eighth lecture, Cauchy gave tests
of convergence for the MacLaurin and Taylor series expansions. Moreover, he
observed that a convergent MacLaurin series can have a sum different from
the function it represents.
In fact, Cauchy had discovered a counterexample, the function e- 1/ x ',
continuously extended to 0, whose MacLaurin series expansion yields the null
series. He first gave this counterexample in the note 'Sur Ie developpement en
series et sur l'integration des equations differentielles', which he presented to
the Academie on January 22,1822. The context was a critique ofthe method of
the undetermined coefficients used in the theory of differential equations:
Cauchy deduced from his counterexample that the MacLaurin series
expansion of a function f(x) is the series expansion of many other func-
tions, for instance, f(x) + e- 1 / x2 and, consequently, that the method of the
undetermined coefficients could not guarantee the generality of the solution of a
differential equation. The mathematicians of the era, such as Lagrange,
Laplace, and Lacroix, regarded the Taylor series expansion as the basis of
differential calculus. That Cauchy should have relegated his treatment of this
topic to the end of the discussion on integral calculus is closely connected with
his adoption of the method of limits as the cornerstone of analysis.
Until Cauchy, in fact, it was thought that a function continuous in the sense
of Euler could always be expanded in a power series whose coefficients were
determined by use of Taylor's formula. Lagrange made use of this Taylor series
expansion in his Theorie des Fonctions Analytiques of 1797 to define the
derivative: if f(x) is a given function, called the primitive function, then the
80 5. Teaching at the Ecole Poly technique

derivative f' (x) is the coefficient of the second term in the infinite series
expansion of f(x). The attention that Cauchy gave to problems relating to the
convergence of series, from 1815, and later, his discovery of the counter-
example e~ l/X', convinced him of the insufficiency of Lagrange's point of view
as the basis for analysis (40). However, in light of the importance of the
MacLaurin's and Taylor's formulas in analysis, Cauchy decided to include a
new proof of them that did not use the integral calculus. This proof directly
led to formulas with Lagrange remainder. Hence the MacLaurin and
Taylor formulas could be presented in the course on differential calculus
(41).
Far from pleasing the Conseils at the Ecole, Cauchy's Calcul Infinitesimal
provoked the same criticisms that his Analyse Algehrique had. On December
29, 1823, the matter came before the Conseil de Perfectionnement (42). A
discussion ensued, and the outcome was the appointment of a commission by
the Minister ofthe Interior. Laplace, Poisson, and Prony were its members. Its
purpose was to work 'in concert with the professors of analysis and mechanics'
(i.e., with Cauchy and Ampere). This commission was authorized 'to make
such modifications in the course brochures (that Cauchy was preparing) that it
might deem necessary, so as 'to increase their usefulness', and to present its
results at the next meeting of the Conseil de Perfectionnement, in order 'that
revised brochures could be distributed to the students the next year'.
To expedite the commission's work, Cauchy and Ampere were to hand over
their brochures to the commission, which functioned as a veritable board of
censorship, during the first semester of 1824. Ampere furnished part of the
material on the second-division course in analysis (43). Cauchy, meanwhile,
stalled for time. He was late in submitting the material for the first-division
course. On May 6, 1824, he explained to the Conseil d'Instruction that 'he
hopes to be able to devote more time to it [the rewriting of the material for the
second-year course], since the revision of the study that he has been working
on is now finished' (44). 'But', he went on, 'there might be another delay
because the royal printing service has all of its presses busy printing material
on the laws of finance'. In effect, then, between May and July 1824, 13 lectures
for the first-division course were printed. (As a matter of fact, only part of the
thirteenth lecture was included.) This material made up the first part of the
second volume of the Resume des Le(:ons Donnees a£ Ecole Poly technique sur
Ie Calcul Iriflnitesimal. However, the printing of this material was suddenly
interrupted during the summer of 1824, so that Cauchy'S course material, like
Ampere's, remained incomplete (45).
These 13 lectures dealt with ordinary differential equations. Just as in the
course on integral calculus, where Cauchy had transformed a method of
approximation into a definition of the integral, he gave here the first rigorous
proof of the local existence and uniqueness of the solution of a first-order
differential equation y' = f(x, y), satisfying a given initial condition y(x o) = Yo
by the method that would later be called the Cauchy-Lipschitz method. This
method, derived from Euler's method of polygonal approximation, was based
 5. Teaching at the Ecole Poly technique 81

on a thorough study of the differences equations:

YI - Yo = (Xl - xo)f(xo)
Y2 - YI = (X2 - xl)f(x l )
. .
. .
Y - Yn-l = (X - xn- df(x n- l ),

when X tends to Xo (46).

Nothing in the minutes of the meetings of the Conseil d'Instruction or the
Conseil de Perfectionnement explains the sudden interruption of the printing
ofthe material from Cauchy's and Ampere's lectures. Thus, lacking evidence to
the contrary, one suspects that the commission, highly displeased with the
material that Cauchy and Ampere had already handed in, was responsible for
the interruption of the printing. Laplace even asserted on one occasion that
'some of the material handed in (to the commission) by one of the professors of
analysis [Cauchy] was so unintelligible to him that he could only make sense
of it after a third reading' (47). In light of the disagreement between Cauchy
and Ampere and the commission, the printing would not be resumed during
the following year.
During the academic year 1824-1825, Cauchy's performance as teacher of
the second-division course was continually the subject of criticism in the
Conseil d'Instruction. The harshest critic was Arago, the professor of applied
analysis. Arago was in a good position to judge the level of training in
mathematics of the students coming from the Ecole, since he was an
admissions examiner at the Ecole du Genie et de l'Artillerie in Metz (48). The
Inspecteur des Etudes, J. Binet, also reproached Cauchy with digressions from
the official program, as well as omissions, particularly where geometry was
concerned. Cauchy replied to his critics, asserting:

It should be noted that the lectures that were devoted to this material
were not a loss for the students insofar as the discussion of these topics
would serve to shorten the treatment needed for other topics later on.
This will not result, therefore, in the need for additional lectures.
However, it will not be possible to cover the entire course in the number
of lectures alloted by the official program. The impossibility of doing so
is a point that this teacher has continually raised over a ten-year period
and that he has again striven to demonstrate by detailing the various
parts of the course and by indicating the number of lectures that he
devoted to each part, a number that he deemed to be too small. Over the
preceding years, he [Cauchy] was granted a quarter-hour extension on
the half-hour that is devoted to the question-answer period that
precedes each lecture. This year, he used the entire half-hour for the
question-answer period; and this, accordingly, diminished the length of
the lecture by one fifth. Thus, it is impossible to cover the entire course
within the time limits set by the official schedule (49).
82 5. Teaching at the Ecole Polytechnique

When the school opened in 1825, the Conseil d'lnstruction decided to make
significant changes in the analysis program. Real changes in the program were
what Cauchy and Ampere had been requesting. They had complained for a long
time that the program was unwieldy and that it was impossible to cover all the
material in the allotted time. Now there would be changes, but only at a price!
A very high price it would be, too, because, 'with these changes, there will no
longer be any preliminary algebraic material before differential calculus,
which will begin the course immediately' (50)! Thus, the changes conceived by
the Conseil d'lnstruction struck a blow at Cauchy's very method of teaching,
for the avowed aim ofthe official scheme was nothing less than a return to the
old way of studying calculus by means of infinitesimals, a method that the
Conseil regarded as not only simpler but more efficient as well.
Astonished and offended to see the heart of his instructional approach
reduced to nothing, Cauchy presented his arguments to the Conseil
d'lnstruction. The minutes of the meeting include verbatim his reply to those
who had criticized him and his work:
To be sure, there are individuals who believe that certain portions of the
analysis and mechanics course, particularly the first-year course,
demand too much of the students. Whether that belief is well founded or
not, it has nothing to do with the professor's method. It has rather to do
with the large number oftopics that have been added to the course since
the reorganization of the Ecole, to the program for the first-year course,
and to the level of rigor that the professors require in their proofs and
arguments. By simply comparing the new methods [i.e., Cauchy's
methods] to those formerly used, anyone can see, without too much
trouble, that the new methods are simpler [than the old ones], when
they are not more rigorous. As for the remainder, I suggest that if some
rigor be sacrificed, as one of the analysis professors proposed at the
meeting last November 24th (51), then experience will soon show that
the new methods, far from leaving the students in the dark, will enable
them to learn in less time and with less effort all that they might have
learned by the former methods. That much can easily be assured. Thus,
for example, the professor of analysis in the first-division course was
able to explain and develop the second part of the infinitesimal calculus
in fewer lectures than the (official) program allots to it (52).
After this last speech in his own defense, Cauchy was obliged to submit to
the Conseil's demands and to give up the publication of Calcul Infinitesimal
(53). In response to the criticism that he was sacrificing geometric applications
to pure analysis, Cauchy undertook the publication of a three-volume work
entitled Le~ons sur les Applications du Calcul Infinitesimal ala Geometrie (54),
in which he presented the mathematical framework of his work in mechanics.
The first volume, devoted to differential calculus, appeared in July 1826. In this
elementary treatment of differential geometry, Cauchy developed, among
other things, the theory of the radii and centers of curvature of an arbitrary
 5. Teaching at the Ecole Poly technique 83

curve and the theory of the orders of contact of curves and of curved surfaces.
In the fifteenth lecture, he examined problems relating to the centers,
diameters, and axes of curved surfaces, paying special attention to the
quadrics. This study led him to the problem of determining the three
eigenvalues of a linear symmetric mapping of a three-dimensional space. These
values, he showed, are always real (55). In addition, from May 1826 to April
1827, he published numerous studies in Exercices de M athematiques that were
based on his mechanics teaching at the Ecole and at the Faculte des Sciences
(56).
Starting in November 1826, the courses in analysis and mechanics-
particularly Cauchy's courses-were closely supervised by the Conseil de
Perfectionnement. Each year, the examiners of the graduates made a report to
the Conseil on the courses taught by Cauchy and Ampere. Prony was
responsible for 'overseeing' Cauchy's teaching. He was generally critical in the
series of annual reports that he sent in for 1825-1826, 1826-1827, 1827-1828,
1828-1829, and 1829-1830. In his report on the academic year 1825-1826,
Prony criticized the excessive sophistication of Cauchy's teaching:
I will finish my observations on the course in pure analysis by
manifesting the desire to see the use ofthe algorithm of imaginaries [i.e.,
complex numbers] reduced to what is strictly necessary. I have been
astonished, for instance, to see the expression of the element of a curve,
given in polar coordinates, derived from an analysis using this algorithm;
it follows much more quickly and with greater ease from a consideration
of infinitesimals. It is quite true that the introduction of the imaginaries
into analytic calculations is often very useful. However, the fact remains
that by using them unnecessarily in a mathematics course at the Ecole
Poly technique one deviates from a very important goal: the goal of
learning to exercise thought and develop powers of judgment (57).

November 17, 1826, the Conseil de Perfectionnement appointed a new

commission in analysis, mechanics, social arithmetic, and geodesy. This new
commission, made up of Laplace, Biot, Binet, Poisson, Pro ny, and Puissant,
was to function within the framework of a general reorganization of the
academic programs, because the pamphlets and brochures that professors
Cauchy and Ampere were required to submit to the Conseil de Perfectionne-
ment were not always published. On this occasion, the minutes ofthe Conseil's
meeting state that:

The Marquis de Laplace called the Conseil's attention to the attempts

made by the Conseil over several years to improve education and
training in mathematics by a simplification of instructional methods. He
also referred to the fact that the commission that had been appointed by
the Minister of the Interior to implement this goal, of which he [Laplace]
was president, had tried in vain for three years to get the professors to
submit written material on their courses that would satisfy the
 84 5. Teaching at the Ecole Poly technique

commission. He added that, as a consequence of the requests of the

Conseil de Perfectionnement at its last meeting, and as a result of the
Conseil's urging, the Minister of the Interior had directed the professors
in question to present printed material on what they intended to do in
their courses... but the other professor concerned, Monsieur Cauchy,
has only submitted some pamphlets that the commission found
unsatisfactory. As a result, it has been impossible, at least up to now, to
get him to comply with the Conseil's instructions and the Minister's
decision (58).
Following this declaration from Laplace, the Gouverneur of the Ecole
declared that 'in case other attempts are made, and are unsuccessful, then he
[the GouverneurJ would consider himself as obliged to propose to the
government that it take harsh measures, which he would find very distasteful'.
The grave warnings from the administration of the Ecole notwithstanding,
the development of the materials on the analysis and mechanics course made
no progress in the following years. Prony's report on the academic year 1826-
1827 insinuated that the students did not make use of Cauchy's teaching in the
examinations for passing from the second to first division:
According to the unfavorable opinion expressed by the analysis
professor on the performance of the students during the 1827 [i.e., 1826-
1827J year, one would have expected some fairly unsatisfactory results
on the examinations. However, these results were better than anyone
had grounds for expecting. Of the 143 students tested, 100 to 110 showed
a more or less sustained mastery. It is to be believed that those students
who neglected their studies during the course worked hard to catch up
the last time during the 5 or 6 weeks between the end of the course and
the examination period. But this kind of improvised instruction has
drawbacks that could become serious, and it is necessary to take definite
steps to prevent the students from being again in the situation of having
to substitute learning by cramming. Learning, indeed, will leave no
traces in memory, it will not interest and profit the students ifit does not
follow a methodically gradual course. This graduation is a condition
that is incompatible with great haste. Moreover, it should be considered
that such cramming has a bad influence on the health of the students, and
one can cite examples of this (59).
In the same report, Prony censured Cauchy's mechanics teaching, which
seemed to him too abstract.
On the basis of a report made by Poisson, in the name of the commission
appointed on November 17, 1826, the Conseil de Perfectionnement decided on
February 15,1828 to take charge ofthe printing of the brochures. The plan was
for stenographers to make exact records of the lectures given in the course on
analysis and mechanics, as well as in other courses. These records were then to
be revised and corrected by the professors and then printed. In addition, the
 5. Teaching at the Ecole Poly technique 85

professors of analysis and mechanics were to hand over to the commission

complete topical outlines of their courses. These outlines would serve as basic
guides for the construction of new syllabi and programs. Once this material
had been approved by the Inspecteur des Etudes and by the examiners, it
would be printed. 'The material for both divisions', reads the instructions,
'should not exceed two volumes of 400 pages in quarto format, one for analysis
and the other for mechanics. The main purpose of these volumes will be to
guide the students in the direction of those applications that are most
necessary in the public services'. The commission set very strict timetables for
the course stenographers to complete their tasks, as well as for the professors
to have the topical outlines for the new course material written and submitted.
These arrangements prompted protests of indignation from the professors
of analysis and mechanics. Cauchy sent a letter (which, unfortunately, cannot
be found in the archives of the Ecole Poly technique) to the Minister of the
Interior. Ampere resigned from the Ecole in May 1828, after declaring that it
was impossible to complete the required two volumes of 400 pages in quarto in
one year. Finally, the Conseil d'Instruction decided against making steno-
graphic records, because their corrections would have required too much
work from the professors (60).
As with all the other decisions made by the various conseils at the Ecole, the
resolutions of February 15 were thus dead letters, completely ineffectual.
Cauchy was content with the June 1828 publication of the second volume of
the Le(:ons sur les applications du Calcul Infinitesimal ala Geometrie, which was
devoted to the applications of the integral calculus. In this work, Cauchy
investigated the classical methods used for the rectification of curves, the
quadrature of surfaces, and the computation of volumes.
Prony was still unsatisfied with Cauchy's teaching. He wrote in the report
on the academic year 1827-1828 (first-division course):
The professor considered several objections that were made last year
concerning the instructional system. Nevertheless, there remains a
vagueness in the exposition and the use of general theories. Sometimes a
student who can discuss generalities quite well comes to a complete stop
or is embarrassed when it comes to applications to particular cases. It is
to be feared that abstract theories, isolated and on their own, are not
remembered. He [Prony] continues to insist on the suppression of
considerations that have to do with the method oflimits and that extend
proofs without making them concrete. This observation is of particular
importance when it has to do with a course for youths who will become
engineers (61).
In May 1829, Cauchy published his Les:ons sur Ie Calcul DifJerentiel, which
replaced the first part of Calcul Infinitesimal of 1823, which, by then, had gone
out of print (62). This work was more logically developed than the work of
1823. In this new study, Cauchy introduced the theory of the order of
infinitesimals, which he used to prove Taylor's formula (with Lagrange and
86 5. Teaching at the Ecole Polytechnique

Cauchy's remainders), now included once again in the treatment of differential

calculus. Moreover, in the eleventh, twelfth, and thirteenth lectures, he
developed a theory of functions of a complex variable. This theory had not
been presented at all in Calcul bifinitesimal of 1823. In developing it, he
extended the definition of the derivative of a real function to functions of a
complex variable, and this extension allowed him to apply Taylor's formula to
complex functions. Unfortunately, however, the crucial point, the difference
between differentiability in the real and complex cases, eluded him altogether.
Thus, he continued to believe that, just as in the case of functions of a real
variable, a continuous function of a complex variable is differentiable.
Obviously, Lefons sur Ie Calcul Differentiel did not meet the terms that had
been laid down by the Conseil, and Prony was obliged to declare to the Conseil
de Perfectionnement that:
The professor of analysis and mechanics has now published the portion
of his course material on differential calculus. It can be seen that he
intended to comply-up to a certain point, at least-with the condi-
tions imposed by the Conseil. However, those conditions have been only
imperfectly fulfilled in this new work; and in general, the methods used in
this publication are not the same as those that have been taught (in his
classes). That being the case, the present students, likelhose in the past,
have no way of reviewing and studying the material they have covered in
class except by way of their in-class notes (63).
By the end ofthe academic year 1829-1830, Cauchy had finally written and
gradually submitted the complete topical outlines for his first- and second-
year courses in analysis (64). But, no work was done on the mechanics course.
Here, matters rested. During the summer of 1830, following the July
Revolution, Cauchy left France and did not return to the Ecole when the
November 1830 term began. Neither the third volume of his Lefons sur les
Applications du Calcul bifinitesimal a La Geometrie nor Lefons sur Ie Calcul
Integral were ever published (65).
 Chapter 6
From the Theory of Waves
to the Theory of Light

Mathematical physics, like teaching, was an extremely fruitful source of

inspiration for Cauchy. The study of various physical phenomena, of course,
provided a natural setting for many difficult mathematical problems. The
challenge facing Cauchy was to develop and articulate new mechanical models
consistent with the experimental facts and to solve the equations of
equilibrium and motion derived from them. In these areas, Cauchy undertook
fundamental research work, which required a great deal of time and effort.
His main contribution to mathematical physics remains today the continuum
theory of elasticity; however, his longest and most ambitious research efforts
centered on the molecular theory of light. In fact, the purely mathematical
theories, such as the theory of elliptic functions and the theory of algebraic
equations, having few physical applications, never particularly interested
Cauchy. So, apart from his arithmetical works, most of his studies dealt with
mechanics or mathematical physics in one way or another, especially those
on linear differential equations, complex function theory, and linear algebra.
In this chapter, we will adopt the current tensorial notation with the
summation convention.! This unimportant anachronism simplifies the
involved mathematical writing that Cauchy used in his papers on mechanics.
Cauchy undertook his first work on mathematical physics for the Grand
Prix de Mathematiques at the Academie des Sciences in 1815. The subject
treated in this competition was the theory of wave propagation on the surface
of a liquid (1). Laplace had already examined the question in the case of a liquid
of constant depth. He concluded from his investigation that waves on the
surface of a liquid are propagated with a constant velocity. Not long after that,
solving the problem for the case where the depth of the liquid is very small and

1 The symbol of derivation relative to time is Go; Gi , Gi' and Gk are the symbols of derivation relative
to the space coordinates. According to the summation convention, whenever a lowercase italic
subscript appears twice in the same monomial, this monomial stands for the sum of the three terms
obtained by successively giving to this index the values 1, 2, and 3.

87
88 6. From the Theory of Waves to the Theory of Light

constant, Lagrange determined that the wave velocity was proportional to

the square root of the depth. By assuming that the wave propagation was
perceptible only at a very small depth, he generalized his formulas to the case
of a liquid of indefinite depth in his Mecanique Analytique (2). On January 3,
1814, the commission of the Grand Prix, which had been appointed on
November 15, 1813, a few months after Lagrange's death, proposed a problem
dealing with the matter for the competition:
A heavy mass of fluid, originally at rest, has been set in motion by the
action of a given cause. Determine the shape of the boundary surface at
the end of a specified time and the velocity of each of the molecules
belonging to this surface.
In principle, the competitors were to work in secret, without discussing their
results with others. However, Cauchy kept Laplace informed from the very
beginning, of the conclusion he reached in his research, namely, that wave
propagation within an incompressible fluid, that is, a liquid, is not uniform,
but is uniformly accelerated. A discussion with Poisson, who was also
investigating this problem (and who had also reached the same conclusions,
albeit by a heuristic approach), convinced Cauchy that his findings were
correct. Accordingly, on July 24, 1815, he presented a note entitled 'Sur Ie
probleme des ondes' to the Academie. In this unpublished note, he gave the
four laws of wave propagation within a liquid that were later stated in the
third part of the award-winning study (3).
Cauchy submitted his paper anonymously on October 2, 1815. He
generalized the problem that the Academie had originally posed and
investigated the motion of an arbitrary molecule, that is, particle of the liquid
in the case of a potential flow caused by a perturbation that he assumed to be
initially weak and confined to a small area of the boundary surface. The initial
conditions are the ordinates 3 3 = F(~l' ~2) of the boundary surface and the
initial impulse Q(O) on it.
In the first part ofthis study, Cauchy examined the state of the liquid at the
initial time. He pointed out that if the initial impulse q(~b 0) is q(O) then the
initial velocity vlO) = Vi(~j' 0) of any particle with the Lagrangian coordinates
(~i' t) can be expressed in the form

vlO) = - -1 a.q(O)
, p'

where p is the density of the liquid. Thus, in modern terms, the initial velocity
q(O)
field derives from the velocity potential - . By using the theory of
p
determinants he had developed in 1812, Cauchy then established the equation
of continuity for a liquid in Lagrangian variables (~i' t):
 6. From the Theory of Waves to the Theory of Light 89

where x;(~j' t) are the coordinates of the particle in a given rectangular

Cartesian coordinate system. In particular, for t = 0,
det(o;~J = 1.
From this Lagrangian expression, he deduced the equation of continuity for a
liquid in Eulerian variables v;(Xj' t) at the initial time
o;vIO) = °
and obtained the Laplace equation
0uq(O) =0
for the initial impulse q(O) within the liquid, since the flow is potential. He
applied these equations to the boundary surface and expressed the solutions
satisfying the boundary conditions by means of Fourier integrals. Thus, he
obtained, in integral form, the values of q(O) and vlO) at all points of the liquid.
In the second part, Cauchy investigated the state of the liquid at an
arbitrary time. Using the so-called fundamental property of the fluids, that is,
the equality of the pressures around any point, he derived Euler's dynamic
equation
1
a; = F; + -ViP,
p
(6.1)

where a; is the acceleration of a particle, F; is a given specific body force, and p

is the pressure on the particle. He expressed the acceleration a; first in
Lagrangian variables and then in Eulerian variables. Then, he proved that
there is potential flow at any time:
1
v;(Xj' t) = - - 0iq(Xj, t), (6.2)
P
and he deduced from the equation of continuity o;v; = 0 combined with
Eq. (6.2) a Laplace equation:
Ojjq = 0.
Then, Cauchy assumed that F; is derived from a potential <p and transformed
Euler's equation Eq. (6.1), into the Cauchy-Lagrange equation:

P- 00q V;Vi A-. _ 0

p +2-'f'-'
If <p is the gravity and Vi is infinitesimal, this equation yields the value of the
pressure P at an arbitrary point in the liquid:
P= 00q - pgX3'

Cauchy applied these results to the boundary surface where P was assumed to
vanish. Thus, knowing the impulsion Q on the surface, he had the ordinates X 3
90 6. From the Theory of Waves to the Theory of Light

of the surface:
1
X3=-8 oQ
pg
and, after differentiating, its vertical velocity:
1 2
V3 =-8 oQ (6.3)
pg
From Eqs. (6.2) and (6.3), it follows that
g8 3 Q = - 8~Q.
This equation substituted in the Laplace equation yields the partial differential
equation
(6.4)
for the impulse at any time on the boundary surface. Cauchy solved Eqs. (6.2),
(6.3), and (6.4) once again by Fourier transforms. The solutions have to satisfy
the initial conditions. In this way, he obtained the values of the unknowns q, Q,
Vi' Xi' p(X;), etc., depending on the initial conditions, in integral form. He
discussed the results in the final part of his study. He showed that this
discussion amounted to investigating nonconvergent integrals of the type
[+00
K = J0 cos j2k;; cos Jl dJl (k > 0)

and to approximating them by means of asymptotic expansions. So, he

obtained the laws of the wave propagation within a liquid and the shape of the
boundary surface at any time.
During the time of his research, Cauchy made additional investigations: he
studied the case in which the depth of the liquid is finite, and he discussed the
equation of continuity and the equations of motion for an elastic fluid, in order
to determine the wave motion on the boundary surface between air and water.
He never did publish these supplements (4).
Cauchy'S paper easily won the prize in the competition. Still, it did not
attract much attention, because it was eclipsed by two studies that Poisson
presented when the competition was over. Whereas Cauchy had to wait until
1827 to see his paper published in the Savants Etrangers, Poisson published his
own studies in the Memoires ofthe Academie as far back as 1818. Poisson used
methods similar to Cauchy's. However, by giving a clearer and more thorough
discussion of the integral formulas, Poisson established the existence of a wave
propagation with a constant velocity in addition to the uniformly accelerated
wave propagation. Cauchy took advantage of the delay in the publication of
his paper of 1815 by adding, in 1821, 2new notes to the 13 initial notes. He later
added 5 other notes to this paper before giving the final manuscript to the
secretary of the Academie on May 27, 1824 (5).
 l':-l ~ ___ ,-"'I...
,,; • .!;-... , • t • ~ t~&...~~ . \00.:--...... _ ; ..... ... ~ ,=rt- :;. , •..£.-.
..i .... 1. !'... , ~.~) ; _
f "' ~ :!:~I' - :!:-;,I .~~.I 'r ~,~ ..~::;-,_ .r;: ... 11
i.,· .. ." :./ .~,! ;,,' to':- I. .. .. , '
':-"""'., r: " .. " , I.~ .. , ( -'I " , - ·t· g l. ' • 'I t ,,. .... : '.1., . ,,"·· ',f t: . ,, ~' " ...,. ,,,,·C 't' flo,}
~~*"J . , 1 ;.4(=- ~"" .. ..,'...
"-,,,C, ..J,:h' S':' .', ';',·Ii .. "I:::~
.'.,', , ':..
.. ,I..:....:. ... )(,'~ " "'1 )' ,~,,:: "' :(' ,1',1. •. j.~:. • "U "1 ' . . . t. ~. ' ..,,'.;, ;;. ,~,jJ' ,. ~ . ,:.:." , ...,:" . . .....". ' ... ..-.:"'..,,,
~".U • • 0;1 .. ......... :- , ::,,: .. 1
~~~~..;~ 'I~'~:' -"?,I,,') ......... ,... .. {~·-!·1 " :~.! .. .... . ,, ! ( .. ,.:o, j!' . -:-,,'. ..,:0•• :',,' "~'I' " ~ •. 4.') t ~)'·,··f l,f1d
", .'''" ",:",1 -If,:O: .. t.Jf·).
",' 'l ::,
';. ·f jt( ':'{ ,or, ,,. ;H '( ·:f
)f :',
:-r. ' ... ·;o . ~ .. ~;-':t ~<i.~I'r ';i' ~: ..' ~ (:~.;: 1/(I' U 4 f1'f.#;~ .·'::'''Jd'';~;
. ...,,: .' ~J.J.1"' J. 4~ ·(· 1 ... , .. ;.,~:~ ,,~";I, ..... , . ~. ~.~ t l t :" .&n""'· I.'!'I':"
~~,\"; ,:::~~'-, ".;I!"·' ...·!-:"t, .," ....
""'''',r ?'
'" ...... I, ="""1 I.i~· "" . ... :-) ( CI'')''( •... I:i t · : ,'-:',:r . ... ~;~ . .. ;:1,·, .....'... ,.. ....... ..,1'>11) . ., .a·. ., . . ,.':)
I.··.·. .. .,"" 'T1
''\.r .'.;of.".... f- \)../;". I~_~ "~t ~:..~~~. .. " J-,,:- (3
F" ." .f; ,:,. ,,,, "=- :"""':\~ .:5~'v'. ··I.l.· · ··t_, :. ~ ': , ,, ..,, .. : • • ~. .. -r'!l ••
,.•'" . . . ,.t.. _..;..L ('-<;..::c. _ ..:. ; - : . .. ~ , . . t,.~ 3
~; .... . : ..-1 ':.'''-i~.. .. . &- __ " , - .~ ..
'J: ;: I,:.::~~; :: ~~~:~~: . ;.
~::'>r::~:~:~~: ~~ ',>~r: ~:~~ . ~.: '.: o'' 'r :: ;.:':~-: ,' . t:.. ., '., i~·" ... <>
~ .. ," ' .. .'
l .. .:..lf~ , _(., .... ,,~f .".f,,: ~. "
';": 'j_·u
r ;l
.;!tt .. l. 'If :. r., ;'.4 ~ '1';'11 ~ , .; rJf- I '>'~ .. ",t _- J:."'I- ...
...".. . ill'''') " Ij"-"'''? ' ,f':'.~;t ll i f;. • h: ,,.·.;,1 !'. r...~.r ' .... "' ... ']"~'::l;J'"
""~""·..1 .• ; ..:,-1" :; f)( "", .:,;'::-1; -,.,.;,: ..' --'~ " "',I:' f_ ~:'f .. ! ,,":- . :~~ ."
1: ·...·£':, j'
~dH"i i1.r~:"~" ,~ ,·· r..; ':i :"i(~ ..· f • • rhl~·" ·~ O'.rr;:i' r"'J f.,){,rr .,''',1:'.' tll~ "f>f 2
SU.,:":' \ ';f.:~l l. hJ)""~ .. "'IlH 1.0'(." "jr"-"" h, .. II~'· ,.j,,!.,: ~~ jJ •• ' :
.::1
~, .. ~" ,"14 ift';" .. } f':~::"): i-;P'iJS :';"·"', w ,{» I;,"': 4: rh ..~ ' ~
~~. (.j~:;:; ~i~::~~:. _~~:::~ ~;::e;· _';i:·~:': ': ii::::~; _;E~~::·..;;:"E: H":'''1'.· ! .
ii ..,;,;.
,. ", ';''' ~
~
'n ...,-( :-_'ld.t. .,h , !r~ ! ",, ; ..j.i 'n::I ... f ,-"lJl~ '''.,(--.: .. 1.~;(:dI ":,,,'(1,. ~':',~;iJ • ..... ·.~.IJ :,ff::"" "',,,:..:») .? ""JJJ I!f'.... )r.1
f." .- .JI
'
">:, •• " .... """.1,.!t/
, - lofi 'f of- ?,II?t - III, j;lt~ .." • • If ll ; - ' ..... i r.,. "...... ,.:. .. "1" "'-' <i
-~, '''!~ '"
~J~', '] J~1~ttr... < f:,;!.....·~ .. ~}"~·.. f I;:~ ~-~_..J .' f, ·)":I:-:'" 9;:(,#; , -l:,~""": ~.:!t)t,.. '~~";:l; ;::..,.H ~r,~"i'"'' .:J'~?;/t': "-,fIJI.lJf ~~,,~';"~I I.~ ...... ~. .~ ·H't·' ,,.~ rcr~I!'1 0-
~-9';~1'" :-1,/.f,~.. _lf", ~:.'., ~~f.,I;:)~-rf l 'p.,i.. .,. :'.a."!-'- ',"" ~ ;' '';' ;( - "J':-~if ...... ';$.( _.' • .)~jI "'~I . . . . ., . . . " _"

J .... t" :f:'il!~! !3i,t.S{": !~t:(l .•r ?;~1,il:' ~~.,!)." , ~~"·."'r !;.·r; .... 1 :;.i'd·1z . . j.!;~?i'.,~ ~,,~.,.! ~!"... ,. ., ! ?4i:H #',..~(. . ~ ~ ..'~IJ,-1 ;f.-t "1
.......,: ... g-
..
, -'" j"'~ + l~!', .,?'f ~!O -.Jot -!;'.",I'.';- ,' f ,t .. (:II'f' ;1'
~':'I_ ,:,. - ... ,~ .... n.~' ( ':' _ ? , ... r, .... •·.. ~"t _ ~. " J,., ... ~ .. ,", _ .. ,.... " .. . ., ..,.. -: _ " ........ ~!"!.'" ~!r" .. :'
"Is. "AI ;,r,:f~t f~·/(~a:J. '(ft~it.i (t;;';IJ." U:·ll~i' ~f.": ".~" ,,'J",:;;, .:u"'·'t" 'I I ...;.,.', r"!IJ .I.. p:~"")" ~:, ; •.Uf ItOll,' '~::' )/: ~:'/;_,,_ , -t-' .1, . ;l
A(".t: ).' .'/~"". I •
1)- ~"O'&;l 'f" :TI .J4 r,' _ ';ft:!"'? ... .:I·. . ,-'I:.~-"'C,1~" . :: .. f.·_, . . ~ - ;,:':i.;, -f- + ')''.ll1 _ •. r,•. <t ..... " .':"_ .... ~I .... of ... ~"t' ___ . u,o)
",::ia. - ".:. . . 2
~'.I,I' 11;.:11:. ::,:,.It. I,.tu~·.! '::, ..1;, ...... 11'.1 ' r" (:"':, ~:"(·~4 1. 1;~,t.u h#,.,:;." 11:,:"":' it~'·'~,-:)~;., :'~-: ,.,J'>,)f. J::i:~I! ' 1''''.'-'' ~")j'.~)" ~Cf;~ 4;., -t" • , I I " ' ) .::1
,- ':(j-:lt~ .... :,f. ...!~r - .. -~., "'jt ~ c.,,= ,>·': .. c;,••~: •• 'i;, ..n~ - .:: , . ,~ .,..:r ,' . ~ - ~4.?::; of" : • • '-? .. _ , :.'>,b ..; ... ..,£ _ ' . _..... : + ". "fI'-~ _ ' ......... t
..... , •• t ...... I t, ~
'J,., ' "1i'I't.i.. ",?C'",'; 1It .. ;r~'tf I.. ·~ .' :, ~ .. 41!'.' t'"'
'.'#'_4,~. 1.~;~:'~lt :'#.1H" -:"':"71.'1J f· ".,I .. 1f;'·'~' .. ' i":.,,,r ....r:o",. '1"'.'>'-' I!. t:•..,,,.-. f)".~,"'6....":( ... ~;I
' ." - •fl. 7')r~"'" ;'1, "Jh'i -'I,"f.,.:'t:, ; .. '~;':">1 ·t _r'\( li:l ... ""I" .I(')~ .,.:~Ij, I'(f t ..:~~ ,/,.; .. ~,,,'(~;,,, t .~:':-/~1.1 .... .1, ~:.... r 1, ~'.-4 ... ..... ~'fl: .,.~, ' ;'" • j

' .,.·t ,~· .. •, •• -'1 _ ".~ ... , _ •. I',!~'

0;;'
,,,,,.,,1 .(.::"t'.1 'l"l p",f~ Cl't·,,"l'!'1 ,'J,ttl,'" 'I ,U :li~" t.."·i- ~..."':'.:t. t H'.1 ,:u r""iU, ~t,H ." ",,!-;J~" .-;ora,>l. J§,"'~<t •
'1.''>, ~j~""''.-I.('J'-~O''''''''Jnf .. -r\.''/('''!''~~ •.t. -.·'d!1 . -t' ..... , ' \ - ~ , ·1.1; .... .... ;,s;,· _ •. ,.,~+ . ...... , ...._..., ;:r
_., ~,II(
') ".,ff
')- lI~"'t'.l "'~" ... ~1 i~LU.~.t. ,~;"fjt... '~' J.1 ' \'.~I . #.' • • ~ .. .... (
~_ : . , : , , , 1- ;.I'!t~- o,~~",,~ .... .J~t· .. ....I . . . ~~ ~ ~,.. ... - .., ...... ~ ...
_,J, ~11,'"
I •• )"J )f,-',d .. l 'i~tt'f.CJ :(~~~:7"''''' !'~i:,:to/ . . -!~;)!~, ":Il-,.IC: s :-• .lt.
-'~ _ ~I;I-.t ' .... l' ; : , : - ","7,. ~ .,~I.l~_'.4"'~"': "'f';",:';' ~ . ~;-. ....., ......
1-:. .. 1·... ')
(5~d~ot ']lrt.! ~J :l~t;."'~ ~~~,: .... ., t~"'" ;.'....... ".' r ).(
t ? __ .. ,S~1,~t,?,L!- .. ~(2n .T",: ..J. ', - ........ )' ,""';;:;,:'-.J e .. '''.-1''
- .,~,:,(
, •."rll ."'u",
. .. t ,,;,,1~" ,.1~fl .. 1 ~;",,,lJ,.. ~I ... )I""· , - . +., ......
.....:-' - .": " ; - 00) t ",..~.,:.
?-1,'1J'7 -+f.')hj. ""Ii"]"'" "IJ"i~· ~ _.,~.;" .. '.0,:'( '.
'"
Computations relative to the wave propagation on the surface of an incompressible fluid. Manuscript by Cauchy, 1815,
Sorbonne Library, ms 2057. Published by permission of the Ecole Poly technique.
92 6. From the Theory of Waves to the Theory of Light

Over the following years, Cauchy did not present any further studies on
mathematical physics or mechanics to the Academie, even though he was a
member of the mechanics section. Nevertheless, he did not completely
abandon these disciplines; he continued to teach them at the Ecole Polytechni-
que and, in 1817, at the College de France. This teaching gave him the
opportunity to master the mechanical concepts and the mathematical tools he
needed to create continuum mechanics. Cauchy stated the fundamental
equations of hydrostatics and hydrodynamics by Euler's method in his course
on mechanics at the Ecole Poly technique. Unlike Euler, however, he based the
theory of perfect fluids on the property that the hydrostatic pressure is normal
to the surface on which it acts, and he argued that the property of the
hydrostatic pressure expected to be equal in all directions can be derived from
it.
In October 1821, Cauchy began teaching a course on mechanics at the
Faculte des Sciences, and this course no doubt provided the inspiration for
further research in mathematical physics: on the one hand, he carried on the
investigation of the linear partial differential equations with constant coeffi-
cients, using Fourier transforms, and on the other hand, he began to develop
his continuum theory of elasticity. Thus, he submitted a major paper to the
Academie on continuum mechanics on September 30, 1822 (6). In this paper,
entitled 'Recherches sur l'equilibre et Ie mouvement interieur des corps solides
ou fluides, elastiques ou non elastiques', he set forth the basis of his continuum
theory of elasticity, which he thought to be applicable to nonmaterial media,
such as luminiferous ether or caloric fluid, as well as to solids and fluids.
By this time, the problem of the elasticity had begun to be thoroughly
investigated from three different points of view: an empirical point of view in
the research on the strength of the materials, a physical one in Fresnel's
research on the wave propagation oflight, and a mathematical one in a series
of studies on the small deformations of curves and surfaces (7). As an engineer,
Cauchy knew the empirical approach of the study ofthe strength of materials.
Inspired by the works of Coulomb, he had undertaken some investigations on
the strength ofthe bridges and the theory of the arches at the Ecole des Ponts et
Chaussees in 1809 and 1810 (8). He had also worked on the site of the Ourcq
Canal under Girard, who had published Theorie Analytique de la Resistance
des Materiaux, a work of considerable renown, in 1798. He remained interested
in these practical problems, and on August 9, 1819, he even presented a
detailed report to the Academie on a paper of Duleau's on the strength of
wrought iron (9).
On the other hand, the theory of elastic surfaces had been of interest to the
Parisian scientific community since 1810, when, following a series of
spectacular experiments by Chladni on the nodes of a vibrating plate, the
Academie des Sciences announced a competition dealing with this matter for
the Grand Prix de Mathi:matiques: the Academie demanded that the
competitors 'give a mathematical theory of the vibrations of elastic surfaces'
and 'compare this theory with experimental results'. In spite of the difficulty of
 6. From the Theory of Waves to the Theory of Light 93

the problem, Sophie Germain successfully determined the solution and was
awarded the prize in the 1816 competition, after an unsuccessful showing in
the 1812 competition and an honorable mention in that of 1814. In the
meanwhile, Poisson presented the paper 'Sur les surfaces eIastiques' to the
Academie on August 1, 1814 (10). In this work, he used a molecular hypothesis
to obtain the equation of a flexible elastic surface equally stretched in all
directions, which Sophie Germain had already established with help from
Lagrange in 1813. Sophie Germain published her results in 1821 in a work that
she promptly sent to Cauchy (11).
In fact, a few months before, on August 14, 1820, Navier had submitted to
the Academie a paper on the same theory of elastic plates (12). He had also
distributed lithographic copies of his paper to a few members in the scientific
community, Cauchy especially. Motivated by an engineer's concern, namely,
the application ofrational mechanics to the study ofthe strength of materials,
Navier calculated the small deformations of a weighted plate equally stretched
in all directions. His paper began with a derivation of the equation that Sophie
Germain and Poisson had obtained for an elastic surface. However, his
derivation took the thickness of the plate into account. In his analysis of the
elastic forces, Navier considered separately the forces produced by elongations
and contractions of the plate and those produced by the bending. In order to
obtain mathematical expressions for the latter forces, Navier examined an
arbitrary small element of the plate that, before the bending deformation, has
the shape of a right circular cylinder whose elevation is precisely equal to the
thickness of the plate and that, after the bending, is assumed to have the shape
of a truncated cone, one base of the cylinder dilating and the other contracting.
Navier assumed, in general incorrectly, that during the bending the elastic
forces act normally to the faces of the cone (13).
A reading of Navier's study inspired Cauchy's paper of September 29,1822.
As has been noted, Navier had separated the forces produced by the
contractions and dilations of the plate and the bending forces. In contrast,
Cauchy combined the elastic forces transmitted onto an arbitrary isolated
region inside the solid. In order to describe these internal forces, which are
known today as stresses, Cauchy took the very natural route of generalizing
the well-known definition of hydrostatic pressure (14). Like hydrostatic
pressure, the stresses at a point inside a solid body are contact forces that act
on each surface element passing through the point. However, there is a great
difference between fluids and solid bodies. In a fluid, the pressure is always
normal to the surface element on which it acts. We have seen that Cauchy, at
the Ecole Poly technique, based his lectures on hydrostatics on this principle
(15). In a solid, however, the stresses generally are not normal to the surface on
which they act, contrary to Navier's assumption.
The application of the c1assicallaws of equilibrium to conveniently chosen
elements of volume allowed Cauchy to express the stresses at a given point in
terms of nine stress-components T u, of which six are pairwise equal (Tij = Tj ;),
with respect to a rectangular Cartesian coordinate system. In modern terms,
94 6. From the Theory of Waves to the Theory of Light

the state of stress at a point of a continuum was mathematically defined by a

symmetric tensor, namely, the stress tensor T. Cauchy's method of proof"is
today regarded as a classic. In the case of a fluid at rest, the nine stress
components reduce to three nonnull components that are all equal to the
hydrostatic pressure p. Cauchy developed these first results by 1821.
I was at that point [Cauchy wrote in 1822J when M. Fresnel came to talk
with me about some investigations on light that, as yet, he had only
presented in part to the Institut. I learned that he had obtained a
theorem analogous to my own, his result being based on certain laws
according to which the elasticity emanating from a single given point
varies in different directions (16).
In two supplements to his study 'Sur la double refraction', one of which was
presented in January 1822 and the other the following March, Fresnel had, in
fact, determined the molecular forces acting upon a single, slightly displaced
molecule of ether from other molecules around it; the results were thus similar,
in this very special case, to the ones Cauchy had obtained in the general case of
a continuum (17). In developing this analysis, Fresnel was attempting to
construct a molecular ether across which transverse vibrations could be
propagated. In this way, he would be able to substantiate his hypothesis that
light waves were exclusively transverse.
Still, as Cauchy wrote in 1823, he was as yet a long way from being able to
set forth the general equations of equilibrium and motion inside a solid.
Accordingly, Fresnel's achievement was, in a sense, a source of inspiration for
Cauchy and an encouragement to continue his own investigations. Finally,
during the second half of 1822, he obtained the desired results. Let us see how
he proceeded. Cauchy considered only the infinitesimal displacements U i of
any point Xi in the neighborhood of a referential point xIO). Since the Ui are
infinitesimal, the velocity Vi of Xi is reduced to 00u i and its acceleration to 00 Vi.
By examining the stresses on two opposite faces of an arbitrary small
parallelipiped ofthe continuum, subjected to a specific body force F i , Cauchy
first determined the equation of equilibrium. Under the same hypothesis,
assuming in addition a specific inertial force - ai' he easily deduced the
equation of motion in a continuum, which generalized Euler's dynamic
equation, Eq. (6.1). In tensorial notation, this equation, known today as
Cauchy's dynamic equation, may be written as

(6.5)

However, Cauchy still had to express the six independent stress components at
a point as functions of the displacements. In his study of 1822, he assumed that
only the linear strains, i.e., linear condensations and dilations about a point,
needed to be taken into account. The key idea, no doubt inspired by Fresnel's
work, thus consisted of investigating the geometrical properties of the stresses
 6. From the Theory of Waves to the Theory of Light 95

DE J-JA. PRESSION OU TENSION

DANS UN CORPS SOLIDE.
_ _ rz

Las geomMres qui onl \'ecberche les equations d'equiIibre OJl de mouvemen t dcs
lames au des surfaces elastiquea au non clastiqucs, ant diltingue deUJ< esp~ces de forces
produitcs les unca par la dilalolion au la conlraction, les aulres par.la fl~):ion do ces
memes surfllCCS. D. plus, ils ant generalcment SUPPOle, dans feurs calculs, que Ics
forces de la premiere esp~ce , nomm6ea lensions, reslent pcrpcndiculaircs aux JilSoe.
conlre lesquellcs elles . 'exercont. Ilm'n sembl6 que ce. deux cspeces de forces pouv.icnt
clre ro!duiles II uno seule, qui doit conslamment s'appoler lellsion ou preuion, qui
ogil sur cha'!uc clement d'une section foite b voljlnte, non - sculemont dans uno sur-
face flexible, rnai. encore dans un solido cl.sti'!ue ou non ~1 •• lique, ct qui csl do I.
m~me nature que 10 pression hydl'ostatiquo oxerceo par un fluide en ropos can Ire III
surf~ce ext6rieure d'un corps. Seulement la nouvcllil' pression ne demeure pas toujour.
perpendicul.ire nux faces qui lui soot spumises, oi In memc dans lous les sens ell. un
poinl donne. Eo developpalll cello jd~e, js sui. parvenu a rcconnallro que In prossion
ou lensioIl. exercCe conlre un plan quelconque en un point donne d'UD corps solide sa
<leduit lres-Biscmenl, lanl en grandeur qu 'on direclion, des pression. au lension!>
exereccs cootre lrois plans reelanllulaires mones par Ie ",Gme point. Celie proposilion ,
que rai d6jb indiqueo dans Ie Bl1llclin dela Socielo philomatique de janvier 1825,
peul 6tre elablie 11 I'nide des coosidcl'alions suivnnle •.
Si, dans un corps solide elaslique ou non olastique, on viani 11 rendre rigide ct inva-
riablo un pelit 61eroont de volume lerminc por des faces quelconquos, ce pelil el~rocllt
oprouvcro sur ses di[crenlos faces, et en eh"que poinl de chaoune d'elles UlJe pression
ou lension determinee. Celie prossion ou lension s~ra semblablo 11 I. pression qll'un
lluide CXerCc coo Ire un clement de I'eoveloppe d'un corps solide, ovec celle seu le ' dif-
ference que la pression exercec par un flu ide en rcpo COoll'8 10 surface d'un corps
solido, esl dirigeo perpcndiculairemool h celIe surface do debot's en dodan., ol ind
peodaole en cbaque pOiDI de I'inclioaison de la surface par rapport nux pIa os coordon-
ntIs, landis quo 10 pr65sion au lonsion oxercee ell uo p,)int donne d'un corps eolide
conlro un Ires-pelil elemenl de surface pos.ant par co poinl, pOlll dIre diri~e pcrpcu-
diculoircment Oil obliquemonl h cello surface, ll111t8t de dehors on dedans, s'il y a con
deosa lioo, 100101 de dedoos (In dehors, s'il yo dilatalion, eL peul depcndre de I'jncli~

Definition of the stress in a continuum_ 'De la pression ou tension dans un corps solide',
Exercices de Mathematiques, 2, 1827, p. 42. Published by
permission of the Ecole Poly technique_

and strains. As a matter of fact, Cauchy had already developed the methods of
linear algebra he needed to calculate the moment of inertia of a solid and to
solve the linear differential equations. Cauchy succeeded therefore in reducing
all the stresses, pressures, or tensions at a point to three principal stresses
acting along the axes of a suitable chosen quadric, namely, Cauchy's stress
96 6. From the Theory of Waves to the Theory of Light

quadric, generally an ellipsoid. The principal stresses, showed Cauchy, are

normal stresses.
The analysis of the state of strains at the same point yielded a similar
geometrical representation: the linear contractions or dilatations e(JI) along all
the directions II about the point (i.e., the rates of extension) can be expressed by
means of three principal strains along the axes of a second quadric, namely,
Cauchy's strain quadric. Formally, Cauchy obtained the symmetric strain
tensor U, whose components are Vij = O(iUj) [o(iu j ) = t(OiUj + OjUi)]; he showed
that e(JI) = VijJ.liJ.lj and that the tensorial invariant Vij represents the cubical
dilatation. However, Cauchy did not investigate the kinematical properties of
strain and never used the fundamental concept of shear.
Cauchy was now able to obtain, in the case of very small deformations, the
expressions of the stresses as functions of the strains by using a double
hypothesis, which he articulated in the following terms:
That given, it is clear that in an elastic solid, if the tensions or pressures
depend only on the condensations or dilations, then the principal
tensions or pressures are directed in the direction of the principal
condensations or dilations. Moreover, it is natural to assume, when the
displacements of the molecules are very small, that the principal tensions
or pressures are respectively proportional to the principal condensations
or dilations (18).
Cauchy generalized Hooke's law. He implicitly assumed the isotropy of the
continuum, obtaining the equation
(6.6)
Cauchy considered not only homogeneous (k constant) but also hetero-
geneous (k function of xIO») isotropic media. In order to formulate the
equations of motion in a continuum, a substitution had to be made in his
dynamic equation, Eq. (6.5), replacing the stress components by their
respective expressions as functions of the strain components given by the
linear relation, Eq. (6.6). If the continuum is homogeneous, this substitution
yields the equation

(6.7)

from which

(6.8)

Cauchy applied his theory to the study of the propagation of sound in an

isotropic elastic body and obtained an equation identical to that of the
propagation of sound in air. He also examined the case of a nonelastic body in
which the stresses depend only on the instantaneous strains. He thus
 6. From the Theory of Waves to the Theory of Light 97

established a significant analogy between the propagation of heat and the

wave propagation in an entirely nonelastic body.
In this work, which marks a turning point in the history of continuum
mechanics, Cauchy gave a stunning exposition, a scientific study resplendent
with intellectual clarity. 'Never had Cauchy given the world a work as mature
from the outset as this', remarked Hans Freudenthal (19). It was not surprismg.
Cauchy's study of hydrodynamics, for which he had derived the equations
in 1815, was a reliable guide to his research in continuum mechanics.
Following Euler, Cauchy systematically developed the analogy between fluids
and solids in his investigation of the continua, using the classical methods
he taught at the Ecole Polytechnique (20). This remark in no way detracts from
the considerable merit of Cauchy's achievement.
Strange to say, six years were to pass before Cauchy published the results
contained in the paper of 1822. In fact, Cauchy informed the Academie of his
investigation on September 30,1822, without giving a reading or lecture on it;
he did not even give the manuscript to the Academie. Rather, he published an
abstract in the January issue of the Bulletin of the Societe Philomatique and
gave a reading of this resume at the Societe's meeting on February 22 (21).
The work did not appear until 1827 and 1828 in a series of articles in
Exercices de M athematiques (22). These articles contained a number of
improvements. However, Cauchy never published his paper, 'Equations du
mouvement des fluides eIastiques ou incompressibles', which was presented to
the Academie on January 27, 1823 (23). Why did Cauchy wait almost four
years to publish the details of his investigations of 1822? We think that two
possible reasons can explain this surprising decision.
The first of these reasons has to do with the conditions surrounding both
the presentation of the paper of September 1822 and the publication of the
resume in the January 1823 issue of the Bulletin of the Societe Philomatique.
When he presented his paper to the Academie, Cauchy acted in a way that was
certainly clumsy, if not downright tasteless. He acknowledged that the work
he was now presenting was inspired by Navier's paper of August 14, 1820,
which actually had not yet been evaluated by the commission responsible
for examining it. On the other hand, Cauchy's study focused on the same
topic as Navier's famous paper, 'Sur les lois de l'equilibre et du mouvement
des corps solides elastiques' of May 14, 1821, which was also awaiting
evaluation by an academic commission. On October 6, 1822, a week after
Cauchy's. paper was presented to the Academie, Navier wrote a letter to the
president of the Academie demanding that both of his papers be evaluated
promptly (24). In this letter, he noted (but made no official protest or charges
on the matter) that Cauchy had been doing research that was very similar
to his own investigations.
Up until this time, the affair had been confined to the Academie. But, by
publishing the abstract that he had read at the Societe Philomatique on
February 22, Cauchy brought matters to a head. On March 1, 1823, Navier
called the Societe's attention to the existence of his two papers (25). Thereafter,
98 6. From the Theory of Waves to the Theory of Light

the scientific community began to take sides with Navier. The Societe
Philomatique resolved that Navier's observations would be included in the
minutes of its meetings and would also appear in the coming issue of its
Bulletin. Moreover, Augustin Fresnel published a quite harsh note about
Cauchy and the quarrel over priority in the same Bulletin (26). First, Fresnel
called attention to the fact that Navier's paper of August 14, 1820, had not
been published (Fresnel himself underlining the word) and remarked (incorre-
ctly) that Cauchy was the reporter of the evaluating commission (27). After
questioning Cauchy's interpretation of Navier's paper of August 14, 1820,
Fresnel went on to call attention to the existence of Navier's second paper of
May 14, 1821. He concluded his note with these words:
The work that M. Cauchy has just published seems to have the greatest
similarity to the paper that was discussed here [that is, Navier's second
paper], and it is important that the date of this paper be recalled and
certified.
Fresnel was personally concerned, as it were, in this matter and clearly
suspected that Cauchy might have plagiarized Navier. But such was not the
case with Fourier. In the Analyse des Travaux de l'Academie for 1822, after
having mentioned the study Cauchy announced on September 30, 1822,
Fourier nevertheless pointed out that Navier and Fresnel had investigated
the same problem (28). Cauchy seems to have been aware of these remarks
and rumors, mutterings that were so often repeated that they could not have
been made in innocence. He thus gave up the idea of publishing his study
right away and of presenting a second paper on the theory of thin plates
that he had been working on (29). Finally, he kept his research works on
continuum mechanics to himself until Navier published his own paper in
1827 (30). The postponement was unfortunate. Cauchy nevertheless took
advantage of the long delay between the presentation of his study before the
Academie in 1822 and its publication in Exercices de Mathematiques of 1827
and 1828 to make an important change in his theory of elasticity for isotropic
bodies.
As has been noted, in 1822, he had assumed a linear relation between the
principal stresses and the principal strains that was determined by a single
elasticity coefficient k (see Eq.6.6). This continuum theory with a single
coefficient for isotropic media was incompatible with the theory that Navier
had developed from a molecular model in 1822. In the study 'Sur les equations
qui expriment les conditions d'equilibre ou les lois du mouvement interieur
d'un corps solide elastique ou ineIastique', published in 1828, Cauchy gave a
new hypothesis. Regarding each principal stress as being composed of two
distinct parts, he assumed that one was proportional to the principal strain
and the other to the cubical dilatation Uii. Thus, in tensorial notation, the
relations between stress components and strain components yield the
equation
(6.9)
 6. From the Theory of Waves to the Theory of Light 99

instead of Eq. (6.6) and, if the continuum is homogeneous,

(6.10)

from which

(6.11)

In this way, Cauchy obtained the continuum theory with two elasticity
coefficients K and k for isotropic media. Navier's theory, which was derived
from the molecular hypothesis, was thus the particular case k = 2K of the more
general theoretical framework that Cauchy had now formulated.
Another reason probably prompted Cauchy to delay the publication of his
1822 study. He had based his theory of elasticity on the classical hydrostatic
model and, accordingly, had regarded solids and fluids as continua. But,
advances in molecular physics had made this concept out of date in the 1820's.
The French physicomathematicians now tried to explain all sorts of pheno-
mena, such as heat, sound, light, electricity, and magnetism, as well as the
elastic properties of bodies, in terms of attractions and repulsions between the
material molecules constituting the bodies and, in a wider sense, between the
immaterial molecules constituting fluids like caloric, light and electricity. In
France, the majority of the great figures who worked in mathematical physics
were won over to this theoretical position: Laplace, of course, and his disciple
Poisson, but also, to a certain extent, Ampere, Fresnel, and Navier. The
molecular model was thus, in various forms, so much a part of the thinking of
that era that many physicomathematicians, such as Fresnel in optics and
Ampere in electrodynamics, interpreted physical theories they had conceived
and developed by means of other schematic conceptualizations of the real
world in terms of this model. Navier adopted the molecular model in his paper
of May 14, 1821, regarding an elastic body as 'a collection of molecules that are
located at extremely small distances from each other'. Moreover, in a resume
published in the Bulletin ofthe Societe Philomatique, he (incorrectly) gave the
view that in 1820 he had deduced the expression for the elastic moment of a
plate from a molecular hypothesis. In reality, however, no such notion is found
anywhere in the original study.
Cauchy, of course, could not be indifferent to the success that the molecular
hypothesis was enjoying among physicists. The papers of Poisson, Navier,
and Fresnel encouraged him to work on the basis of this hypothesis (31).
But, as we saw earlier, Cauchy had waited until Navier published his paper
of May 14, 1821 before he shared his research with the Academie. On October
1, 1827, Poisson, outstripping Cauchy, asserted in a note that he read at the
Academie that he was 'presently engaged in a very far-reaching study of the
laws governing the equilibrium and motion of elastic bodies' (32). In order
to preserve his priority rights, Cauchy, on that very same day, announced
100 6. From the Theory of Waves to the Theory of Light

(in a hurried fashion) a study in progress 'Sur l'equilibre et Ie mouvement

interieur d'un corps solide considere comme un systeme de molecules
distinctes les unes des autres' and registered a rough copy of it; the work,
hastily written and edited, was deposited in a sealed envelope with the
Academie (33). Accordingly, on April 21, 1828, when Poisson read the paper
he had announced on October 1, 1827, Cauchy made an objection, probably
in order to defend his priority rights and to recall the existence of his sealed
envelope (34).
This incident underscores the unhealthy atmosphere in which the theory of
elasticity developed after 1814. Irrelevant bickerings and quarrels and
continual protests, claims, and objections on matters of priority rumed the
peace of the little world of the scientists who labored there. No doubt they were
all responsible for the situation, and Cauchy was particularly blameworthy.
But it is likely that Poisson should bear a larger share of the responsibility
than the others. After all, did he not often use his position to take more than his
due share of the glory? There were frequent enough complaints as regards his
behavior in this respect; and Cauchy, who had been on bad terms with him
since 1825, simply did not trust him. From that day on, until 1830, these two
scholars, brilliant scientists with roughly concurrent research interests,
worked in an atmosphere that was poisoned with suspicion and hostile
polemics.
Shortly after the presentation of his paper to the Academie, Poisson
published an abstract of the work in the April 1828 issue of the Annales de
Chimie et de Physique. This publication was followed by endless polemics with
Navier in the succeeding issues of the same periodical. Cauchy kept away from
all the clamor. However, he was probably upset at seeing the theory evolve
without his participation, particularly as Poisson was preparing a printed
edition of his paper (35). So, during the summer of 1828, he actively worked on
three new studies that were to be published in Exercices de M athematiques.
The first of these studies was a major investigation, 'Sur les equations qui
expriment les conditions d'equilibre ou les lois du mouvement interieur d'un
corps solide, elastique ou non eIastique'. As was noted earlier, in this study
Cauchy presented a corrected version of the continuum theory of elasticity
that he had first articulated in 1822. There is no going back on it. The other two
works, 'Sur l'equilibre et Ie mouvement d'un systeme de points materiels
sollicites par des forces d'attraction ou de repulsion mutuelle' and 'De la
pression ou tension dans un systeme de points materiels,' were both based
on a molecular hypothesis (36). No doubt fearing that he might be overtaken
in a race to publish first, he once again deposited sealed envelopes containing
the rough drafts of both of these works with the Academie on August 18,
1828 (37).
The study 'Sur l'equilibre ... ' was an improved version of the rough copy he
had deposited in a sealed envelope with the Academie on October 1, 1827. This
work rested on the analysis of the motion of a given molecule M under the
action FM of all the neighboring molecules M p. Cauchy assumed that between
 6. From the Theory of Waves to the Theory of Light 101

two molecules M and M p there exists a central force of attraction or repulsion

which is proportional to their masses m and mp and is a function of their
respective distance rp' Then, the action F M is of the form

Let us denote by ui the displacement of the molecule M, by Ui + ApUi the

displacement of the molecule M p, by Jl p the unit vector of direction MMp
and by l;lp p) the rate of extension in the direction M M p' Thus Cauchy obtained
the equation

(6.12)

for the acceleration of the molecule M. By assuming that Ui is very small in

comparison to r p' he showed that

and

(6.13)

The hypothesis that the forces existing between the molecules decrease very
rapidly with the distance enabled him to neglect infinitesimals of the second
order in the infinite series expansion of the finite differences ApUi' Afterward,
Cauchy made some assumptions relative to the distribution of the molecules in
the system. He first considered a crystal with three rectangular axes subject to
an external body of force F i• From Eq. (6.13), he deduced the equations of
motion:

O~Ul = F 1 + (L + G)oiu 1 + (R + H)O~Ul

+ (Q + I)O~Ul + 2Ro 12U2 + 2Q031U3,
O~U2 = F 2 + (R + G)oiU 2 + (M + H)O~U2
+ (P + I)O~U2 + 2P0 23U3 + 2Ro 12U1, (6.14)
O~U3 = F 3 + (Q + G) oiU 3 + (P + H) 8~U3
+ (N + I)O~U3 + 2Q013Ul + 2P 032U2,

which depend on nine constants. If G = H = I, L = M = N, and P = Q= R

(three constants), the crystal has cubic symmetry, and if G = H, L = M = 3R
and P = Q (five constants), the crystal is uniaxial with axis X3'
102 6. From the Theory of Waves to the Theory of Light

This theory was thus more general than the one he had obtained in his
investigation of the problem in 1822, in which the bodies were regarded as
isotropic continua. Cauchy encountered difficulties in deriving the case of
isotropic elastic bodies from the molecular model. After a first attempt in the
rough draft of October 1, 1827, an attempt that was based on a debatable
hypothesis, Cauchy developed a molecular theory with two constants for
isotropic elastic bodies. For this theory, he used his molecular theory of the
crystals with cubic symmetry as a starting point, considering the invariance
of the elasticity in an arbitrary rotation about each molecule. Then, the
equation of motion in an isotropic medium is
O~Ui = F j + (R + G)OjjUj + 2R op jj , (6.15)
where Vij = o(iuJ).
Wrongly, Cauchy still thought that this molecular theory with two
constants was equivalent to the one he deduced from the continuum
hypothesis [compare Eq. (6.15) with Eq. (6.10)] and that it was more general
than the molecular one-constant theory Navier and Poisson had developed
(38). Cauchy took up this point again in the second paper, 'De la pression ou
tension dans un systeme de points materiels'. In this study, he introduced the
concept of stress (pressure or tension) into the molecular theory of elasticity. In
case the original state ofthe system is stress free, the constants G, H, I vanish.
Thus, for an isotropic body, Cauchy found the stress-strain linear relation
Tij = R(2 Vij + V kkbj)P (6.16)
[compare with Eq. (6.9)] and the equation of motion
O~Uj = F j + R(ojju j + 2ojVj) (6.17)
instead of Eq. (6.15). This equation with one constant is identical to Navier's.
During the following months, Cauchy was very occupied with the problem
of applications of his molecular theory. In his study of April 21, 1828, Poisson
had deduced the equations of equilibrium and motion of elastic strings, rods,
membranes, and plates. Cauchy simply refused to be outdone. Accordingly, he
presented to the Academie a number of studies on the problems of the motion
of plates, laminas, and rods. His first research, like that of Poisson, dealt with
isotropic bodies (39). Later on, he undertook the investigation of anisotropic
plates and rods (40). At this juncture, he noted, for the first time, that the
molecular theory of elasticity depends, in general, on 15 coefficients (41).
Cauchy also applied his molecular theory to the problem of the equilibrium
and motion of perfect fluids. As we saw earlier, this problem had interested him
since 1815 and was in a sense at the origin of his continuum theory of elasticity.
He advanced the hypothesis that in a fluid the action that the neighbouring
molecules exert on a given molecule is imperceptible; this assumption enabled
him to replace the finite sums by integrals in the equations of equilibrium and
motion. He showed that one could easily deduce the characteristic property of
fluids, that is, the equality of the pressure in all directions. The proof of the
characteristic property of fluids within the framework of molecular theory
 6. From the Theory of Waves to the Theory of Light 103

gave occasion for a sharp dispute between Cauchy and Poisson in April 1828
(42).
In the meanwhile, the theory of light became the newest area of application
of the theory of elasticity (43). Between 1816 and 1822, Fresnel had developed a
wave theory of light that allowed him to explain a number of optical
phenomena: interference, diffraction, polarization, reflection, and refraction in
refringent and birefringent bodies, with one or two optical axes. Fresnel's
theory was based on careful experiments. Analytical formulas, such as, for
example, the expression of the Fresnel surface that allows the calculation of the
direction of refracted rays in a doubly refracting crystal with two optical
axes, had been obtained by inductive methods. However, in 1821, Fresnel
sought to construct a molecular model whose properties would account for
the nature of light. According to this model, light waves would correspond to
molecular vibrations in the ether. For light waves, Fresnel showed, the ether
molecules-unlike the molecules of air for sound waves-vibrated transver-
sely. This hypothesis presented some difficulties and was rejected by Poisson
(44).
As we have seen, Fresnel's work inspired Cauchy, who, at the time, was
working on his continuum theory of elasticity. However, Cauchy had not
attempted to apply his continuum theory to the study oflight waves, and it was
only in 1828 that he first took up this problem. Could it have been that he
wanted to leave to Fresnel, who died prematurely in 1827, the exclusive honor
of making fundamental investigations in this area? It is more probable that
Cauchy delayed embarking on any research in this area until he had overcome
certain mathematical obstacles.
On the one hand, in order to examine the problem of double refraction, he
needed to generalize his theory of elasticity to some anisotropic crystals.
Accordingly, he developed his molecular theory, which, in turn, would serve as
the basis for his theory of light. On another level, he needed to solve the
differential equations representing the motion of the ether molecules. At any
rate, in 1827, Cauchy published a study entitled 'Sur l'application des residus
aux questions de physique mathematique' in which he explored a general
method of solving linear partial differential equations with constant
coefficients.
In a short study published at the beginning of 1829, Cauchy presented the
first conclusions he had reached on the basis of his research on the theory of
light (45). He pointed out that a shock initially produced at an arbitrary point
of an isotropic system of molecules is propagated in the form of two spherical
waves, one of which vanishes with the initial cubical dilation, while the other
corresponds to the molecular vibrations parallel to the plane containing the
initial vibrations. In the case of a uniaxial crystal, he had recourse to Huygen's
theorem on double refraction.

We think [Cauchy wrote] that it should be concluded that the equations

of motion of light are contained in those that express the motion of a
104 6. From the Theory of Waves to the Theory of Light

system of molecules in which the molecules deviate but a little from an

equilibrium position.
Here, Cauchy referred to Eqs. (6.14) (with F; = 0), which are the basic
equations of his theory of light (without dispersion), and to those that can be
derived from them, such as Eq. (6.15). Cauchy took more than a year to realize
the program he had enunciated in his paper of 1829. This delay probably stems
from theoretical difficulties for integrating Eqs. (6.14). During the following
months Cauchy came to grips with these difficulties. In several papers,
especially in a paper of April 12, he showed by using a change of variables that
if a differential equation is homogeneous, the sextuple integral representing his
general solution reduced to a quadruple integral (46). He further proved that in
the case of an initial shock at a given point the motion of the molecules is
propagated like a wave, the vibratory phenomena being perceptible only in the
neighborhood of a wave surface that can be determined at any time. The
theorem can be applied to a system of ether molecules. Once this latter point
was established. Cauchy, in 1830, was able to use two distinct mathematical
methods to derive his theory of light from the analysis of the propagation of a
local disturbance in the system of ether molecules.
The first method, which Cauchy published in Exercices de M athematiques,
was based on Huygen's principle of superposition of little motions (47). It
consisted of first finding only the simple solutions ofEq. (6.14) corresponding
to a given plane wave. Let us denote by n the unit vector that is oriented in the
direction of propagation of this wave front. Then, for a molecule whose
distance from the plane wave at the initial time is h, Eq. (6.14) can be written

(6.16)

where aij(o) is symmetrical.

By using an orthogonal transformation, Cauchy reduced the problem to
solving the equation
06Ui = o~ s3)(0) uj '
where the s3) (0) are the three real roots of the characteristic equation
det (a;io) - s) = O. Cauchy assumed that these roots are positive. Thus, he
obtained three plane waves, whose equations are given by
h - s(i)(o)t = O.
For each s(i)(o), the displacements of the molecules are parallel to the principal
direction. In other words, the three plane waves are polarized. In order to
construct the wave surface that is propagated from a local disturbance,
Cauchy considered wave fronts that are slightly inclined to each other (the
direction 0 varies continuously). The wave surface is the envelope of all these
planes. In general, it has three sheets, each of them being the envelope of the
system of wave fronts that corresponds to one of the three principal values
s(i)(o).
 6. From the Theory of Waves to the Theory of Light 105

The second method developed by Cauchy led to the same results. It

consisted of directly investigating the general solution (given in the form of a
quadruple integral) of Eq. (6.14). Cauchy exhibited this development in
lectures at the College de France in May and June of 1830. He was preparing
the publication of these lectures when the July Revolution of 1830 broke out
(48).
In the case of an isotropic system of molecules (Eq. (6.15) with F j = 0),
Cauchy showed that the wave surface separates into two spherical waves,
with the velocities JR + G and J3R
+ G. The first spherical wave is the
wave of light: The displacements of molecules are transverse, as Fresnel has
assumed. The second one is longitudinal. It vanishes if the initial dis-
placements of the molecules are transverse. In the case of a uniaxial system
of molecules whose initial state is stress-free (i.e., if G, H and I are null), the
three sheets of the wave surface are one sphere and two ellipsoids. The sphere
and the first ellipsoid yield Huygen's construction for the ordinary and
extraordinary rays. The second ellipsoid is assumed to be invisible. Cauchy
also considered the case of a system of molecules with three rectangular axes
(Eq. (6.14) with F j = 0), in order to deduce the optical surface in a biaxial
chrystal. As a result, by assuming some specific relations between the
coefficients L, M, N, P, Q, R, he obtained a wave surface which is only an
approximation of Fresnel's. Moreover, Cauchy'S theory required the polariz-
ation plane to be the plane which contains the direction of wave propagation
and that of the displacements of molecules. This last definition differed from
Fresnel's. But the main difficulty of Cauchy's theory of birefringence is that one
of the two optical waves is not exactly transverse.
In 1830, Cauchy also developed a theory of reflection and refraction oflight
on the boundary between two isotropic bodies. He obtained Brewster's law on
the complete reflection and Fresnel's laws for reflection and refraction of
polarized waves, by assuming that the density of ether is constant in all bodies.
Finally, Cauchy undertook, on Coriolis' advice, to examine the problem of
light dispersion, which Fresnel had not explained. His theory consisted of
using a Fourier series representation for the displacement of the molecules.
Then, the displacement U j of a given molecule at the point whose ray vector is r
becomes

Uj = L (ak(r, t) cos k'r + bk(r, t) sin k ·r).

k

Cauchy chose an arbitrary simple displacement ak(r, t) cos k'r + bk(r, t) sin k·r.
He calculated the finite difference L\u j in this simple displacement and
substituted its expression into Eq. (6.12). By assuming that the molecules are
symmetrically distributed around each other, he was able to obtain a
symmetrical ordinary differential system
05Uj = bij(k)u j , (6.17)
which could be solved by classical methods of diagonalization.
106 6. From the Theory of Waves to the Theory of Light

Cauchy deduced from Eq. (6.17) the same results as he did from Eq. (6.16).
But in the theory of light dispersion, the wave velocity V depends on the
magnitude k of the wave vector k, and thus on the wavelength), = 2~. Cauchy
assumed that color depends on the wavelength and, in consequence of this
assumption, he effectively showed the index of refraction depends on the color
of the light ray and on the nature of the system of molecules. Cauchy intended
to develop the theory of light dispersion in a paper 'Sur la dispersion', but its
publication was suddenly interrupted by his departure from France in
September 1830 (49).
By the end of the 1820s, Cauchy had used his molecular theory of elasticity
in applications. However, he had still not completely abandoned the Eulerian
point of view that he held in 1822. In May 1830, in an article of the Exercices de
M athematiques, he presented a general continuum theory of elasticity for the
first time (50). This occurred after Poisson presented a paper to the Academie,
in which he suggested that in the general case the continuum theory leads to 36
coefficients of elasticity. In his article, Cauchy compared his own 2 theories, the
continuum one, with 36 coefficients, and the molecular one, with only 15
coefficients. In the special case of isotropy, two coefficients are obtained by the
first theory [Eq. (6.9)] and only one by the second theory [Eq. (6.16)]. Cauchy
then showed that the molecular theory could be derived, by reduction of the
coefficients, from the continuum theory. Such, then, are the origins of the
famous discussion on the 'Cauchy relations', which connect these two theories.
By 1830, Cauchy seems to have chosen the molecular theory, not for
doctrinal reasons but on the basis of experiments Savart had undertaken on
isotropic bodies. Three years later, in his lectures given at Turin, Cauchy
unhesitatingly opted for the molecular theory (51). He then justified his choice,
not on the basis of experiments, but on the basis of dogmatic arguments of a
philosophical and theological nature.
From 1815 to 1830, Cauchy was relentless in his pursuit of a single project
common to all the great physicomathematicians ofthat era, Laplace, Poisson,
Ampere, Fresnel, and many others. This project was the inquiry into the
mathematical laws governing the propagation of physical phenomena within
a medium, whether for liquid waves, as in 1815, or for light waves, as in 1830.
On that, in the tradition of Eulerian mechanics, Cauchy elaborated the basic
concepts of continuum mechanics in 1822; as a mathematician, Cauchy
thought that he had found in his continuum theory of elasticity a universal
model from which it would be possible to deduce laws governing such diverse
phenomena as sound, heat, and light. When, in the second half of the 1820's,
Cauchy substituted a molecular theory for his continuum theory, this project
became, in a sense, united with the Laplacian goal to create a universal
molecular physics. However, Cauchy treated the molecular theory merely as a
mathematical tool, albeit that it also became a metaphysical dogma for him,
and he neglected its physical implications. These features were especially
evident in his research on the theory of light.
 Chapter 7

From the Theory of Singular Integrals

to the Calculus of Residues

Cauchy's crowning achievement in analysis was unquestionably his theory of

complex functions, a branch of mathematics that, except for some interrup-
tions, commanded his attention from 1814 until his death. Few mathema-
ticians concerned themselves with Cauchy's theory before the late 1840s.
Almost all of the progress made in this area until that time was due to Cauchy.
The principal lines followed by his work are well known. Cauchy first
embarked on what would become his complex function theory by way of
studying the integration along closed paths in the complex plane; he did not
undertake to investigate analytic functions of a complex variable until 1831
and only much later, in 1846, did he begin to work out the fundamental
notions that would govern his complex function theory.
Actually, Cauchy constructed this theory in order to develop new
methods for calculating integrals, roots of algebraic and transcendental
equations, and solutions of linear ordinary and partial differential equations
with constant coefficients. These mathematical problems were closely connec-
ted with the study of mechanical topics. Nevertheless, the historians of
mathematics who have examined Cauchy's papers relative to this branch of
analysis have generally failed to consider this context. Accordingly, in this
chapter we explain how the development of the complex function theory fit
into the general framework of Cauchy's overall scientific concerns.
Cauchy presented his first paper on analysis, a major study entitled 'Sur les
integrales definies', to the Academie on August 22, 1814 (1). Accordingly to
Cauchy, this paper was the first work of reference concerning his complex
function theory. The original version of the paper, without the footnotes
Cauchy added in 1825, however, was in the tradition of Cauchy's great
mentors, Laplace, Legendre, and Poisson (2). In his 'Recherches sur les
approximations des formules qui sont fonctions de tres grands nombres' of
1785, quoted by Cauchy in the introductory section of his paper, Laplace, just
as Euler before him, had made use of complex numbers in integrating certain
107
108 7. From the Theory of Singular Integrals to the Calculus of Residues

real functions of a real variable, for instance, cos bxe - x 2 , between 0 and + 00
(3).
Laplace's method, however, was based on a simple induction that followed
from the then-accepted principle of the generality of analysis. On such a basis,
Laplace applied the usual methods of integration to complex functions of a
complex variable. These methods', he admitted, 'although used with great
care and due caution, nevertheless need prooffor the results they generate' (4).
Laplace examined this method again when, in 1812 and 1814, he published the
first two editions of Theorie Analytique des Probabilites. At that time, he
interested his protege Cauchy in the investigation of this question.
'I have conceived', wrote the youthful scholar in the introduction to his
paper, 'the hope of basing the passage from the real to the imaginary domain
on a direct and rigorous analysis'.
Cauchy first considered a real function feu), where u was supposed to be
itself a real function of two real variables x and y. He easily established the
equation

a [ au] a [ au] (7.1)

oy feu) ax = ax feu) oy .

Thus, Cauchy assumed that f and u were complex functions, which he wrote
u = M(x, y) + N(x, y)J=1 and feu) = P'(x, y) + P"(x, y)J=1.

Substituting these expressions in Eq. (7.1) and setting

s= p,oM _ p"0N U=p,oM _p"oN

ax ax ay ay

v=p,oN +p"oM
oy oy'
he stated the imaginary differential equation

oS + oTJ=1 = oU + av J=1
oy oy ox ox
equivalent to the two real differential equations

oS = au and aT = oV (7.2)
oy ox oy ox'
from which he deduced the classical equations
ap' oP" oP' oP"
-=- and -=--
ax oy oy ox
in the special case [M(x,y) = x and N(x,y) = y]. The differential equations,
(7.2), as he put it, 'summarize the whole of the theory relating to the passage
 7. From the Theory of Singular Integrals to the Calculus of Residues 109

from the real to the imaginary [domain]'. After having formed the double
integrals

and

(7.3)

over a rectangle R = [xo, X] x [Yo, y], where S, T, U, and V were supposed 'to
keep a determinate value', he obtained two fundamental equations between
two simple integrals by permuting the order of integration:

I x [S(x, Y) - S(x,Yo)dx
Xo
= fY [U(X,y) -
Yo
U(Xo,y)] dy

and

I x [T(x, Y) - T(x, Yo)]dx

Xo
= (Y [V(X, y) -
JyO
V(Xo, y)]dy. (7.4)

These formulas are equivalent to the famous Cauchy's integral theorem

applied to various closed paths, which depends on the mapping u of x and y. By
means of this theorem, Cauchy calculated a number of definite integrals.
The second part of the study was longer and still more innovative. Cauchy
assumed up to that point that the functions to be integrated were regular (i.e.,
had a 'determinate value') between the limits of integration. He focused his
attention on certain cases where this regularity failed to hold at certain isolated
points. It was then no longer possible to permute the order of integration in
Eq. (7.3), so, he posed, in section 11.2, the general problem of determining a
double integral:

fan ibn oK
a' b'
-dxdz
OZ
when the primitive K = 4>(x,z) assumes an indeterminate value at a point
(X,Z) in the rectangle [a',a"] x [b',b"]. Then, the value of the integral
depended on the order of integration. Cauchy obtained the difference A
between the two values depending on the order of integration in the form of
one or several singular integrals:

lim ft 4>(X ±~, Z ± Od~.

£-0 0
( .... 0

Singular integrals were nonnull definite integrals taken over infinitely small
intervals. This class of integrals, which Cauchy introduced into analysis, was
entirely new.
110 7. From the Theory of Singular Integrals to the Calculus of Residues

Likewise, in section 11.3, Cauchy examined the simple integral

f" l{J'(Z)dz

in the case where the primitive l{J(Z) of l{J'(Z) is discontinuous at a point Z in the
interval [b',b"]. If Z was a pole of l{J'(z), he obtained a determination of the
integral that he called his principal value in 1822. For instance,

f + 4dZ

-2
-=Log2
z
and

f + 2dZ =
-2 Z
O.

In section 11.4, Cauchy investigated the double integrals [Eq. (7.3)] in the case
of a function f =~. The conditions Cauchy stated for g and h were rather

imprecise, but actually he dealt only with functions f real valued on the real
axis and with simple poles Ui = U(Xi'Yi) on the border or inside u(R). Equations
(7.4) are thus generally false. In order to calculate the differences between both
sides, respectively, A and A', Cauchy substituted the first-order expansions

of M(x,y) and
N( »): ON(Xi' yJ ON(Xi' yJ
Xi' Yi +.. OX + 11 oy
of N(x, y) into the two singular integrals

lim
8-+0
.,-+0
Jo[8 S(Xi ± e,Yi ± 11) de
and

lim
8-+0
.,-+0
J[8 T(Xi ± ~'Yi ± l1)d~.
0

In this way, he obtained the values

A = 2J1.1t,
A' = 2A1t, (7.5)
where
JI. = - 1m [2: g(Ui) ]
h'(Ui)
 7. From the Theory of Singular Integrals to the Calculus of Residues 111

and
A = Re [ L :,~:i;) ]
if all the Ui are inside u(R). This result was equivalent to the residue theorem
applied to closed paths depending on the mapping of U (rectangle if u = idR ,
semicircle, triangle, etc.). If some U i were on the border of u(R), the values of A
and A' relative to these poles were p,n and An. This first version of the method of
residues was called by Cauchy the theory of singular integrals.
The final section of the second part of the paper, as well as the supple-
mentary parts written on the request of the evaluating commisioners, were
devoted to applications. Cauchy used the theory of singular integrals to
calculate integrals between - 00 and + 00 of real-valued functions with simple
poles in the half-plane Imz ~ 0, assuming only that the functions vanish at
infinity in the same half-plane. In the case of functions with poles on the real
axis, Cauchy obtained the principal value of the integrals. Unfortunately, he
also considered functions with an infinity of poles on the real axis [for instance,
if h(x) = cos xp(x)]. Such an applications could, of course, lead to incorrect
results.
The paper 'Sur les integrales definies' was only the most remarkable part of
a whole set of research studies on questions relative to the definite integrals:
calculation techniques, transformation properties, and applications to other
areas of analysis. It opened the way for a less famous study, 'Sur diverses
formules relatives a la theorie des integrales definies' which was presented to the
Academie on January 2,1815 (5). The key feature of this paper was a discussion
of methods of integration used in Laplace's Theorie Analytique des Pro-
babilites and, to a lesser extent, in Legendre's Exercices de Calcul Integral. In
the first part of his paper, Cauchy recalculated certain integrals that Laplace
had deduced by passing from the real to the complex domain. In performing
these reevaluations, Cauchy used a new method that was based on the
consideration of double integrals. This method differed entirely from the one
he had used in his paper of August 22, 1814.
The third part of the paper of January 2, 1815, was devoted to the
transformation of finite differences into definite integrals, one of Laplace's
favorite topics. Here, using new methods, Cauchy derived several important
formulas of the Theorie Analytique des Probabilites. Moreover, he introduced
a new class of integral (just as he had done the preceding year with singular
integrals) that he labeled 'extraordinary integrals': iff (x) is a function such that
f(O) -=I- 0, the integral fX ~;~)1 dx takes an infinite value, but the integral

fo
Xf(X) - ¢(x) dx, wher: ¢(x) is the nth-order expansion of f(x) in the
xa+ 1
neighborhood ofO, generally takes a finite value if n is the integer part of a. This
. Cauchy d
last integral, whIch '
eSlgnate d by I' xf(x) d . d'
a + 1 X, IS an extraor mary

integral.
112 7. From the Theory of Singular Integrals to the Calculus of Residues

During the following years, Cauchy developed the applications of the

theory of singular integrals. He used it in his paper 'Sur la theorie des ondes' of
October 1815 to establish, for a ~ 0, the inversion formulas

4>l(m) = Jr tXl cosmJ.lF 1 (J.l)dJ.l

Jr tXl
and

4>2(m) = sin mJ.lF 2(J.l) dJ.l (7.6)

of the Fourier transforms

Fl(a) = Jr teo 4>l(m) cos am dm

and

F 2(a) = Jr teo 4>2(m) sin am dm (7.7)

by showing that
OOfOO n
f cos am cos mJ.lF 1 (J.l) dm dJ.l = -Fl(a)
00 2
and

ff 00
o
00
0
sin am sin mJ.lF 1 (J.l) dm dJ.l = ~ F 2(a).
2
(7.8)

His proof, based on the use of the convergence-producing factor e- am and on

permutations of the signs Jand lim was not rigorous (6). Cauchy argued that

fo
OO 1(r.t.
cos am cos mJ.le- am dm =-2 2 (
r.t. + J.l-a
)2
r.t.)
+ r.t. 2 + (J.l+a )2·

Since lim e- am = 1, the left sides of Eq. (7.8) become, respectively,

and
. 1
hm-2
f+ 00
F 2 (J.l) 2
r.t.dJ.l
( )2'
a"" 0 0 r.t. + J.l + a
which are singular integrals. By the change of variable J.l = a + r.t.¢, Cauchy
obtained for these singular integrals the expressions
 7. From the Theory of Singular Integrals to the Calculus of Residues 113

and
1 r+
2, F 2(/1) J0
oo d~
1 + ~2'

equal, respectively, to IF 1(/1) and IF 2(/1)·

Thereafter, between 1817 and 1821, Cauchy's productivity in mathematics
declined sharply. Absorbed in teaching at the Ecole Poly technique, he
presented only six papers to the Academie during this four-year period, and
none at all for the two-year interval from November 1819 to October 1821.
However, he continued to work on applications of singular. integrals.
Cauchy first presented his findings at the College de France in 1817 (7). In
his lectures, he extended the use of his method of 1814 to the calculation of
improper integrals:

f +oo

-00
u(x)
4J(x)-() dx,
vx
where 4J(x) is the real part of a complex-valued function 4J(x) + X(x)j=1 of a
real variable, and applied the formula to the special case

4J(x) + x(x)j=1 = cos rx + sin rx j=1.

He also investigated the properties of the Fourier transforms, Eq. (7.7), and of
the inversion formulas, Eq. (7.6), called by him 'reciprocal functions'; he
proposed various applications of them, which he had already discussed at the
Societe Philomatique on March 20, 1816 (8). From these lectures, Cauchy
extracted an article on reciprocal functions for the Bulletin of the Societe
Philomatique (9). On the other hand, Cauchy developed some considerations
about the theory of singular integrals and the reciprocal functions in his
analysis teaching at the Ecole Poly technique during the year 1817-1818
(10).
Fourier now made an objection, claiming prior discovery (11). Actually, he
had already set forth the properties of his transforms in 1807 and, using them,
had developed a method for solving the linear partial differential equations
related to the theory of heat. This method was identical to Cauchy's. Fourier's
paper had been deposited in the archives of the Academie since 1811. However,
Cauchy was completely ignorant of its existence, for it was not yet published,
and proclaimed his good faith in a second article, which appeared in the
December 1818 issue of the Bulletin ofthe Societe Philomatique (12). Without
questioning whether or not Cauchy had indeed acted in good faith in this
matter, we are nevertheless inclined to believe that Fourier's work indirectly
influenced Cauchy's research.
In fact, Poisson and probably Laplace, unlike Cauchy, knew of Fourier's
unpublished study on the theory of heat. Poisson made use of formulas that
were identical to those appearing in Fourier's paper in his own 1815 paper on
114 7. From the Theory of Singular Integrals to the Calculus of Residues

the theory of waves. Cauchy was on very friendly terms with Laplace and
Poisson in 1815, and as we have seen, he had discussed the theory of waves
with both of them at this time. Accordingly, through Laplace and Poisson,
Cauchy might well have been aware of Fourier's results. Such an hypothesis
might explain why in his research he failed to use a basic method that had
already enabled him to integrate the equations of the theory of waves and
instead employ the method used by Fourier and Poisson (13).
After an interruption of two years, Cauchy came back to the theory of
singular integrals in 1819. On November 22, 1819, he presented a paper to the
Academie on the determination of the roots of an algebraic or transcendental
equation in integral form (14). The method was based on the theory of singular
integrals. Cauchy considered a function f(u)/uF(u) with real simple poles 0,
a, a', a", etc., between - 1 and + 1 and complex simple poles oc + {J~,
oc' + {J'~, etc., inside the upper unit semicircle. Integrating f(u)/uF(u) along
this circuit and using for the first time integrals of a complex function, he
applied the residue theorem and wrote

I" f(cosp + sinp~) d

Jo F(cosp + sinp~) p

=F-1f+1 f(r) dr
-1 rF(r)

f(O) f(a) f(a') ]

+ 11: [ F'(O) + aF'(a) + a'F'(a') + ...
f(oc + {Jj=1)
+ 211:[ + f(oc'+{J'j=1)
+ ... ]
(oc + {Jj=1)F(oc + {Jj=1) (oc' + {J'j=1)F(oc' + {J'j=1) .

Unfortunately, he never published the paper.

This work gave an impulse to new investigations relative to the theory of
singular integrals. During the following months, Cauchy wrote a valuable
paper, entitled 'Recherches sur les integrales definies qui renferment des
exponentielles imaginaires' (14). In this paper, he systematically used the
residue theorem of 1814 applied to circular paths. Moreover, he definitively
gave up the idea of separating the real and imaginary parts of complex
equations, which characterized the paper of 1814. Therefore, he expressed his
findings by means of imaginary integrals:

L: u(x) + v(x)F-1 dx

containing imaginary exponentials

erxp = cos rx + sin rx~.

Thus, we know that he exhibited important results in this paper, such as the
 7. From the Theory of Singular Integrals to the Calculus of Residues 115

integral expression of the nth derivative:

!f+"
2 _"
e- npFi f(b + ePFi)dp = ~ dnf~b),
n. db
(7.9)

Cauchy's integral formula in the special case of the unit circle:

11" [ f(e pFtPFt) + f(e- PFt

-2 1
) ]
-1Ft dp1 = nf(a) (a < 1), (7.10)
° -ae -ae
and the average property:

~ f: [f(eP Fi ) + f(e- pFi)] dp = nf(O).

Probably, it is not certain, he also obtained the integral formula

cjJ(u o) + cjJ(u 1 ) + ... + cjJ(Um-l) = 2n f+"

1 _" rePP f(reP Ji ) dp, (7.11)

where cjJ is a continuous function, Ui the (simple) roots of the equation

F(u) = 0 inside the disc with center 0 and radius r, and f(u) the function
cjJ(u) F'(u). Strangely enough, Cauchy did not present this paper to the
F(u)
Academie until September 16, 1822, nearly three years after it was begun, and
he never published it.
One month after, on October 28, Cauchy presented another study on the
theory of singular integrals to the Academie (15). In this paper, entitled 'Sur les
integrales definies, ou ron fixe Ie nombre et la nature des constantes arbitraires
et des fonctions arbitraires que peuvent comporter les valeurs de ces memes
integrales quand elles deviennent indeterminees', he summarized all the
improvements he had introduced into the theory of singular integrals since the
presentation of his memoir 'Sur les integrales definies' of 1814, which was still
unpublished. Moreover, he made important theoretical innovations: he
defined the new concept of principal value and generalized and radically
simplified the intricate formulas of 1814 by means of imaginary integrals. The
paper was unpublished but a 'very short analysis' of it immediately appeared
under the same title in the Bulletin of the Societe Philomatique (16). In 1823, he
twice summarized the theory with its improvements and new applications, first
in an appendix to an article on the linear partial differential equations (17) and
then in his Calcul I rifinitesimal of 1823 (18).
Cauchy began by considering a real function of a real variable f(x) with
some poles xo,x1, ...,Xm-l within the interval [x',x"]' Then, he defined the
principal value of the integral f:'''f(X)dX as the sum
lim fXo-, f(x)dx + f X' -' f(x)dx + ... + (X" f(x)dx.
£-0 Xl xo+e JX m -l+£
116 7. From the Theory of Singular Integrals to the Calculus of Residues

For instance, the principal value of f -1

+ 1 dx is 0, the value that Cauchy had
x
given in 1814. If the integral is taken from - 00 to + 00, its principal value
IS

lim IXO-E f(x)dx + f X1 - Ef (X) dx + ... + fIlE f(x)dx.

e-O -1/f. xo+e Xm -l+C:

In order to obtain the general value of the integral, which is indeterminate in

most cases, it is necessary to add singular integrals of the form
-IIEI"
lim I f(x)dx,
-liE

lim [( fX'-EI"'f(X) dx + fX'+E f(x) dX)J,

xo+ e Xi +eVj

and
IIE
lim I f(x)dx,
1 lev

where f.,/" f.,/,i' Vi' and V are arbitrary positive constants. If f- = lim xf(x),
x--+ - 00

fi = lim (x - xJf(x), and f+ = lim xf(x) exist and are finite, Cauchy
X-Xi x- + 00
obtained the formulas

lim [(fX'-EI"'f(X)dX + fX'+E f(X)dX)J=fiIOgf.,/,i,

£-+0 Xj-e Xi+eVj Vt

-1/EI"
lim I f(x) dx = f-Iogf.,/"
E-O -liE
and

limf IlE
E-O lin
f(x) dx = f+ log-
1
V

by using the integral mean-value theorem Eq. (5.2), with x(x) = _1_ and
X-Xi
1 .
x(x) = -. For mstance,
X

I
+1 -dX = lim (I-El1dX
- + fEdX
- = lim [ log (ef.,/,) + log- 1J = (f1)
log -

I
-1 X E-O -E X EV X E-O ev V

and the general value of + 1 dx is indeterminate, since it depends on f.,/,

-1 X
and v.
 7. From the Theory of Singular Integrals to the Calculus of Residues 117

Cauchy then examined the case of two real functions of two real variables
4J(x, y) and X(x, y) satisfying the relation
o4J(x, y) OX(x, y)
(7.12)
ax ----ay'
He proceeded in the same manner as in 1814. From the equations

ry y
o4J(x, y) dx dy = IX r ox(x, y) dx dy,
IXo JyO ax
x
Xo JyO oy

he deduced the relation

I x
Xo
4J(x,Y)-4J(x,Yo)dx= r x(X,y)-x(xo,y)dy,
JyO
y
(7.13)

which is to be compared with Eq. (7.4), on the assumption that 4J(x, y) and
X(x,y) are finite and continuous in R = [x o, X] x [Yo, Y]. If 4J(x,y) and X(x,y)
have one pole (a, b) in R, Eq. (7.13) deals only with the principal values of the
integrals, and we have, therefore, the equation

I x
Xo
4J(x, Y), - 4J(x,yo) dx = r x(X, y) -
y

JyO
X(Xo, y) dy - Ll, (7.14)

where

Ll = limr x(a + k, y) - x(a - k, y) dy.

k~O J
y

Yo
(7.15)

Cauchy applied these formulas to complex functions. This extension was

justified in his own opinion by the definition of imaginary expressions as
symbols equivalent to two real expressions, which he had provided in his
Analyse Algebrique of 1821. Cauchy argued as follows: let f be a complex
function of a complex variable and u a complex function of two real variables.
The relation Eq. (7.12), is obviously satisfied by 4J(x, y) = f(u) au and
ax
x(x,y) = f(u) :;, and therefore, we have Eq. (7.13), if the functions are finite and
continuous, or Eq. (7.14), if they have one pole Uo = a + b.j=l inside
u(R).

For instance, iff =~, with g continuous and finite and h having simple poles
g
inside u(R), Eq. (7.15) yields

Ll = 2n.j=l p. + Il.j=l),
which is equivalent to Eqs. (7.5). In the special case u(x,y) = x + y.j=l, it
follows from Eqs. (7.14) and (7.15), by substitutingf(x + y.j=l) for 4J(x,y),
118 7. From the Theory of Singular Integrals to the Calculus of Residues

and Fif(x + yFi) for X(x,y), that

L:f(X + Y F i ) - f(x + yoFi) dx

= Fi i f(X + yFi) -
JyO
Y
f(xo + yFi) dy - A, (7.16)

where

A = Filim fff(a+k+yFi)-f(a-k+yyC1)dy. (7.17)

k-+O J
Yo

In order to determine the singular integral A, Cauchy used the substitution

y = b + kz and the mean-value theorem Eq. (5.2). Thus, he obtained the
formula
A=2nfyC1, (7.18)
where

f = lim kf(uo + k). (7.19)

k-+O

The limiting value f is, of course, the residue off at the simple pole U o, and
accordingly, the residue theorem can be expressed in the form

f~f(X + Y F i ) - f(x + yoFi)dx

= F-IfY f(X + yF-I) - f(xo + yF-I)dy

Yo

(7.20)
where fo, flo ... , fm- 1 are the residues offat the simple poles Uo,U 1, ••• ,Um - 1
inside R.
For a = Xo or a = X (with Y > b > Yo), however, one must take the principal

l:
value of

f (X + Y yCl) - f(x + YoyC1) dx

and attribute the value nfJ=!"
to A in Eq. (7.15) and, likewise, if b = Yo or
b = Y (with X > a > xo).
Cauchy also discussed the case u(r, p) = reP R . If r = 1, he obtained the
equation

f:,," ePR f(epR)dp = 2nFi(fo + fl + ... + fm-d, (7.21)

where the residues fo, f 1 , .•• , fm- 1 are at the poles inside the unit circle. The
 7. From the Theory of Singular Integrals to the Calculus of Residues 119

formulas of the paper of November 22, 1819, are immediate consequences of

this equation.
In 1822, just as in 1814, Cauchy investigated only functions with simple
poles inside R. In a footnote to his article of July 1823 (19), he also gave the
value of the residue f of f at a multiple pole U j of order p in the form
. dP- l kPf(uj+k)
!~ dkp-l (p _ I)! ' (7.22)

which generalizes formula (7.19). Cauchy gave no proof of Eq. (7.22), but
further articles allow us to restore his arguments: he made the change
of variable y = b + kz in the singular integral Eq. (7.17) and then he ex-
panded f[uo + k(1 + z.j=1)] = k P(1 + z.j=1)P f[u o + k(1 + z.j=1)] to
the (p - l)th order with respect to k, obtaining:
P-l(1 Il)P-l
f(uo) + k(1 + zF-1)f'(u o) + ... + + Zy - lc f(P-l)(UO)
(p -I)!
+ w[(uo + k(1 + zF-1)].
The substitution of this expansion in the singular integrals yields a sum of
p integrals, all vanishing except
f(P-l)(U
0
1
J=1 +
)f+oo1
J=1 dz.
(p - I)! -00 1 + z -1 -1 + z -1
Thus. he once more obtained the residue formula, Eq. (7.18), but with
f(P-l"J(u O) .
f= , that IS
(p -I)!
f r
dP-l kP f(u o + k)
= k~ dk p - l (p -I)!

Therefore, Eqs. (7.20) and (7.21) also deal with functions having multiple
poles.
Cauchy published all these results on the theory of singular integrals while
working on the resolution of the equations by means of integrals. In fact, it was
by way of his research into the problem of solving linear ordinary and partial
differential equations with constant coefficients that Cauchy was led to perfect
his theory of singular integrals. For this reason, we will examine the work that
Cauchy did between 1815 and 1823 relative to this topic before pursuing the
historical development of his complex function theory.
In his paper 'Sur la theorie des ondes' of 1815, Cauchy exhibited the general
solution of the equations he had examined, Laplace's equation and the
equation of propagation,
04y o2y
OZ4 + g2 OZ2 = 0
in the form of Fourier integrals containing an arbitrary function that was
determined a posteriori from the initial and boundary conditions. He did not
120 7. From the Theory of Singular Integrals to the Calculus of Residues

set forth in precise terms the methods by which he had arrived at the formulas.
It was probably by induction. However, he tried to establish their generality by
giving their series expansions and by comparing them to the results he
obtained by the method of undetermined coefficients, 'a method', he wrote,
'that possesses all generality possible' (20).
Over the following years, Cauchy did not get an opportunity to develop his
solution method to problems of mathematical physics. His appointment to the
chair in mechanics at the Faculte des Sciences of Paris in 1821 led him back to
this research area. Poisson, Cauchy's arch rival in analysis, had just published
an important study in which he solved equations relating to vibrating elastic
surfaces, to the distribution of heat in solid bodies, as well as the motion of
fluids (21). Cauchy also presented several papers to the Academie in these
same subject areas. In these papers, he developed a general method involving
Fourier transforms for solving linear partial differential equations with
constant coefficients and applied it to the main equations of mathematical
physics.
In the first paper, 'Sur l'integration generale des equations lineaires a
coefficients constants', which he presented to the Academie on October 8 1821
(22), Cauchy substituted a single Fourier transform with an imaginary
exponential for the two transforms with circular functions for the first time.
Thus, he expressed a function f(x, y, z, ... ) of n variables contained in
[Jl', Jl"] x [v', v"] X [m', m"] x ... by a multiple integral:

( _ 1
)nf
+OO f+oo f+OO fl'· IV. fro.••• • •• e.(I'-X)~e/i(V-Y)~
2n - 00 - 00 - 00 I" v' w'

eY(W-Z)~ ••• f(Jl, v, m···)da.dP dy···dJldy dm ....

Then, he showed that the integral formula

( _1 )nf+oo f+oo f+OO fl'· IV. fro. ...e9ite·(I'-X)~e/i(V-Y)~

...

2n - <Xl - 00 - 00 1" v' ro'

eY(ro-Z)~ ••• f(Jl, v, m···) da. dP dy··· dJl dv dm···

represents a solution 4J(x, y, z, ... , t) satisfying the initial conditionf(x, y, z, ... )
for t = 0 ofthe homogeneous linear partial differential equation (ofmth order
relative to t) with constant coefficients:

( a ' ay'
Fax
a a a)
az ' ... , at 4J(x, y, z, ... , t) = 0, (7.23)

if OJ (a., p, y, ... ) is one of the m roots of the algebraic equation

F(a.J=1,pJ=1, yJ=1, ... ,0) = O.
Finally, he sought to determine the solution of the equation, knowing the
values of 4J and of its derivatives relative to t up to order m - 1 for t = 0 and
from this to deduce the solutions of the main linear equations of mathematical
 7. From the Theory of Singular Integrals to the Calculus of Residues 121

physics satisfying given initial conditions in integral form. In an unpublished

portion of this work, he generalized his method to inhomogeneous equations.
While this study was being written, Cauchy remained unquestioning in his
belief in the generality of the solutions he had obtained by his method, since he
could always set them in infinite series expansions equal to the expansions
deduced by the method of undetermined coefficients. However, in the study
'Sur Ie developpement en serie et sur l'integration des equations differentielles'
he presented some three and a half months later to the Academie (23), he
showed that the method of undetermined coefficients is deficient. Indeed, from
the existence of the function

whose series expansion in the neighborhood of 0 is identically the null series,

he deduced the fact that a given MacLaurin expansion can actually
correspond to several functions and, consequently, the methods of undetermined
coefficients do not guarantee the generality of a solution of an ordinary or
partial differential equation. For this reason, in an addendum to this paper,
Cauchy expressed his misgivings about 'the discussion in its present state'
relative to the generality of the formulas representing the solutions of partial
differential equations of mathematical physics (24).
A year later, on September 16, 1822, Cauchy presented a new paper,
'Sur l'integrations des equations lineaires aux differences partielles a coeffi-
cients constants et avec un dernier terme variable' to the Academie (25). In
this paper, which was an expansion of his October 8, 1821, study, he set
forth the main results given in his lectures at the Faculte des Sciences.
Cauchy began by expressing F ( :x ' :y' :z '... :t) in Eq. (7.23) in the form

L
p;m (0 0 0 ) oP
jp ;)':1':1"" "P' Then, developing an idea that was present in the
p;o uX uy uZ ut
paper of October 1821, he reduced the problem of solving a linear partial
differential equation, with given initial conditions for ¢ and for its derivatives
relative to t up to order m - 1, to the simpler problem of solving the linear
ordinary differential equation of mth order,

p;m (jP
L jp(IXJ=1, pJ=1, yJ=1"")"P
p;O ut
S(t) (7.24)

satisfying the initial conditions S(O) = 1, :t S(O) = u ... , :t:~ll S(O) = urn-l,
where u is some variable. Cauchy solved this equation by the classical method
he expounded in his first-division course at the Ecole Poly technique.
Moreover, in a paper that he presented to the Academie on May 26,1823
(26), Cauchy gave a new method of solving linear ordinary differential
122 7. From the Theory of Singular Integrals to the Calculus of Residues

equations with initial conditions, such as Eq. (7.24). This method was based on
his theory of singular integrals. It consisted of applying the residue theorem,
. F(O)
Eq. (7.20), to the functIon (u _ O)F'(lJ)' where F(lJ) is the characteristic
equation. Thus, the solution was expressed in integral form. It was in this
context that Cauchy extended the residue theorem to functions with multiple
poles in order to treat the case of characteristic equations with multiple roots.
At the end of 1823, the construction of a unified theory of singular integrals,
a task that had been begun 10 years before, seemed to have been achieved.
When his study 'Sur la tMorie des ondes' was registered for publication on
May 17, 1824, Cauchy appended a note to it on the theory of singular integrals
in which he made no additions to his results of 1822 and 1823 (27). But,
suddenly, early in 1825, just as he was teaching mathematical physics at the
College de France, he announced discoveries of fundamental importance
relative to the basis and applications of his theory of singular integrals. In fact,
between January 31, and February 18, 1825, he presented three papers on
analysis to the Academie, in which he investigated the calculus of residues and
the theory of curvilinear integrals of complex functions. The circumstances
that led Cauchy to write these papers remain a mystery to this day, since the
study of February 14 has not been published and the other two works are
known only by way of what appears to be later versions. Although we are
unable to completely explain this affair, we will nevertheless attempt to shed
some light on it. In order to come to grips with the mystery surrounding these
discoveries, we have to consider the influence of three quite different
mathematicians.
The first ofthese mathematicians was Poisson. In 1820, Poisson published a
paper on definite integrals, which was perhaps, as is frequently stated, the
origin of Cauchy's innovation (28). Poisson proposed to integrate a real
function having a real pole between the limits of integration by passing
through a neighborhood of the pole along a path in the complex plane. The
value thus obtained, he asserted, was equal to the difference of the values of the
primitive of the function at the two limits of integration. Twice, in 1822 and
1823, Cauchy criticized this method of integration (29). He was especially
annoyed with Poisson for attempting to preserve the definition of definite
integral deduced from the primitive. He also rejected the idea of 'passing
through a sequence of imaginary variables' in order to integrate a function
between real limits, arguing that this conception led one to giving imaginary
values to integrals taken between real limits. In spite of these criticisms,
Cauchy was probably prompted by the paper of Poisson to develop his own
ideas on complex integration.
The second mathematician, Bamabe Brisson, is less famous. Some 12 years
older than Cauchy, he was a brilliant chief engineer at the Ponts et Chaussees.
He was also interested in pure and applied mathematics, to which he seems to
have devoted his spare time, in particular to the theory of linear partial
differential equations (30). In 1804, 1821, and 1823, Brisson presented papers
 7. From the Theory of Singular Integrals to the Calculus of Residues 123

on this subject to the Academie. Cauchy was appointed reporter for the
evaluation commission on the paper of 1823, a work that was never published
and seems to have been lost. In this study, Brisson appears to have developed a
symbolic calculus derived from the Leibnizian analogy of powers and
differences. He had already stated the elements of this calculus in his 1821
study. In the 1823 paper, he used these first results to obtain solutions for linear
ordinary and partial differential equations in symbolic form. From these
solutions in symbolic form, he proceeded to obtain series expansions and
integral representations. Because of a disagreement between the members of
the evaluation commission, Cauchy delayed submitting his evaluative report
to the Academie (31). But, without delay, on December 27,1824, he introduced
Brisson's symbolic calculus, as revised and corrected by himself, in a long
memoir devoted to his own method of solving the equations using Fourier's
formula (32). The aim of this paper was to simplify this last method by using a
concise notation. Thus, toward the end of 1824, Brisson's work had prompted
Cauchy to focus attention on the possibilites of a new calculus for linear
differential equations.
Elsewhere in his study, Brisson used an integration formula 'that rests',
Cauchy critically wrote in his evaluative report, 'on the consideration of
definite integrals that would assume imaginary values while the function under
the f sign retains a real value' (33). In a new study that he was working on (34),
Brisson, perhaps mindful of Cauchy's criticisms, used integrals taken between
imaginary limits of integration. Cauchy, who quickly grasped the fruitfulness
of Brisson's idea, became interested in this new class of integrals.
We turn now to the role of yet another mathematician, Michael Ost-
rogradski. Ostrogradski had come from Russia to Paris in 1822 to study
mathematics. Though still very young, he was nonetheless, Cauchy wrote,
'blessed with great abilities and well versed in infinitesimal analysis' (35).
Ostrogradski presented two highly significant papers to the Academie during
the summer of 1824. In his paper of July 24, 'Sur la difficulte que se rencontre
dans Ie calcul des integrales definies lorsque la fonction Ii integrer est
discontinue entre les limites d'integration', he criticized Poisson's 1820 study
on definite integrals (36). Ostrogradski's criticisms coincided with Cauchy's to
the extent that he reproached Poisson for his notion of definite integrals based
on the consideration of primitives. But, it was Ostrogradski's paper of August
7, 'Remarques sur les integrales definies', that was particularly instrumental in
drawing Cauchy's attention to the young Russian (37). In this study,
Ostrogradski proved-albeit in a clumsy but fundamentally correct way-
that the residue of a function at a multiple pole assumed the value Cauchy had
declared without proof in 1823. In a note to the Academie on February 14,
1825, Cauchy alluded to this study (38). But, rather curiously, while we see this
paper as a study closely related to the calculus of residues, Cauchy merely
mentioned this work incidentally in discussing his own research on integration
along arbitrary paths in the complex plane. Moreover, he even pointed out
(quite incorrectly) that in this study Ostrogradski had used integrals taken
124 7. From the Theory of Singular Integrals to the Calculus of Residues

Mikhail Ostrogradski (1801-1862). This Russian mathematician took part in the

invention of the calculus of residues. Photograph by 1. L. Charmet,
permission of the Academie des Sciences.

between imaginary limits. Ostrogradski, in fact, used only integrals between

real values and merely indicated that, although the function was real between
the limits of integration, it might, under certain conditions, become imaginary.
On January 31,1825, Cauchy presented the paper 'Sur un nouveau genre de
calcul analogue au calcul infinitesimal' to the Academie. One year later, on
February 27, 1826, the manuscript was deposited with the Secretariat, and
several days later the work was published in Exercices de M athematiques (39).
We can reasonably assume that the original version of the work - that is, the
version of January 1825-scarcely differed from the February 1826 version. In
this study, directly prompted by Ostrogradski's paper of August 7, 1824,
Cauchy announced a new calculus, which he called the calculus of residues. He
 7. From the Theory of Singular Integrals to the Calculus of Residues 125

Nouveau Memoire sur Ie Calcul des Residus et sur les integrales Definies, 1825.
Manuscript by Cauchy. Published by permission of the
Academie des Sciences of Paris.

drew a parallel between the differential coefficient of a function J at the regular

point u o, which can be defined as the coefficient of U - Uo in the expansion of
the function in the neighborhood of uo, and the residue of the same function
relative to a pole Uo of order m, which he defined as the coefficient of
(u - uot- 1 in the expansion of the function f(u) = (u - uotJ(u) in the
126 7. From the Theory of Singular Integrals to the Calculus of Residues

t m- 1 )( )
neighborhood of u o, that is, (m _ ~~ or, by setting k = U - U O,

d"'-l kmf(u +k)j<m-l)(u)

!~ dkm 1 (m ~ 1)1 (m _ 1)~ [to be compared with Eq. (7.22)]. Thus, he
transformed Ostrogradski's demonstration into a definition.
The calculus of residues was not really a novelty since Cauchy had already
developed the mathematical tools necessary to calculate the residue of a
function at a simple pole, which he denoted by f; moreover, in 1823, he had
even determined the value of the residue of a function at a multiple pole.
However, up to then these residues had been obtained by means of singular
integrals, so that Cauchy had not thought it necessary to develop a special
terminology and notation. It is not clear, however, whether Cauchy already
possessed the notation for the extraction of residues at the beginning of 1825.
This uncertainty follows from the fact that, prior to 1826, Cauchy did not make
use of this notation which resulted from his collaboration with Ostrogradski.
On February 7,1825, one week after having presented his first paper on the
calculus of residues, Cauchy wrote a note announcing a new paper 'Nouveau
memoire sur Ie calcul des residus et sur les integrales definies', to the Academie
(40). In this note, he indicated a generalization of his calculus of residues to the
case in which the order of multiplicity of a pole was replaced by a fractional
power. In the end, however, Cauchy gave up on this particular generalized
calculus of residues, and he never published his paper. His mistake in this
matter clearly followed from the failure to define the class of functions to
which his new calculus should apply. Between February 7 and February 14,
Cauchy inserted several lines in his note announcing new research on
definite integrals between imaginary limits. This research, he said, was
contained in the last paragraph of his paper. Citing Brisson and Ostrogradski,
Cauchy articulated the crux of what would be his outstanding paper, 'Sur les
integrales definies prises entre des limites imaginaires', which he presented to
the Academie on Feburary 28, 1825.
This latter study was the most important step from the crude theory of
singular integrals of 1814 toward the sophisticated articles on meromorphic
functions of the 1850s. Aware of the significance of this paper, which was the
outcome of 10 years of research, Cauchy had the entire paper edited in August
1825 (41).
The essential innovation was the definition and the use of integration in the
complex plane. Just as he had done in Calcul Infinitesimal in 1823, Cauchy first
defined the integral

I X+Y,J=1
xo+Yo,J=1f(u)du

as the limit, as the intervals [xo, X] and [Yo, Y] are indefinitely subdivided, of
the finite sums:
 7. From the Theory of Singular Integrals to the Calculus of Residues 127

+ [(X2 - Xl) + (Y2 - Yl)J=!J!(X l + Ylj"=l) +

... + [(X - Xn- l ) +(Y - Yn-l)J=!J!(X n- l + Yn-1J=l),
where Xo ~ Xl ~... ~ X and Yo ~ Yl ~ ... ~ Y. Then, for any path from
Xo + YoJ=1 to X + Y J=1,he introduced the parametric representa-
tion c{J(t) + X(t)J=! (with c{J(t o) + X(to)J=! = Xo + YoJ=! and
c{J(n + x(nyCl = X + y yCl).
In the case of a function that is finite and continuous, Cauchy proved that
the value of its integral is independent of the path of integration between
Xo + YoJ=! and X + Y J=! by using the calculus of variation. Following
this, he examined the case in which the function had (simple or multiple) poles
in the domain between two given paths. In an early stage of his investigation,
Cauchy made use of the theory of singular integrals, which he had been
developing since 1822, in order to obtain the residue theorem for the case of
two nearby curves (he also used a method based on the consideration of the
whole part of the function); and in a later stage, he generalized this result to the
case of two arbitrary curves. He represented a path as a curve in the complex
plane and then considered a 'variable and moving curve whose form is
such that over any two distinct time intervals it can be made to coincide
successively with two fixed curves' representing the two paths under con-
sideration. However, he restricted himself to paths in the rectangle
{x + y.J=l:(X,Y)E[Xo,XJ x [Yo, YJ} and indicated without proof that the
theorem 'holds even when such contours (i.e., paths) are not confined to
the rectangle' (42). From the general residue theorem, Cauchy easily derived all
the formulas that had been obtained over the preceding years. In concluding,
he gave the values of a number of improper integrals.
Up until now, Cauchy had formulated all his research on definite integrals
in solidly analytical terms and scarcely paid any attention to formulations
based on geometrical considerations; for, in his view, such formulations lacked
rigor. However, in the present situation, the fact that paths in the complex
plane could be represented geometrically enabled Cauchy to explain simply
their relevant topological properties, even if he also analytically defined the
homotopy of two paths by means of a parameter. Later on, when this paper
was out of the way, Cauchy did not in any sense renounce his view of
imaginaries as symbolic expressions, because such a view seemed to him to be
the only one that could be rigorously justified.
The simultaneous creation of the calculus of residues and the theory of
integration in the complex plane corresponds, in Cauchy's works, to two
radically different directions. The calculus of residues is fundamentally a
formal calculus that arose from the theory of singular integrals, but was
expounded in its own right in an abstract manner. It seems that Cauchy had
been presenting this new calculus in his physics courses at the College de
France during 1824-1825, when he was striving to perfect Brisson's symbolic
calculus. He perceived that the calculus of residues could be of use to him in
128 7. From the Theory of Singular Integrals to the Calculus of Residues

simplifying the calculations arising in connection with the solution of

differential equations. On the other hand, the theory of integration in the
complex plane represented an unusual attempt to reinterpret and generalize
all the results Cauchy had obtained between 1814 and 1823 by using his theory
of singular integrals. Cauchy, of course, was perfectly aware of these opposing
tendencies: he relegated his studies on the calculus of residues to the broad area
of 'calculi'; because, as concerns residues, the focus was on algorithms that
would render his theory of singular integrals more amenable. On the contrary,
he consigned his paper 'Sur les integrales definies prises entre des limites
imaginaires' to the theoretical realm, thus completing the goal he had set for
himselfin 1814, namely, to endow the method of integrating by passing from
the real to the imaginary with a rigorous theoretical foundation. Therefore, he
made no allusions at all to the calculus of residues in the notes that he
appended to his 1814 paper 'Sur les integrales definies', which in its updated
version was handed over to the Secretariat of the Academie on September 14,
1825, for publication (43). For that matter, they were not mentioned in an
important article published on October 1825 in the Annales de Mathematiques
of Gergonne (44), in which he again proved the residue formula he had given in

f::
1814 for the calculation of

f(u)du,

specifying more rigorously the conditions that must be imposed on the

function f(u) and generalizing the formula to the case of a function with
multiple poles in the half-plane 1m u > O.
It was not until 1826 that he began to publish his first works on the calculus
of residues in Exercices de Mathematiques, which had begun to appear at this
time. He proposed the specific notation

:J::;o«(f(u)))
for denoting all the residues of f inside the rectangle R = {x + yj=1,
(x,y)e[xo,X] x [yo,yJ). The expressions

x ("'y -I..(u)« (u))) x ("'y «cjJ(u)))

xouyo'l' X 'xo<""yO X(u)
and
x ("'y cjJ(u)
xo <""YO «X( u)))

. cjJ(u) cjJ(u) .
were used to denote the resIdues of cjJ(u)· X(u), X(u) and X(u) ,respectIvely, at the

poles of X(u), cjJ(u), and X;U) contained in R. Moreover, Cauchy called the
integral residue, the sum of all the residues of the functionf(u). One of the first
 7. From the Theory of Singular Integrals to the Calculus of Residues 129

SUR UN NOUVEAU GENRE DE CALCUL

ANALOGUE AU CALCUL INFINITESINAL.

On sait que Ie calcul dill't!rentiel qui a tant conlribuci aux progres de I'analyse, e t
ronde sur In consideration des coefficients differentiels ou fonctions derivees. Lorsqu'oll
allrihue b une variable independante X l1n accroissement infiniment pelit " une
fonction f( x) de celie ,·nriable l'8~oit clle-meme on general un nccroissement infini-
mcut pelit dODt Ie premier term a est proporliouuel b " et Ie coefficient fini de
dan~ I'accroisscweot ,Ie I. fonction cst ce qu'on nom me Ie coefficient dilferentiel. Co
coefficient subsiste. quel'l0c soit ru, at ne peut ,'evanouir constamment que dans Ie
cas ou la fonction proposCe se rcdllit 11 une quantile constanta. II n'en est pas de memo
d'un autre coefficient dont nous allous parler. ct qui est generalement nul. e:,<cepte
pour des voleur, particulierO$ do la variable ro. Si. apres avoir chcrche lei valcm's
de x qui rend.mt In fonction f( x) infinie, on ajoule 11 I'uoe de cos valeur,. 0.;-
signce pu (I),. I. quantile infiniment petite " puis, quo I'on d6vcloppe f(x, +.)
su ivan! lea puissances ascendanLes de In m6mc quantite!. Iss premiers lermes du deve-
loppement renfermcront des puissances negalives de " et ['un d'cul< ser~ Ie produil
de -'- par un coelliciunl fini ,que nou~ appeUeroD5 Ie resid" de la fODction f( x) re-
•
[otiC k 10 valeur parliculi~re x, de 10 variable ro. Les resi.lus de celie cspecc 5e
presenteDt naturellement dans plusieurs !'rancbes de I'analyse algebriquo et de I'analyse
iufiniitsimsle. Leur consideration fouroil des methodes simples ct d'un usage facile, qui
s'appli'lucnt II un grand nomhre de 'lucstions diverses, et des formules nouvelles qui
paraissen! meriler I'atltmtion dtl$ geomctres. Ains;, par excmple. on dedui! immediate-
ment du cDlcu[ de, ~idus In formulc d'interpolation de Lagrange, [n decomposition
des fractions rationnelles dans Ie cas des racines egalcs ou inegales, des formules gene-
rales prepres ~ dCtel'mincr les valeuTs des integrales dcfinies, la sommation d'une mul-
titude de serie. et particulierement de series pcriodiques, I'integration des equations li-
neaircs oux dill'erences finies ou infioimcoL petites et b coefficients constants, avec ou
sans dernier terme variable, la serie do Lagrnnge ct d'ouLrcs stlries du meme genre, 1(1
resolulion des l:quaLions algebriqucs au trsnscendantes, ctc, ...

Definition of the residue of a function at a pole. 'Sur un nouveau genre de calcul

analogue au calcul infinitesimal', Exercices de M athematiques, 1, 1826, p. 11. Published
by permission of the Ecole Poly technique.
130 7. From the Theory of Singular Integrals to the Calculus of Residues

applications of the calculus of residues consisted of representing the singular

part of a [meromorphic] function f by the integral residue

(45)

Developing a notion that originated in the concept of the extraordinary

integral, he then investigated the whole part of the function J, that is,

CQ(x) = f(x) - £ «(f(z)))

x-z
and thus obtained a method for decomposing rational fractions into simple
fractions and later, more generally, for expanding functions into series (46).
Cauchy also obtained a simple proof of the residue theorem, first for
rectangular contours by the same procedure and shortly after this, for circular
contours (47). All calculations discussed were applied to functions for which
the number of poles increased infinitely with the length of the contour. In 1827,
he showed that it was necessary to consider in the calculations the integral
residue corresponding to a circle with center 0 and a radius that tends to
infinity. This he called the principal residue of the function (48).
Important areas of application of the calculus of residues were in the
research on the solutions of algebraic and transcendental equations as well as
on the solution of linear ordinary and partial differential equations with
constant coefficients and-in some cases-with variable coefficients. As we
saw earlier, it was within the context of the solutions of such equations that
Cauchy had created his new calculus. Furthermore, he generalized the formula
he had obtained in 1819 for the sum of similar functions 4J(u i ) of the roots Ui of
F(u) = 0 [see Eq. (7.11)] to the case of multiple roots, and presented this
generalized result in the form
£ 4J(u)F'(u)
L 4J(ui ) = «F(u))) .

in Exercices de Mathematiques (49). Likewise, he gave the general solution of

the linear homogeneous differential equation
d"y d"-ly d"-2 y dy
-+al--+a2--+···+a
dxn dxn - 1 dxn - 2 n
_l-+ay=O
dx n

by the formula

where F(r) is the characteristic equation and 4J(r) is an arbitrary function that
can be determined from the initial conditions (50). Finally, in several other
papers, Cauchy tried to apply the calculus of residues to linear partial
differential equations by combining it with the Fourier transform method.
 7. From the Theory of Singular Integrals to the Calculus of Residues 131

Thus, he was able to obtain solutions for some types of equations, especially
when a variable t could be separated from the other variables (51). In 1829 and
1830, Cauchy made use of this method in his theory of light (52).
In his research work on the application of the calculus of residues to
mathematical physics, Cauchy obtained representations offunctions by means
of residues in the form of series expansions, especially Fourier's series
expansion. He thought that he had proved the convergence of such series (53).
However, this was not the case, because in 1829 Dirichlet showed the
insufficiency of Cauchy's argument (54).
Aside from the calculus of residues and the first definition of a complex
function of a complex variable, which he gave in 1829 in his Le~ons sur Ie
Calcul Differentiel, Cauchy did not develop any new results in his theory of
functions until 1831. Indeed, during this period he seems to have been in a sort
of retreat from the position he had held in 1825; he made no use of integration
in the complex plane for arbitrary regions, but limited himself to rectangular
and circular areas. With the exception of Ostrogradski, no mathematicians
seem to have been interested in Cauchy's theory of functions during these
years. This lack of interest, however, was not due to any lack of effort on
Cauchy's part, because he certainly tried to stimulate an interest and
understanding of this theory by his publications and in his teaching at the
Ecole Polytechnique and at the College de France. But, for a considerable
time, his methods were considered as being too complicated, and it was only
after 1840 that the theory began to find acceptance in the French, German, and
Italian mathematical communities.
 Chapter 8

A Mathematician in the Congregation

Augustin-Louis Cauchy was now 28 years old and still living at his parents'
home, near the Palais du Luxembourg, where he had returned in 1812. Louis-
Fran~ois decided that it was now high time for his eldest son to get married,
since he now had a secure place in life. The elder Cauchy's choice of bride for
his son was the only daughter of the bookseller Marie-Jacques de Bure, Alolse
de Bure, 23 years of age. The de Bures were an old, solidly bourgeois family.
They had been in the book trade since the 17th century and were connected
with quite a few publishers, particularly with Didot and Saugrain. The two de
Bure brothers, Marie-Jacques and Jean-Jacques, had followed their father
Guillaume in the book trade in 1813 and were later associated with the King's
Library, where they compiled many catalogs. Serving in this capacity until
1838, they became well known as collectors of books and prints (1).
The marriage took place amid great pomp and ceremony on April 4, 1818,
at the Church of Saint-Sulpice in Paris. Aside from her personal effects, which
had an estimated value of 5000 francs, Alolse brought a considerable dowry to
the marriage: a perpetuity of 1300 francs; another perpetuity and income of
1000 francs, to be paid annually in cash by her parents; 2500 francs in silver;
and 10 shares of common stock in the Bank of France. In addition to the
foregoing, she brought a gift of 3000 francs from her uncle, Jean-Jacques de
Bure. Augustin-Louis brought a more modest financial contribution to the
marriage, including 15,000 francs that his parents had advanced as part of his
inheritance. He also brought 900 francs in state credits; 4000 francs in hard
cash, and his personal effects, including scientific instruments valued at some
3000 francs. In effect, almost this entire amount came from Augustin-Louis'
own earnings and thrift, which confirms the fact (as if it needed to be
confirmed) that during those days one would be unlikely to enter the field of
mathematics to make a fortune.
The marriage had been announced under the most favorable conditions,
with Louis XVIII and the entire royal family attesting the goodwill and esteem
they bore toward young Cauchy by signing the marriage contract (2). A
132
 8. A Mathematician in the Congregation 133

./t
r; ," 1'/.,,_ Si P... - lUI""""" 1...., fl ~"".9 ... v,.... ''''~ I'.;:;..c::; ,..I';
-f,.,~ !. /,~C;t. '" '".:- ~" "'AA'~__ '.... vir ay.l-. .S>..J._
lit... bl';"~ 9_ d!. .$.1,,,- 9~ ,,/J#..
./,
/ ;,. ~ (d1...1. &r..
-/rL,;..... tlu.:..- 9.. L tdJl-l/.~ 9~ 4,.

a . .
01'- ,...... J.,1~- '''' .1,. ~.t._. "'-' .~ ,,-1;1;:. 9-__ 11 t;"".."

9...L 1.1","- L j;"''7'~ '4i"'""" ..--t !itf'~ 9",,-r c.tt::-p.-

I

. ~-- /d- fl.f-/!;>- , ", .. q,-T' L ~ :.,.:.. vI/".;u..:, L

:iJ.i ..h.." ft..;..J , '/.... "',~ 9.;...,.. ., .,,,,3..,. 'x -ff'"fd:L ""-

I'~i;:{- ~ ",••~- c;J",-r / ... / '1.....,.9.. v....... ~ l'ii~

...9-::;.,/ ...J,( ', of "Iv- t*"-. :r- It._:r-

0. "44- "".<, df..l?- .I"':"'"
/, . 1/
v 10...;, !.. $,.it.;; .f...;.,;
~~jJ~ ....-t~
.;,ip...::r4 .. :cr

Petition by Augustin-Louis Cauchy to the Director General of the Ponts et Chaussees

to be licensed to marry Alolse de Bure. Archives Nationales, F14 2187 2 , Cauchy file.
Published by permission of the Ecole Poly technique.
134 8. A Mathematician in the Congregation

number of Peers of France attended the ceremonies, among them several

Congregationalists, such as the Count of Polignac, the Marquis of Semonville,
and the Viscount of Dambray. Also present were the Director of the Ecole
Poly technique; mathematicians, such as Laplace, Reynatld, Dinet, and Binet;
and several ecclesiastics, among whom were Teysseyrre, Genoude, Desjardins,
and Legris-Duval. The book trade was represented by Firmin-Didot and
others. A little pamphlet was published in celebration of the marriage (3). This
publication opened with an address by Augustin-Louis Cauchy to the
Duchess of Angouleme requesting that she and her husband attend the
wedding ceremonies. It contained some verses written in honor of the young
couple by Louis-Fran«ois and Alexandre Cauchy, Pierre Didot, and Charles
Magnin and a poem Augustin-Louis addressed to his bride. This last poem is
the sole remaining testimony on how he felt toward Alolse. It reads:

If ever there was a blessed moment in my life,

A happy day
It was most assuredly, my sweet friend,
The day you gave your love to me.
God Himself, has just blessed
The marriage that crowns all my desires.
His blessed law wills that I should love you
Ah! It is but a command that I am happy to obey.

And I will love you, my beloved friend,

Right to the end of my days;
But even then there is another life
Since your Louis will love you forever.

According to Valson, 'Cauchy's first years of married life were smooth,

untroubled by any rancors and problems' (4). The couple lived in the Hotel de
Bure in the rue Serpente, No.7. The house belonged to Marie-Jacques and
Jean-Jacques de Bure. Starting in 1822, Augustin-Louis habitually spent
summers in Sceaux, near Paris, at the Maison Trudon, a peaceful, shaded
property that had been purchased by Marguerite de Bure, Alolse's
grandmother.
In 1819, the couple's first daughter, Marie Fran«oise Alicia, was born; and,
in 1823, the second and last daughter, Marie Mathilde, was born. In any event,
it appears that Augustin-Louis Cauchy accorded his wife and two children no
important place in his life: his scientific work and other preoccupations took
all of his time. In fact, we will see that in 1830 Cauchy did not hesitate to leave
his family for several years and go into voluntary exile. It is, of course, a very
dangerous thing to speculate on the motives that compel the behavior of so
complex a person as Augustin-Louis Cauchy, particularly when so little
relevant evidence is available. Yet, it would seem highly likely that Cauchy's
marriage was a marriage of convenience, a union entered into in order to satisfy
his parents. Since early youth, his sensitive and passionate nature had been
 8. A Mathematician in the Congregation 135

strained to the utmost by his somewhat arrogant und uncompromising search

for truth. In short, then, domestic life was simply a bother to him; it was stifling
and placed an intolerable strain on his fragile nerves.
In August 1826, Cauchy contemplated taking a month's vacation in the
Pyrenees Mountains for his health. Lamennais, who also suffered from a
nervous condition would also go with him on the trip (5). In any event, the
vacation never took place. As the years passed, Cauchy's condition did not
improve (6). However, he refused to rest, to slacken his pace, but rather
continued his regular participation in meetings and working sessions at the
Academie. In addition to this, he produced a prodigious amount of research-
treatises, studies, articles, and mathematical studies.
Although he lived with his in-laws in the rue Serpente, he was still a regular
visitor at the Palais du Luxembourg, where he spent time with his parents
and his sisters and brothers. Until his retirement in 1825, Louis-Franyois
continued to support his eldest son, advancing his career whenever possible.
Through his influence, Augustin-Louis was awarded the Legion d'Honneur in
August 1819 and an appointment as engineer-ordinary first class in March
1820 and then as chief engineer second class in May 1825 (7).
In spite of the number and variety of mathematical and scientific works
he produced between 1816 and 1830-a list that, aside from the many
unpublished courses he taught at the Ecole Poly technique, the College de
France, and the Faculte des Sciences, included 92 papers read, presented, or
registered at the Academie, 41 articles published in scientific reviews, and 10
longer published works-he devoted a significant part of his time to other
activities. In particular, he was a regular participant in the various charitable
works founded by the Congregation, which he had entered in 1808, later being
joined in this association by his two brothers, Alexandre and Eugene.
The Congregation had officially begun to hold meetings at the start of the
Restoration in 1814. It was managed now by a Jesuit, Father Legris-Duval.
Taking full advantage of the new political climate, it prospered under the
protection of the authorities. Little by little, the Congregation increased its
activities and spheres of operation through its affiliates. These were specialized
organizations that the Congregation created to take care of particular areas of
concern. The first affiliate was the Societe des Bonnes Oeuvres, which was
established in 1816. This organization's area of concern basically consisted of
three charitable practices: visiting hospitals, visiting prisons, and instructing
young chimney sweeps in the catechism. Cauchy was a member of a
commission that, in April 1816, was charged by Father Legris-Duval with the
responsibility of forming the Societe. In this capacity, he served alongside the
extremely conservative leading figures, such as Mathieu de Montmorency and
Alexis de Noailles (8). He worked tirelessly. In fact, Lamennais, to whom
Cauchy was closely bound, stated in a letter written in July 1818 that he
(Cauchy) was absorbed from 'dawn to dusk' with charitable works (9).
Cauchy's close association with Lamennais had no doubt come by way of his
(Cauchy's) prior friendship with Teysseyre.
136 8. A Mathematician in the Congregation

Later on, he also worked with the Societe Catholique des Bons Livres, an
organization that had been created in Paris in August 1824. Its goal was to
finance the publication of good books, to propagate their message, and to
distribute them as cheaply as possible. The Duke Mathieu de Montmorency
was head ofthis organization until he died, when the Duke de Riviere replaced
him. Cauchy was one of its five directors. He served in this capacity along with
the Abbe Perreau, Grand Almoner of the Chevaliers de la foi; the Abbe
Dufriche-Desgenettes, Curator of the Church for Foreign Missions; the Abbe
de Salinis, a close relative of Lamennais; and Pierre Sebastien Laurentie, an
ultraroyalist journalist on La Quotidienne. In July 1827, a serious dispute
broke out between the Societe and the Grand Master of the Universite,
Monseigneur Frayssinous.
At the behest of Laurentie, the Societe Catholique des Bons Livres planned
to create and publish the Catholic Encyclopedia of the Sciences, a major work
that was to have extreme conservative leanings. Abbe Clausel, a member of the
Royal Committee on Public Instruction and an ardent Gallican, prompted by
Frayssinous, violently attacked the encyclopedia project and the Societe itself
in three very sharply worded pamphlets. Clausel laid particular blame on
Laurentie, whom he regarded as the great enemy of the Universite and of
censorship. Cauchy, Salinis, Perreau, and Laurentie then handed in their
resignations as directors of the Societe to the Duke of Riviere, as an expression
of protest against the publication of the pamphlets. Following various
negotiations and compromises, Laurentie was forced to give up his post as a
director; the other directors, including Cauchy, resumed their duties. Follow-
ing this dispute, the Societe lost its ministerial protection, and as a result, it
slowly declined; finally, in August 1830, it was disbanded altogether (10).
In the meantime, Cauchy became further involved in the extreme right-
wing battle that was now raging between the Jesuits and the Universite, with
Cauchy on the side of the Jesuits and against the Universite monopoly. In
order to understand the nature of this dispute, it is necessary to briefly recall a
few points relative to the evolution of the political climate in France during the
years from 1821 to 1828 and the role Cauchy played at the Academie des
Sciences on the side of the clerics.
Ever since the assassination of the Duc de Berry, eldest son of the future
Charles X, in February 1821, the government had been breaking away from
the more liberal political policies that it had followed during the preceding
period and pursued an increasingly reactionary road. At the end of the reign
of Louis XVIII, the Prime Minister VilleIe instituted the repression against the
Charbonnerie (a liberal secret association) and used all means at his disposal
to reduce the influence of the liberal parliamentarians in the two houses.
Charles X came to the throne in 1824, a few months after the election (skillfully
prepared by Villele) of a Chambre retrouvee with the same leanings as the
famous Chambre introuvable of 1815.
The ascension of Charles X further strengthened the reaction. Measures
were instituted in 1825 in favor of the landed aristocracy (the Milliard for
 8. A Mathematician in the Congregation 137

Cauchy as young academician. Half-length portrait by Boilly, 1821. Published by

permission of the Ecole Poly technique.

the indemnification of the emigres). The reestablishment of birthrights was

projected, but it was rejected by the Chamber of Peers in 1826. As for the
bourgeoisie, they saw their interests threatened, in 1824, by a scheme for
the conversion or funding of stocks, and again in 1827, by the dissolution
of the National Guard in Paris. The guard was guilty of having protested,
during a review, against the notorious 'Law of Justice and Love,' which
curtailed the freedom of the press.
More than anything else, however, it was the so-called 'alliance between the
throne and the altar' that aroused deep-seated passions in France during this
time. Influenced by the ideas of Maistre, Bonald, and Lamennais, a large part
of the French right regarded the Great Revolution of 1789 as nothing less than
'Satan's handiwork', in the fullest sense of the term. Therefore, in order to
reestablish political order and the traditional civil arrangements-both of
which were undermined by the catastrophe of 1789-it was necessary to
support religion. Bit by bit, a faction of the counterrevolutionary groups
longed for a kind of theocratic purification (11). The king, a very pious man,
138 8. A Mathematician in the Congregation

wanted to see the reestablishment of the prerogatives that the Catholic religion
had enjoyed prior to the Great Revolution. The law on sacrilegious persons,
the law on religious communities for women (which, though instituted merely
by royal ordinance, had the weight of a legal statute), and the solemn
coronation at Rheims all seemed, in 1825, to foretell a 'government by priests'.
In the provinces, the various missions that had been organized by the many
congregations and were supported by the civil authorities redoubled their
efforts to lead the masses back to the cult and faith of Catholicism. Augustin-
Louis Cauchy fully approved of these policies and echoed this conservative
religious mood in the Academie, where, on several occasions, he did not
hesitate to raise objections and denounce doctrines that were contrary to
religion. This behavior, of course, exasperated and angered the Academie's
liberal wing.
On July 19, 1824, he presented a note pertaining to Geoffroy Saint-Hilaire's
verbal report on Serres' work entitled 'Anatomie comparee du cerveau dans les
quatre classes d'animaux vertebres'. Going outside the framework of a purely
scientific debate and inquiry, he severely condemned Dr. Gall's 'very eminent
philosophical principle, a principle that rejects both the true philosophy and
the vital doctrines on which rest the peace and well being of society' (12). This
action by Cauchy, however, met with sharp and determined resistance from
the Academie. In his note, he pointed out that the Academie had itself never
placed Gall on its list of candidates for the medicine and anatomy sections
because of his theory of cerebral protuberances. The President of the
Academie intervened in the dispute, forcing Cauchy to withdraw this intrusive
observation (13).
A short time later, at the Academie's session of October 4, 1824, Cauchy
gave the evaluative report on a study of Souton about the theory of light. In
this report, he reproached Souton for having erroneously asserted (following
Voltaire) that Newton doubted the existence of the soul. Cauchy took
advantage of this opportunity to attack Voltaire and to reaffirm the
superiority ofthe Christian religion (14). This was sufficient to set off whispers
and murmurs among the academicians and to ignite a sharp polemic in the
press. In the October 6, 1824, issue of Le Corsaire, a liberal periodical, there
appeared an article in which was denounced 'this little discourse that rather
more resembles a homily than a scientific report'. In a bantering tone, the
journalist continued:

Now, it was certainly a curious thing to see an academician who seemed

to fulfill the respectable functions of a missionary preaching to the
heathens. It is not a novelty, except when it happens at the Academie, to
hear a geometer busily trying to prove-not by way of equations and
solidly logical arguments but by declamations that are both hackneyed
and completely out of place in an enlightened age-that materialism has
taken over the domains of science, a dubious proposition that ought not
to be difficult to disprove. Equally strange to hear was that the names
 8. A Mathematician in the Congregation 139

of Voltaire and other philosophers are found mentioned-by what

standards we do not know-in this peculiar little sermon, which as a
sermon, ended up very amusing indeed.
Le Memorial, a Catholic publication, replied to Le Corsaire's 'indecent article'
by citing Cauchy's outstanding qualities:
When he was no older than the age at which most youths have scarcely
completed their elementary school training, he was called to take his seat
in the most illustrious academy in the kingdom, a singular distinction
that was due to nothing more than the brilliance of a first-class mind, a
brilliance that outstripped his years. That much so, one should not pay
much mind to the insipid carpings of a literary corsaire (15).
In the meantime, Cauchy had earned a reputation in liberal circles as an
inquisitor in the breast of the Academie.
In the June 1, 1825, issue ofthe New Monthly Magazine, Stendhal captured
the perceived personality of the controversial savant in a definitive phrase,
writing:
Nowadays, M. Cauchy ofthe Institut, a veritable Jesuit in short frock, is
charged with the honorable task of ferreting out and bottling up
physiology, a science that of late has been advanced so far smartly
by the experiments of MM. Flourens, Magendie, and Edwards (16).
A short time later, in the November 1826 issue of the same publication,
Stendhal reported to his English-language readers an incident that befell the
Academie:

Some time ago, a naturalist whose name I will not mention out offear of
doing him injury, read a paper on the various phenomena that can be
observed in the life of certain insects. Of fundamental importance in its
own right, the subject was here treated in a very witty manner. At the end
of the reading, whispers and murmurs of approbation could be heard,
whereupon M. Cauchy took the floor and made the remark that the
Academy should not honor this curious little discussion of animal life by
applauding it. Said M. Cauchy: 'Even if it were admitted that the things
just told us were true-in my opinion they are false-it would still not
be useful to communicate such truths to the masses, given the devilish
state into which our misbegotten Revolution has hurled public opinion.
Any such talk can only harm our holy religion. They clearly show the
influence of physical causes and tend to confirm Cabanis' mischievous
doctrines'. These words from Mr. Cauchy, who is himself a dignitary of
the Academy of Sciences, were greeted with great bursts of laughter.
Although this little anecdote, which combines the ridiculous with the
downright odious, was particularly stinging, an examination of the minutes of
the Academie's meetings shows this incident was merely a clever, malicious
140 8. A Mathematician in the Congregation

invention that was doubtlessly inspired by Cauchy's actions against Gall's

system. Using the same ironic tone, Stendhal kept up his attacks on the
scientist, writing:
This courageous man, who is so enterprising and who apparently wants
to be a martyr to contempt, requested at the meeting on Monday last
that the Academy remove from its library all books that were tainted
with philosophical leanings. Up to now, the Academy has not dared
reply to M. Cauchy's request. Further to the point, several men of
science, whose livelihoods depend on salaries from some insignificant
employment, have been obliged to refuse to publish their most recent
discoveries and findings in physiology out of fear that they might be
accused of throwing some new light on the interplay between man's
physical and moral aspects' (17).
This was all baseless gossip, pure and simple, and Cauchy paid no attention
whatsoever to these malicious criticisms. In Dupin's Observations sur les
Recherches Statistiques sur ['Instruction et sur la Moralite Comparee des
Differents Departements de la France, which was presented on January 15,
1827, he found arguments that allowed him to set forth once more a
justification for the Church's role in education, particularly that of the Freres
des Ecoles Chretiennes in Brittany (18).
Such clericalism-and Cauchy was the champion of clericalism at the
Academie-naturally angered the liberal movement; upset the bourgeoisie,
which was still largely attached to the principles of 1789; and caused the lower
classes-particularly in the town and cities-to become disaffected. The issue
was clear, and the stakes were high: would the clergy go against the
educational establishment, the press, and the legislature in order to reassert,
with the support of the State, its ideological control over society? Oddly
enough, the liberals found allies in the ranks of the extreme royalists, who were
attached to Gallican traditions and were hostile to the ever-increasing role
played by the congregations that acted outside of any control by the Church
of France. As was seen in the crisis that shook the Societe des Bonnes Oeuvres
in 1827, the advocates of Gallican liberties had powerful supporters in the
government.
The publication of a pamphlet against the Jesuits by the Comte de
Montlosier, an old reactionary royalist with strong Gallican leanings, in
February 1826, was the signal for a vast anticlerical campaign. This time
around, the liberals were allied with the Gallicans in denouncing the secret
role of the 'priestly party', whose driving force was the Jesuits, whose tool
was the Congregation, whose doctrine was ultramontanism, and whose
ultimate aim was nothing less than the establishment of a theocracy. The
liberal press, the deputies of the opposition parties, and many sensationalist
pamphlets all joined forces to popularize the idea of a clerical plot and
clamored for the dissolution of the Congregation, as well as the dispersion of
the Jesuits.
 8. A Mathematician in the Congregation 141

In spite of the official illegality of the noncertified Jesuit residences, the

Minister of Ecclesiastical Affairs, Monseigneur Frayssinous, who had
nevertheless become suspect in the eyes of the ultramontains, defended the
Society of Jesus. Thus, the Jesuits were not only able to continue as before, but
even to expand their activities. For example, in 1826, there were 476 Jesuits in
France, as well as 2 noviciates and 8 small seminaries that the bishops had
entrusted to the order. In reality, these seminaries were colleges, the most
famous being that of Saint-Acheul. The Jesuits' power was extended by the
Congregation on which they exerted an unquestionable influence. At this time,
some 1200 persons, many of whom occupied leading positions in society,
belonged to the Congregation (19).
Meanwhile, Villele's government, increasingly isolated by its reactionary
clerical policies, was seeking to curb the opposition groups, especially in the
press, by applying repressive administrative measures and police tactics, all of
which were unpopular and ineffective. Moreover, Villele's enemies-that is,
the liberals and the Gallicans-expected to win the next elections that
were announced November 1827, because the enfranchised classes were
busily disassociating themselves from the extreme conservative party and
its unrealistic reactionary policy. In order to calm the liberals, Martignac, the
new Chief of the Ministry, obtained two ordinances from the King in June
1828. According to these ordinances, all church schools were placed directly
under the control of the Universite and members of the unauthorized
congregations were prohibited from teaching. In effect, these regulations
amounted to a ban on the Jesuits, and they, quietly submitting to the royal
ordinances, dispersed.
A few days before this decision was made, the directors of the Societe des
Bons Livres, among whom was Cauchy, came together and founded the
Association pour la Protection de la Religion Catholique. The aim of this
organization was 'to unite the efforts of all persons of good will in order to
defend the Catholic religion'. The association consisted of 45 members and a
board of directors, of which Cauchy was one of the members. The real aim of
the leaders of this organization was twofold: to establish a true ultramontain
party that would protect the interests of the Church and to organize Catholic
and antiliberal opinion by means of propaganda. The Gallicans regarded this
undertaking with horror (20). By merely participating in the creation of such
an organization, Cauchy became even more involved in politics. At the
Academie, in the company of the scholars with whom he associated, he
attempted to recruit others for the cause: Jacques Binet joined the association,
and Andre-Marie Ampere agreed to sponsor, along with Cauchy, preparatory
classes at the College de Juilly, an institution that had been established for the
benefit of students who had been attending the Jesuit secondary schools that
were closed in 1828. The Almones of the College de Juilly, the Abbe de Salinis,
was a director of the Association (21).
In March 1828, the Association began publication of a weekly, Le
Correspondant, Journal Religieux, Politique. Philosophique, with Bailly as
142 8. A Mathematician in the Congregation

editor. Lamennais was hostile to this publication, which was 'plowing the
same field' as the Memorial Catholique, the paper that he had founded.
However, as the Abbe de Salinis explained to him in a letter dated February 25,
1829, 'Le Correspondant was an idee fixe with the majority of the Association'
(22). In truth, however, Lamennais was disappointed by the Association's
timidity. Renouncing the theocratic ideal he had adhered to in the preceding
period, he published a work in early 1829 entitled Des Progres de la Revolution
et de la Guerre contre l'Eglise, in which he associated democracy and liberty
with the progress of Catholicism. However, notions such as these could not
satisfy reactionary ultramontains such as Cauchy. In his letter to Lamennais,
the Abbe Salinis observed:
Cauchy and the Vicar of Missions [Dufriche-DesgenettesJ, who is
basically a fine man, approve of the things you are doing to the extent
that they understand them.
But Salinis was an optimist, because their degree of comprehension was, in
truth, very scant. In fact, the Association so was paralyzed by its own internal
divisions that, on balance, its actual accomplishments were quite small: a few
brochures in 1828 and the journal that vacillated between Lamennais' and the
congregationalists' positions right down to the Revolution of July 1830.
While condemning liberal agitation, Le Correspondant watched the policies
followed by Polignac during the final months of the Restoration with
increasing alarm and assailed the Revolution of July 1830 as an all too
foreseeable misfortune. Adopting 'civil and religious liberty' as its motto,
it continued publication after the Revolution, disappearing only when
Lamennais began publication of his new journal, LAvenir.
Taking advantage of the discomfort of the reactionary congregationalists,
in November 1830, Lamennais, in conjunction with Montalembert, founded
a new association, the Agence Generale pour la Defense de la Religion
Catholique. This organization, which replaced the Association pour la
Protection de la Religion Catholique, can be regarded as the first liberal
Catholic organization in France.
Augustin-Louis Cauchy, of course, never changed his views as regards
liberal Catholicism and throughout his life remained deeply attached to the
reactionary clerical ideals of the Congregation. Deeply involved in the
antiliberal struggle, he regarded the events of 1830 not only as a national
catastrophe, but also as a personal drama.
 Chapter 9

Exile in Turin

The Revolution of 1830 marked a turning point in Cauchy's life and a break in
his mathematical productivity. On August 2, 1830, Charles X abdicated in
favor of his grandson, the Duke of Bordeaux. One month later, Cauchy left
Paris and France and went into voluntary exile, which would last eight years.
His exile was a consequence of the events that took place during the famous
July days.
The Revolution of 1830 had its origins in the crisis between the royal
power and the liberal movement, a crisis that divided the propertied classes
and shook the country for several years. In August 1829, a right-wing cabinet
was formed with La Bourdonnaye as the Minister ofthe Interior and Polignac
as the Minister of Foreign Affairs. This cabinet represented a veritable
provocation to the liberal majority in the Chamber. One year later, on July
26, 1830, following the liberals' victory in the elections, a set of official
proclamations appeared in Le Moniteur suspending the freedom of the press;
announcing the dissolution of the Chamber, which had just been elected;
and modifying the election laws. This was nothing less than a coup d'etat,
and its effect was to provoke Paris into insurrection. On July 27, the first con-
frontations took place between the insurgents and the royal troops under the
command of Marmont. For three days, the 'Glorious Three', Paris bristled
with barricades as the insurrection gained strength. On July 29, the Tuileries
fell to the insurgents, and Marmont was forced to evacuate Paris, leaving
the capital in the hands of the revolution. The wealthy liberal bourgeoisie
and its leaders Laffitte, Thiers, and La Fayette now brought the Duke of
Orleans forward. The deputies had appointed him Lieutenant General du
Royaume on July 31 and King of the French on August 7. During these
hectic days, Charles X fled from Paris and went to Cherbourg where he and
his family embarked for England. Cauchy watched these events unfold from
close range. During the weeks immediately preceding the Revolution of 1830
143
144 9. Exile in Turin

his antiliberal hatred seems to have reached unparalleled heights. On July

13, the day on which the election of deputies for the Department of the Seine
was held, he provoked an incident at the Sorbonne. The incident was reported
that same day by the Journal du Commerce in the following terms:

M. Cauchy, a member of the Academie des Sciences, is an elector and

votes in the 2nd Section of the 7th College, which meets at the Sorbonne.
This morning, when M. Cauchy was called to deposit his paper ballot, he
found two lists on the desk, one consisting of the names ofthe members
of the provisional office and the other the list of candidates for final office.
M. Cauchy was quite loud in expressing his indignation at seeing a list of
liberal candidates posted up right in front of the royalist electors. He
seized the list and crumpled it up in his hand.
Premier President Seguier, who was standing near M. Cauchy, spoke
sharply to him and reminded him that just as he was free to vote as he
saw fit, he must leave to the other electors the possibility of correctly
writing the names of the candidates of their choice. But M. Cauchy,
paying no mind to M. Seguier's observation, continued to speak and
gesticulate vehemently as he crumpled the list even more. M. Cropelet,
the President of the 2nd Section, had to call M. Cauchy to order and
request that he put the list back on the desk and withdraw as soon as he
had cast his vote.

A few days later, Paris was gripped by insurrection. Barricades sprang up in

the Latin Quarter. In the Place Saint-Michel, a few steps away from the rue
Serpente where Cauchy lived, there was a violent confrontation on July 28,
1830. Students from the Ecole Poly technique took an active part in the
uprising. Indeed, on July 28, 1830, the poly technicians left the Ecole, where
they were billeted and, in uniform, took command of small bands of
insurgents. Their action was decisive everywhere, especially in the storming of
the Babylon Barracks, during which the polytechnician Vanneau was killed.
Cauchy was deeply shaken by the collapse of the government that he had so
ardently supported. He did not attend a single session at the Academie
between July 26 and August 30. Never before-not since his appointment in
1816-had he been absent for so long a time (1). On August 30, 1830, he
attended the Academie's weekly meetings for the last time, and during the first
days of September, he left Paris to go abroad, taking neither wife nor children
with him (2).
What reasons could have compelled Cauchy to abandon everything, his
family and his position, during the summer of 1830? Although definitive
evidence is lacking, we can nevertheless advance a likely hypothesis. First,
however, it should be noted that the most common explanation for his
behavior does not fit in very well with the chronological order of the facts:
 9. Exile in Turin 145

Poly technicians climb over the wall on July 28, 1830 and rejoin the insurgents.
Published by permission of the Ecole Poly technique.

Cauchy would have refused to swear allegiance to the new political

arrangement in France out ofloyalty to the now deposed king, he would have
been faced with immediate loss, by forfeiture, of all the positions he had come
to occupy, and in consequence he had to decide to leave France (3).
As a matter of fact, the law requiring the oath of loyalty to the King of
the French from all persons occupying a position in the public service was
not passed until August 30, 1830, after Cauchy had made his decision to
leave the country. Penalties against him by the administrations of the various
institutions with which he was affiliated-the Ponts et Chaussees, the Ecole
Poly technique, and the Faculte des Sciences of Paris - were imposed several
days after his actual departure (4). Thus, Cauchy did not go abroad after
having lost his positions for refusing to take the oath, as Valson claims. A
careful examination of the available facts and documents suggests other
explanations. In truth, when Cauchy left Paris at the beginning of September
1830, he probably had no idea that he was now embarking on a long exile. But,
several compelling reasons required that he leave the capital for a time.
For quite a while, Cauchy had been considering going on a trip in order to
rest. As we have already seen, he had planned on traveling to the Pyrenees in
1826, but had been unable to do so at that time. His exceptional creative
146 9. Exile in Turin

activity during the years prior to the Revolution of 1830 had left him physically
and emotionally drained. Indeed, during the two years immediately preceding
the turmoils of July 1830, he seems to have even accelerated the 'furious pace of
his creative work' (5). In 1828 he published the last installment of Volume II of
Exercices de M athematiques as well as the 12 installments of Volume III.
During the same year, he also published the second volume of Lefons sur les
Applications du Calcul Infinitesimal ala Geometrie and a paper in the Bulletin
des Sciences of Ferussac. Moreover, he presented 10 papers to the Academie
during this same period. The following year, 1829, he published the first 8
installments of Volume IV of Exercices de Mathematiques, the Lefons sur Ie
Calcul Differentiel, as well as 8 articles in the Bulletin des Sciences of Ferussac.
Aside from this, he presented 15 papers to the Academie. Finally, during the
first 8 months of 1830, he published the last 4 installments of Volume IV of
Exercices de M athematiques along with 3 installments of Volume V, 4 articles
in the Bulletin des Sciences of Ferrussac, and 2 studies on the theory oflight, in
addition to presenting 12 papers to the Academie. While it is certainly true that
none of the works were produced during the 2 years immediately preceding the
Revolution of 1830, they can be counted among his great masterpieces. In
terms of sheer quantity, his productivity was impressive, to say the least.
Referring to this period, Cauchy declared in 1831, in a letter to his family, that:

I dare not work any longer in the evenings, as I was once silly enough to
do over a period of time, in order to preserve the good health I am now
enjoying. In this regard, I have gained much, because it is certain that
when I left Paris, I felt so weak and exhausted that I had started to
believe that I could not very well continue keeping the same pace (6).

Intellectually exhausted and physically weak, he had to revive his strength far
from Paris. Cauchy's assertions in this matter are confirmed by letters from his
brother and from his wife, written in September, October, and November
1830. On September 11, 1830, Alexandre Cauchy wrote to the Director
General of the Ponts et Chaussees, explaining Augustin-Louis' absence:

For the past two or so weeks, my brother, not being scheduled for any
active duties at the Ponts et Chaussees, has used the leisure time afforded
by the recess periods at the Ecole Polytechnique and at the Faculte des
Sciences to take the short trip to Switzerland and Italy that his health has
so long demanded (7).

Alexandre Cauchy thus hoped to soothe and put off the Ponts et Chaussees
administration, which was now pressing for Cauchy to take the oath of
allegiance.
In other letters, letters written right up to the end of November 1830,
Augustin-Louis' family continued to pretend against all evidence that his poor
state of health necessitated the extension of his trip and that this necessity was,
 9. Exile in Turin 147

Augustin-Louis Cauchy. Painting by J. Roller, undated. Published by

permission of the Ecole Poly technique.

in fact, the only reason for his absence (8). But, if this was a plausible
explanation in September, it became a poor excuse as time passed, and the
Cauchy family was unable to keep Cauchy from losing all his public positions:
on November 26, 1830, he lost his adjoint professorship at the Faculte des
Sciences, in February 1831 he was stripped of his professorship at the Ecole
Poly technique, and in early March of that year, he lost his rank as an engineer
in the Ponts et Chaussees.
The extension of his leave of absence beyond the month of September
indicates that Cauchy's decision to leave France was not based solely on the
need for rest to improve his health, but that his decision was also the direct
consequence of the July Revolution. Cauchy, his physical and emotional
system already weakened by overwork, simply could not bear the shock of
events brought on by the political turmoil that came to a head in July.
Passionate to the point of blindness, with an emotional sensitivity that was
hidden behind his outward rigidity and coldness, but which often betrayed
itself, as in his tendency to stammer and gesticulate whenever he lost control,
and traumatized since childhood by the very idea of revolution, he resolved at
the beginning of August to leave Paris, to depart with no more of a definitive
plan in mind than simply to avoid turmoil and agitation.
148 9. Exile in Turin

Two facts, no doubt, have a definite bearing on his decision: the massive
participation of his students from the Ecole Poly technique in the revolution
and the violent anticlericalism, whose main victims were the Jesuits, his
friends. On July 28, 1830, the Jesuit residence in Montrouge was sacked and
plundered, and it is very likely that Cauchy sheltered in his home Jesuits who
were attempting to flee from the popular wrath. It seems even more likely that
he accompanied some of them to Switzerland as they made their way into
exile (9).
The exact route that he followed on his departure from France is not
known, but he seems to have gone directly to Fribourg, Switzerland (10). A
small colony of emigres were settled there, having been drawn to this
particular location because of its proximity to the French frontier and the
friendly protection of the canton's patrician government. The Jesuits were very
influential there and operated a college that, since 1827, had welcomed
numerous French students, who were lodged in a private boarding school
there. Cauchy did not remain in Fribourg long, but rather continued on to
northern Italy. By early October, he had reached Turin, and by the end of the
month he was in Modena (11). During the last day of November, he reached
Genoa, where he met the King of Sardinia. By now, however, his trip abroad
had already become an exile.
In effect, at the end of September 1830, while he was still in Fribourg,
Cauchy had to decide whether he would take the oath of allegiance.
Supporters of the now-dethroned Charles X were divided on whether or not
the oath should be taken, with a minority of the royalists regarding themselves
as bound to their sovereign by a feudal code of honor; to take the oath would,
in their eyes, be to commit a felony. This was the way Cauchy saw the matter,
and disregarding the advice of his family, who continued to hope that he would
return, he finally rejected the idea of returning to France and decided to remain
abroad, where he expected to be able to continue teaching.
Cauchy's father, Louis-Fran<;ois, and his two younger brothers, Alexandre
and Eugene, who had succeeded their father in the Chamber of Peers in 1825,
took the oath of allegiance to the new regime and, having done so, served the
new government with all the zeal that they had shown in serving its
predecessor. Moreover, this latest political contortion, so adeptly performed
by the Cauchys, father and sons, earned them an honorable mention in the
Dictionnaire des Girouettes, a satirical work that appeared in 1832 and gave
the following definition:
Cauchy-Honorary Keeper of Archives in the Chamber of Peers. An
official with the Intendancy of Rouen before the great Revolution.
Keeper of the Archives and of the Seal of the Senate during the
Consulate. Secretary to the Curator of Archives of the Senate under the
Empire. Keeper of the archives and Editor of the Verbal Proceedings in
the Chamber of Peers during the Restoration. Honorary Keeper of the
Archives in the Chamber of Peers before, during, and after the July
 9. Exile in Turin 149

Revolution. His son Alexandre, Counselor to the Royal Court and

Titular Keeper, and his son Eugene, Adjoint Keeper. These three are very
cosy at the Palais du Luxembourg ... Napoleon made him a knight of
the Legion of Honor ... Louis XVIII made him an officer in the same
order. Charles X, whose praises he sang, advanced his [Cauchy's]
children. He has not yet sung the praises of Louis-Philippe. Let's wait a
while! (12)
The Cauchy family, of course, fully shared Augustin-Louis' political views, but
while the scientist of the family held to his convictions in a passionate,
uncompromising way, the other members of the family preferred to hold their
views with a prudent reserve.
In early October 1830, while he was still at Fribourg, Cauchy-with the
support of the Jesuits-threw himself into work on the Academie Helvetique
project. This project, in the words of the prospectus, which was written by
Cauchy, aimed to establish courses so that 'young people, who shall have
completed their studies, whether at the college of the city [Fribourg] or at
some other institution, may come and pursue instruction in the philosophical
and literary sciences, in oriental languages, and in the mathematical
sciences ... ' In the prospectus, Cauchy specified that courses at the Academie
Helvetique would be taught by 'scholars from France, who, after having
carefully considered the conditions under which the new laws in their own
country would allow them to retain their professorships, have resolved to
remain faithful to the oath they swore to Charles X' (13). This passage was
omitted when the prospectus was printed.
During the next few weeks, Cauchy devoted all of his energies to bringing
this project to a successful conclusion (14). A major problem facing him at this
point was that of finding the necessary financial means. In October 1830, he
wrote (and had printed) a prospectus that was to launch his fund-raising
efforts. In order to obtain the title of founder of the Academie Helvetique, it
was necessary to take one or several issues of stock worth a thousand francs
per annum. But, this would not bring in sufficient funds, and Cauchy tried to
get the financial support of Europe's most reactionary rulers: he wrote to both
the Emperor of Austria and the Csar of Russia, and he personally met the
Duke of Modena and the King of Sardinia, Carlo-Felice.
Cauchy also developed a working list of prospective teachers for the
Academie, a list of legitimist scholars who, like himself, had gone into exile.
Aside from himself, the list contained the names of Charles-Louis Haller,
former professor at the Ecole des Chartes; Joseph Recamier, a former
professor at the Ecole de Medecine of Paris and at the College de France;
Henri-Francois Gaultier de Claubry and Jacques Clarion, both formerly of the
College de Pharmacie; Charles-Francois Leroy and Camille Menjaud, from
the Ecole Poly technique staff; Meyraux, from the College de Bourbon; the
engineer Hippolyte d'Haranguier de Quincerot; and d'Horrer, a diplomat
(15). The project seemed to be on the verge of successful completion, the
150 9. Exile in Turin

financial problem having been straightened out and the necessary authoriz-
ation papers obtained when, on December 2, 1830, the 'Day of the Sticks' saw
the reactionary patrician regime in Fribourg overthrown and the liberals
swept into power. This spelled the end of the Academie, and the entire project
had to be abandoned.
The months that followed this setback, up until the Autumn of 1831, are one
of the least known periods in Cauchy's life. After his trip to Italy and the
abandoning of the Academie Helvetique project, Cauchy seems to have settled
in Switzerland (16). Events now unfolding in Paris kept him from returning
home: on February 14, 1831, a mob sacked the church of Saint-Germain
l'Auxerrois, and the following day, the Archbishop's palace was also sacked. A
brief trip back to Paris in early March convinced him that, despite the pleas of
his family, it would be better to remain abroad so as to avoid the turmoil in
Paris (17). Thus, he returned to Switzerland, again without his wife and two
children, whom he left with his parents-in-law. This separation would continue
until 1834, and we are forced to wonder if indeed Cauchy was not fleeing from
the responsibilities of family and marriage quite as much as from the
revolutionary turmoil ....
Be that as it may, Cauchy used these months of solitude to regain his
strength. In July, he informed the President of the Academie in Paris that the
'precarious state of his health, weakened as it was by intensive study, obliged
him to extend his leave of absence a while longer' (18). Cauchy's scientific
productivity had now been considerably reduced. During this time, from
December 1830 to June 1831, he published three articles in Italian entitled Sui
metodi analytici in the Biblioteca Italiana of Milan. In these three articles, he
presented an introduction to the methods of his courses at the Ecole
Poly technique (19).
The reason for the publication of these articles was a report by Giuseppe
Cossa that examined Cauchy's Exercices de Mathematiques in the Biblioteca
Italiana. In his report, Cossa advised readers of negligent errors and oversights
in Cauchy's presentations of the foundations of analysis and further observed
that the author of a treatise who fails to attain his end should at least have the
honesty to acknowledge the fact and not hide his own inabilities. In a note
appended to this report, Cossa specifically declared that he was not referring to
Cauchy. Nevertheless, Cauchy felt that he and his work were the subjects of
Cossa's report, and he wrote the three articles to show what he meant by "the
need for rigor" in mathematics (20). Given their didactic character, it might be
thought that they were the summary of a course from this period.
Cauchy soon went to Turin, where a small colony of legitimists were
thriving under the protection of the King of Sardinia, Carlo Alberto.
Unfortunately, it has not been possible to determine the exact date of his move,
however, it appears to have been sometime during the month of August. In any
event, the Academy of Sciences of Turin devoted its meeting of October 11,
1831, to an address given by Cauchy on his paper 'Sur la mecanique celeste et
 9. Exile in Turin 151

sur un nouveau calcul qui s'applique a un grand nombre de questions

diverses'. A short while later, on November 27, 1831, he presented a second
paper to the Academy, entitled 'Sur les rapports qui existent entre Ie calcul des
residus <:t Ie calcul des limites, et sur les avantages qu'offrent ces deux calculs
dans la resolution des equations algebriques ou transcendantes'. These two
studies represent Cauchy's real return to the world of mathematical research
after an almost total absence, which had lasted for a full year.
The paper presented on October 11 was a "composite" study (21). In the
introduction, Cauchy indicated the occasion for which he had undertaken this
study. The methods of calculation used in astronomy since Laplace first
presented them in his monumental M ecanique Celeste were based on series
expansions with convergences that had not been rigorously shown. Cauchy
had been aware of this problem for quite some time and, prior to 1830, had
begun to think about ways that would enable him to strengthen the analytical
foundations of mathematical astronomy. Thus, Cauchy's paper of October
1831, in its most significant aspects, was the result of research that had been
begun several years before the paper was written (22).
It is sometimes erroneously concluded from a remark made by Cauchy that
this paper was written following a conversation that he had with the
astronomer Jean Plana in Turin. Plana, of course, had spoken with Cauchy
several weeks before about the time needed for calculating the coefficients in
the series expansions related to the various perturbations, and Cauchy later
gave him certain formulas by which the numerical calculations could be
performed more simply and more quickly, he himself having already made use
of these formulas. Cauchy observed, however, that in order to obtain these
formulas, it sufficed merely to 'apply to the expansion of the function that, in
the Mecanique Celeste, is denoted by R certain well-known results, such as
Taylor's theorem and Lagrange's theorem on the expansion of the functions of
the roots of algebraic or transcendental equations'. In fact, by citing Plana,
Cauchy dealt tactfully with the most outstanding astronomer in Turin, whose
scientific monopoly could be threatened by the presence of a brilliant and
illustrious mathematician.
The most important part of the study dealt with the general principles of the
calculus of limits. First, Cauchy proved the integral formula

f(X)=~fh x !(x) dp (x=XepJ=!), (9.1)

2n -7t x-x
where f is a finite and continuous function and x is a complex value in the open
disc D with center 0 and radius X. He immediately obtained this formula,
known today as Cauchy's formula, by replacing the finite and continuous
function f(x) by
_f(x) - f(x)
x'-----
x-x
152 9. Exile in Turin

in the mean-value formula

f +"
_" f(x)dp = 2nf(0) (x = X eI';=t).

Cauchy noted that the mean-value theorem, as well as his integral formula,
can also be deduced from the residue theorem applied to the function
cP(x)
x
taken along a circle with a center 0 and radius X:

f_"+" cP(x)dp = 2n~£~:cP(x)

x
By this method, he had established his integral formula for the first time about
1820 (see Eq. 7.11).
Then, Cauchy expanded _ x
x-x
into the MacLaurin series f (;)n,
0 x
which is
convergent in D, and by permuting the signs Land S, he showed that any
function of a complex variable x has a convergent MacLaurin series
expansion, provided the modulus ~ of the variable maintains a value less than
the value for which the function ceases to be finite and continuous. This
statement is sometimes called the Turin theorem.
Cauchy also majorized the modulus of the nth term,

1
2n f:,," (~y f(x)dp,
of the series expansion of f(x) in Eq. (9.1) by (.!. Y/\ f{x) and the modulus of
the remainder of order n

-
1 f+" xnf(x)
dp
2n _" xn l(X - x)

by ~n ;
xn-
f(x)
(X-~)
(and also by /\ [ x?(x)
x n- (x-x)
J). The symbol /\ designates

the limit of the function, that is, the maximum of the modulus of the function
in its domain of definition; in particular, we have
/\ f(x) = supl f(X epJ=l).
p

These majorations are now called the Cauchy inequalities. Cauchy used
them in order to determine upper bounds of errors when evaluating functions
whether explicit or implicit by means of series expansion. This method, which
is known today as the method of majorants, was called by Cauchy the calculus
of limits.
 9. Exile in Turin 153

In his paper, Cauchy presented also a new theory on the variation of

arbitrary constants and, in a very extended second part, applications of this
theory and of the calculus of limits to celestial mechanics especially for the
expansion of the function R.
The second paper, which was presented on November 27, 1831, to the
Academy of Turin, remained almost unknown, in spite of the publication of
several abstracts, until its republication in 1974 (23). It is unquestionably one
of Cauchy's most beautiful studies, and less because of the importance of the
mathematical results it contains than because of the polish and elegance of its
proofs, which are based on the systematic use of integration in the complex
plane, which he had created some six years earlier. Moreover, this new study
marks a theoretical advance over his paper of 1825, since he now introduced
closed paths. For instance, he stated the residue theorem in the synthetic form

£ «(f(z))) =
2nJ=1
1 r J(z)dzds ds
Jc
(9.2)

where C is a closed path of length c, s is the curvilinear abscissa of C, and

£«(f(z))) is the integral residue of J inside C. Moreover, he gave the
majoration c/2n 1\ J(z) for the modulus of £«(f(z))).
By using Eq. (9.2), Cauchy resumed using the residue method he had
developed to investigate the roots of the equation J(z) = O. Thus, he obtained
the formula

I
i
F(Zi) = 1
2nJ=1
r J'(z) F(z) dzds ds
Jc J(z)
(9.3)

for the summation of the similar functions of the roots Zi of the equation lying
inside the contour C, to be compared with Eqs. (7.11) and (7.25); the modulus
of IF(Zi) is majorized by ~ 1\ [J'(Z) F(Z)]. For F(z) = 1 and F(z) = z,
2n J(z)
Eq. (9.3), yields, respectively,

m= 1 r J'(z)dz ds
2nJ=1 Jc J(z) ds '
where m is the number of roots inside C, and

Then, Cauchy developed a new calculus, called the calculus of indices of

functions. The index of the real function /(s) of a real variable relative to the
root (1 is the number
. 1{
1lm- /«(1 + e) J«(1- e) }
£-+0 2 J[I«(1 + e)]2 J[I«(1 - e)]2 '
154 9. Exile in Turin

which can take the values - 1, 0, or + 1. The integral index of f, denoted by

Cauchy's Y«(f(s))), is the sum of the indices of f relative to all of its roots;
Y~~ «(f(s))) is the sum of the indices of f relative to all of its roots between S1
and S2' The calculus of indices was based on the formula

1
2nj=1
f !'(Z)dZdS=YC((X(S)))
c f(z) ds 0 ¢(s)
(9.4)

where X(s) + J=1 ¢(s) is the complex function f(z) taken on the contour C of
length c. From this formula, Cauchy elegantly deduced many theorems on the
roots of algebraic equations, especially the Sturm theorem.
Moreover, using the calculus oflimits, he set forth methods for determining
the series expansion of a single root of an algebraic equation or of a sum of
such roots. A short time later, on September 10, 1832, he presented a study to
the Academy of Turin in which he exhibited a new method for expanding the
real and complex roots of an equation depending on one parameter and, in
certain cases, for determining a bound for the error that is committed by
neglecting a given number of terms in the series (24).
Cauchy was now far removed, physically, from the French mathematical
community, and it took him almost 10 years to make the results of his first
Turin paper known. As for the second Turin paper, Cauchy refused to publish
it at all, for he soon found new methods for attacking the problem it treated.
Well aware of the importance of the results he had obtained, results that not
only generalized his work in 1813 on the roots of equations, but also Fourier's
results and Sturm's famous theorem of 1829, which dealt with the number of
real roots (of an algebraic equation) between two given numbers, Cauchy
attempted to create a formal calculus of indices offunctions that was not based
on the calculus of residues (25).
Cauchy's arrival was an important event for the progress of the sciences in
Turin. The exiled French scientist had the support ofthe Jesuits, who were very
influential at court, and was thus well received in Turin. As has been seen, the
doors to the Academy were wide open to him; but even more important, he
was offered a professorial chair at the university so that he could resume his
work as a teacher. On December 19, 1831, Count Gloria, the President of
Studies, petitioned King Carlo Alberto to reestablish the chair in mathemat-
ical physics at the University of Turin, which the illustrious Avogadro had
formerly held. This chair had been eliminated by King Carlo Felice on account
of the events of 1821. Thus, on January 5,1832, the king established the chair in
physics for Cauchy (26), who was thus able to inform his family of his new
position on January 11, declaring:
At the request of the University of Turin, a chair in sublime physics [that
is to say, mathematical physics, which I taught at the College de France]
has been established, and I am appointed to teach this subject for a
beginning salary of 1000 ecus (27).
 9. Exile in Turin 155

On 15 January, after he had been appointed to the chair, Cauchy visited the
court in order to express his gratitude to his royal benefactor. While there, he
read a paper, concerning which King Carlo Alberto wrote that 'the views he
expressed seem, in my opinion, to be very wise, and I expect to reflect deeply on
the matter' (28).
We do not know exactly when Cauchy began teaching his courses, and it is
not even clear whether he taught during the winter of 1832. However, it is
known that during the month of March he went on a trip that took him to
Paris, among other places. While in his home city, Cauchy attended the
meetings of the Academie on March 5, March 12, and March 19. Taking
advantage of the fact that he was in Paris, he presented the two papers he had
written in 1831 to the Academie. He seems not to have headed directly back to
Turin, because, during the same month, he spent two weeks in Rome, where
he resided at the Hotel Cesario While in Rome, he met the mathematician
Tortolini (29) and was given the honor of an audience with Pope Gregory XVI.
After his return to Turin, Cauchy delayed committing himself to returning
to France, even though his family was urging him to do so. A month earlier, his
youngest brother Amedee had died; and, no doubt, his wife and children were
worrying about him. Moreover, his second brother, Eugene, soon arrived in
Turin as deputy for the family and asked him to hasten his return, which was so
heartily desired. Indeed,
Those reasons most likely to touch him were set forth. Aside from
considerations for his family, he was made aware of the fact that France
was now neither so tumultous nor so distressed as he assumed. It was
explained [to him] that the revolutionary movement had become
progressively more calm -and, moreover, by not letting himself get
involved in any kind of official position, he would surely be able to find
the peace and quiet necessary for his studies (30).
Unfortunately, just as he seemed to have reached the point of relenting and
returning home, new troubles broke out in Paris. The funeral of General
Lamarque, who had died of cholera, was the occasion for a republican uprising
on June 5 and 6,1832. The fighting, though brief, was very violent, and it left
many dead. The ensuing repression, which was a measure of the depth of the
great fear that had shaken the bourgeoisie, was extremely harsh, and Cauchy
quickly seized on these events and used them as a pretext for not returning
home (31).
Cauchy did return to Paris for a two-week stay in October 1832, before he
started teaching his courses at the University of Turin. On October 8, 1832,
Cauchy took part in the commission that was charged with selecting a
problem for the Grand Prix de Mathematiques of 1832, and the fact that he did
so suggests that he was then contemplating a longer stay in Paris. However, he
did not return to the Academie after October 29. Aside from presenting a paper
that had been published using lithography, a study entitled 'Sur la rectification
156 9. Exile in Turin

des courbes et la quadrature des surfaces courbes', dated October 19, 1832, he
had enough time to present a note entitled 'Sur Ie versement des voitures
publiques', in which, drawing on his own experiences as a traveler, he
advanced proposals for reducing the number of serious accidents (32). He
seems to have returned to Turin at the end of October 1832.
Once Cauchy had returned to Turin, his main concern was in teaching
his courses. This remained the case until his departure in July 1833. According
to Joseph Bertrand, Cauchy first planned to give his lectures in Latin, but gave
up on the idea and, instead, lectured in Italian, a language he had first studied
years before at the Ecole des Ponts et Chaussees. This substitution of Italian
for Latin was made on the basis of the belief that those attending his courses
could more clearly understand the former that the latter. Though he lectured
in Italian, the lecture notes and other material for his courses were no doubt in
French, as were his Sept Lec;ons de Physique Generale of 1833 (33) and the
Resumes Analytiques, which were published (thanks to a subsidy of 200 lira
that King Carlo Alberto granted on request from Cauchy) by the royal
printing service between 1833 and 1835 (34). The first of these works presented
(albeit in the very general language of natural philosophy) the principles of
molecular physics. The second work, which, according to the foreword, was a
composite 'of a series of articles aimed at presenting a resume of the most
important theorems in analysis, older results as well as newer ones, and
especially the theories that include algebraic analysis and those methods that
make for an easier exposition of it [algebraic analysis]', seemed rather like a
compromise between a miscellany along the lines of Exercices and a work like
his Analyse Algebrique of 1821.
The only account we have today bearing on Cauchy's performance as a
teacher in Turin is a rather unflattering picture by Louis-Frederic Menabrea.
The courses Cauchy taught were not required. However, attracted by his
tremendous reputation, many students enrolled in his classes. At that time,
education was carried out under fairly precarious conditions. The government
was afraid of student agitation and turmoil, and accordingly, students were
not even authorized to reside in Turin itself. In light of this restriction,
professors only gave private lessons (35). As to Cauchy's teaching, Menabrea
wrote:
[His courses] were very confused, skipping suddenly from one idea to
another, from one formula to the next, with no attempt to give a
connection between them. His presentations were obscure clouds,
illuminated from time to time by flashes of pure genius. The students
found them to be exhausting, and only a few were able to endure them to
the end. In fact, of the thirty who were enrolled in his course with me, I
was the only one to 'see it through' (36).
Thus, if Menabrea is to be believed, it is clear that Cauchy failed as a teacher in
Turin. This failure may, of course, be explained in terms of Cauchy's known
lack of teaching ability as well as in terms of covert meddling and obstructions
 9. Exile in Turin 157

by certain students who were close to Plana, the titular of the chair in analysis
at the University of Turin, for it is known that Plana had been hostile to
Cauchy since his arrival in Turin.
As we have seen, Cauchy made reference to Plana in his study of October
11, 1831. As he read the introductory remarks of his paper to the Academy of
Turin, Plana interrupted him at several different points without, however,
contesting the assertion that he had conversations with Cauchy relative to the
topic under investigation. Plana alleged that he had dealt with one of the
questions investigated in Cauchy's study in papers of his own, papers that he
had registered with the Academy in sealed envelopes on September 2 and 6,
1832. In any event, under questioning by Cauchy, Plana had to acknowledge
that he had not solved the problem of determining the general expansion of the
perturbation function (37). Moreover, according to Menabrea, a short time
later, Plana got the notion that Cauchy was attempting to obtain one of his
[Plana's] positions (38). Whether or not Cauchy did indeed seek to undermine
Plana's position is open to question; however, this affair illustrates the mistrust
that governed relations between the two men. Thus it was that Plana hardly
bothered to conceal his hostility toward the 'new hypertranscendent analysis'
his great contemporary had developed (39).
Cauchy had a much better relationship with Bidone, who years later
recalled 'with genuine pleasure' the talks he had on scientific matters with 'the
great scientist' (40). Finally, it should be noted that Cauchy was very close to
Jesuit mathematicians in Turin, to Father La Cheze, a professor of mathema-
tics and physics at the College of the Holy Martyrs, and to Abbe Moigno,
whom he had known for a long time. After his departure from Turin, Cauchy
proposed that one of these two men should be appointed to fill his chair in
sublime (mathematical) physics at the university (41).
The Revolution of 1830 and his consequent departure and exile from
France unquestionably marked a turning point in Cauchy's life. Deeply
involved in the extreme conservative movement and weakened by intensive
intellectual activity, he could not bear to see a regime that he wholeheartedly
supported collapse in a few days. The legitimate king went into exile; the
Jesuits went into hiding to avoid being slaughtered, and the liberals triumphed
at the Ecole Poly technique. Cauchy felt each of these events to be a personal
affront. He once again experienced the nervous illness that had afflicted him in
1812 when he was in Cherbourg, the symptoms being depression with
moments of excitement and physical weakness. When he left Paris early in
September 1830, his aim was merely to take a much needed rest, but this had
changed by early October and what had begun as a trip to recuperate became
voluntary exile, which Cauchy justified on grounds of his refusal to take the
oath, but which his family regarded as a flight pure and simple.
The events of 1830 also signaled a break in Cauchy's mathematical
productivity. Fourier's death, Legendre's death a few years later, and Galois'
tragic fate: all foretold the end of an era. Cauchy's absence from Paris was now
painfully felt. He was, to .be sure, temporarily replaced (and without any
158 9. Exile in Turin

apparent regrets) at the Ecole Poly technique by Coriolis, his friend and former
repetiteur. But, it was a long time indeed before anyone could even pretend to
take his place at the Academie. Writing to Libri in 1831, Sophie Germain
spoke in darkly prophetic terms when she observed that:
[There is] decidedly a kind offate or spell hovering over everything that
has to do with mathematics. Your own difficulties, Cauchy's problems,
M. Fourier's death, as well as that of the student Galois, who, for all
his impertinence, suggested certain exciting developments and
tendencies ... (42).
A few months after this letter was written, she herself passed from the scene.
 Chapter 10

The Education of the Duke of Bordeaux

At the beginning of the summer of 1833, Cauchy received a letter dated June 22
from the Baron de Damas in Toeplitz. De Damas was the tutor ofthe Duke of
Bordeaux, the grandson of Charles X, the former King of France, who was
now living in Bohemia. The letter stated:
Sir, you can be of great service in the education of the Duke of Bordeaux,
and the king himself ordered me to write to you. His Majesty would like
you to come, if it is at all possible, and work with my student and be in
charge of his education in the sciences, which up to now has been the
responsibility of M. Barande.
M. Barande has just been removed for reasons that grieved us greatly,
but which reflect neither on his honor nor on his abilities. I have thought
it fitting to mention this fact to you as I want to avoid giving grounds for
any unfounded suspicions that might conceivably arise.

With my highest regards and continued affection,

Yours truly,

Baron de Damas
In addition to this official letter, Cauchy received another letter, which was
filled with expressions of goodwill and high esteem for him and which closed
with the following words:
While I cannot speak of your personal position, I do know what your
sentiments are: the heir to our king needs your services, and I have been
given the responsibility of asking for them. I now assure you of the deep
attachment that I swore to you in times better than these and that will
not change.

Baron de Damas
159
160 10. The Education of the Duke of Bordeaux

In a postcript to this letter, the baron declared:

The king would like you to come to Prague and to proceed thence to the
countryside, where His Majesty resides, as soon as possible. Write me of
your travel plans and the presumed date of your arrival.
The king will leave Toeplitz during early July and will return to
Buschtierad, near Prague. His Majesty is quite pleased with his stay here,
since the waters helped him very much (1).

Cauchy gave his last examination at the University of Turin on July 22 and left
Turin for Prague, traveling by way of Geneva, Switzerland, and Bavaria (2).
Thus, a new life began for Augustin-Louis Cauchy, which would last for six
years.
In order to understand Charles X's decision, it is necessary to examine the
complex situation in the little court of exiles presided over by the old king.
Following the July Revolution, Charles X and his family had first taken refuge
at Lullworth, in England and then Holyrood, in Scotland. On August 2, 1830,
at Rambouillet, Charles, as well as his son, the Duke of Angouleme, abdicated
in favor of his grandson, the Duke of Bordeaux. However, on November 27, he
was declared regent until the Duke of Bordeaux reached his majority; that is to
say, until September 29, 1833, when the Duke would celebrate his fourteenth
birthday. This decision alienated the Duke of Bordeaux from his mother,
Duchess of Berry, Marie-Caroline, who was not trusted by the royal family at
all. In any event, the duchess had firmly decided to act on her young son's
behalf. Settled in Italy and surrounded by an entourage of frivoulous-minded
aristocrats, she contrived a grandiose but ill-conceived plan. The plan
essentially consisted of organizing a vast uprising in France to reestablish the
legitimate heir to the throne. When Charles X was informed of what was afoot,
he completely disapproved of the undertaking. However, encouraged by a
number of leading Parisian legitimists, Marie-Caroline paid no attention to
Charles' disapproval of her scheme, just as she paid no attention to the defeat
of the first conspiracy in February 1832, and continued to pursue her designs.
Thus, on April 29, 1832, she secretly landed in Provence. In June, the uprising
miscarried, and after a grotesque escapade, the duchess went into hiding in
Nantes in order to escape Louis-Philippe's police. But, on November 7,1832,
she was finally arrested and imprisoned in the fortress of Blaye.
Marie-Caroline became now a heroine and martyr for the Parisian
legitimists of the Faubourg Saint-Germain. But an even greater shock came in
February 1833 when it was learned that the duchess, having secretly remarried
while in Italy, was now pregnant. Charles, who in the autumn of 1832, had left
his refuge in Scotland and taken his family to Prague, where Emperor Francis
II placed the Hradschin Palace at his disposal, completely disowned his
daughter-in-law as soon as he learned of her remarriage and pregnancy. The
absolutist faction at the court of the old king was lead by Cardinal de Latil,
 10. The Education of the Duke of Bordeaux 161

Baron de Damas, the tutor of the young Duke of Bordeaux, and especially by
the Duke of Blacas, the old king's confidant. It now gained the upper hand over
the liberal faction, which was led by Madame de Gontaut, the governess ofthe
Duke of Bordeaux; the Duke of Gramont; and Joachim Barande, the director
of education of the young duke, and which still supported Marie-Caroline.
Even in France, the legitimists were split; many were simply outraged by the
duchess's conduct, but the most active of the legitimists, those who now went
by the name 'Young France', continued to support her. Chateaubriand came
to Prague to plead her cause. However, his efforts were of no avail.
The dispute between the two factions actually went beyond the duchess's
personality, for the real issue had to do with the kind of education that would
be given to the young man, who might, one day, occupy the throne of France.
Marie-Caroline's supporters thought that the Duke of Bordeaux should be
given a modern education, which would be relevant to the times. The aim of
the education they contemplated would at last be a reconciliation of the
monarchist ideal with the principles of liberty. Adhering to this view,
Chateaubriand had notions of becoming the tutor of the future Henri V.
However, the absolutists had a wholly different view of things. As they saw
things, the Duke of Bordeaux should be educated in the traditional way.
Accordingly, they reproached Barande who, in spite of his known merits, was
regarded as being too liberal minded. In January 1833, the decision to dismiss
him was made, and a replacement had to be found. In early June, the Marquis
of Foresta, the subtutor of the Duke of Bordeaux, returned from Rome and
brought with him two Jesuits, Fathers Druilhet and Deplace, both of whom
had formerly been professors at the College de Saint-Acheul. Damas's plan
was simple: he wanted to retain Barande, a brilliant poly technician and
engineer ofPonts et Chaussees, to teach mathematics and the natural sciences
to the young duke and to entrust instruction in other subjects to the two newly
arrived Jesuits. But, Barande refused to go along with this plan, and Charles X
sent him packing (3). This explains why, at the time Barande was dismissed,
Billot, a professor oflaw, and Cauchy were called (4). In August, when Cauchy
arrived in Prague, matters had just taken a new turn. The announcement of the
two Jesuits as tutors to the Duke of Bordeaux aroused a storm of protests back
in France. The Jesuits' unpopularity in France was a known fact, and this
appointment was seen as an outright provocation. Under heavy pressure from
legitimists in France and after the Emperor of Austria had intervened in the
matter, Charles X decided to dismiss the Jesuits. In August, Father Druilhet
left Prague, and in November, Father Deplace followed. Now, disowned by
Charles X, de Damas tendered his resignation and the liberal faction prevailed.
On September 9,1833, the Duke ofBordeaux reached his majority and several
aristocrats of the Young France persuasion came to Prague. By recognizing
him as king, they hoped to weaken the convervative influence of Charles X and
his family; once more, Chateaubriand arrived in Prague to make a final plea to
the old king on behalf of the duchess.
162 10. The Education of the Duke of Bordeaux

Under such conditions, Cauchy was not all sure that he would be able to
remain in Prague. In fact, on September 24, writing from Buschtierad, he
informed the Count of L'Escarene of the situation he faced in Prague:
Unfortunately, we still have not gone beyond provisional arrangements
here so that it is really impossible for me to say definitely whether or not I
shall remain here or whether I will return to Turin to teach the courses
entrusted to me by the King (5).

Nevertheless, taking part in the intrigues of the absolutist party, Cauchy had a
brochure published in Prague, entitled Quelques Mots Adresses aux Hommes
de Bon Sens et de Bonne Foi (6). In this publication, he declared his intentions of
participating in the education of the prince, writing:

The love which authors have for their own works and the attraction they
feel for the theories they invent are so well known that it is easy to
imagine how much it has cost me to interrupt the scientific works that I
had undertaken. But, what my own weakness kept me from doing while
my king was on the throne and in the Tuileries, I cannot refuse to do for
my king now that he is in exile, and for the child of the miracle who now
wears the double crown of glory and misfortune. A stranger to the
language of courts and to the art offlattering the mighty, I can only bring
a true heart and unblemished life to the service of my king. Many of my
friends have preceded me here to this land, which is so new an
experience to me. I no longer see them. When I look at what I have to
offer, I see new grounds for fearing to involve myself in so difficult and
challenging a career. But, I am well aware that the great ideal that
controls the education of the prince is both a religious and moral ideal,
which can inspire even the most exalted persons and which is the only
ideal that can form great princes and great kings ... I dare, therefore, to
reply to the voice that calls on me. I will not recoil in the face of any
obstacles.

After the dismissal of the Jesuits, Charles X appointed the absolutist

Bishop of Hermopolis, Monseigneur Frayssinous, tutor to the duke, during
final days of September, the Bishop appointed two instructors, Cauchy and
Billot, the two scholars who had been first summoned to Prague to assist
Fathers Druilhet and Deplace. At the same time, Charles summoned the
Marquis of La Tour-Maubourg, a man of fairly liberal tendencies, as
replacement for the Baron de Damas, the dismissed tutor. Citing the state of
his health as a reason, La Tour-Maubourg declined the appointment and, as a
substitute for himself, sent General d'Hautpoul, who arrived in Prague in early
October. From the very first, General d'Hautpoul encountered the under-
handed opposition of Billot and especially of Cauchy.
In his memoirs, d'Hautpoul discusses the reception he received, recalling
that:
 10. The Education of the Duke of Bordeaux 163

I learned that M. Cauchy had been named to replace M. Barande, and I

knew of his reputation as an outstanding scholar. He was totally
committed to the faction that was supporting the Jesuits against the
prevailing popular mood in France. I observed that Cauchy showed me
a very reserved attitude .... He often visited the Bishop of Hermopolis,
but he rarely came to see me, and when he did, he seemed embarrassed by
it (7).
General d'Hautpoul officially undertook his duties as subtutor on November
1, 1830, and an official ceremony was organized for the occasion with
Monseigneur Frayssinous, the Baron de Damas, d'Hautpoul, Cauchy, Billot,
and other members of the young Duke of Bordeaux's household attending.
Cauchy took advantage of this gathering to read a declaration in which he did
not bother to hide his hostility toward the decision taken against the Jesuits.
D'Hautpoul recalled the scene in the following terms:
M. Cauchy took a sheet of paper out of his pocket and read a farewell
speech that he had prepared in advance. Such a solemn way of going on
leave seemed rather bizarre to me, and moreover, the little speech was
given in such a voice that it was quite impossible for me to completely
grasp the sense of what was said. However, I did manage to catch some
words expressing regret at the departure ofM. de Damas and the Jesuits,
which was completely acceptable. Aside from this, however, I believe I
heard him say that he deplored the changes that these dismissals had
brought about and that he was repelled at the thought of having to
participate in them.
Thus, Cauchy, who did not even once mention General d'Hautpoul by name
in his statement, put himself squarely on the side of the Jesuits and against the
'obscure prejudices' by which they had been victimized (8).
As the weeks passed, relations between Cauchy and General d'Hautpoul,
instead of improving, became increasingly worse. Their notions on teaching
and pedagogy were diametrically opposite, for the latter wanted to endow the
young prince with a modem education, a training and outlook suitable for the
times they were living in, while the former filled his teaching with religious
considerations and indulged the misconduct of his high-born pupil to an
extent that d'Hautpoul found downright culpable (9). Moreover, Cauchy
never ceased to decry loudly the dismissal of the Jesuits. The final break
between the two men came in February 1834 when d'Hautpoullearned that
Cauchy had made a farewell statement to de Damas in the Gazette du
Lyonnais. Outraged, he had a meeting with Cauchy and demanded that he give
an explanation for his action. Cauchy, of course, replied to d'Hautpoul's
demands in a very vague, evasive way (to). Dissatisfied, d'Hautpoul demanded
of Monseigneur Frayssinous that Cauchy be dismissed. This dismissal did not
occur, and General d'Hautpoul handed in his resignation and left Prague on
February 20, 1834.
164 10. The Education of the Duke of Bordeaux

Thereafter, tutors came and went, one after the other. But, the young
prince's real education was taken charge of by Monseigneur Frayssinous and
his assistant, the Abbe Trebuquet, aided by the more or less permanent tutors,
Billot, Clouet, Montbel, etc., and, of course, Cauchy in the exact sciences. With
the departure of d'Hautpoul, Cauchy was assured that he would continue to
playa role in the education of the Duke of Bordeaux; and, that being so, he had
his wife and children come to Prague, to the Hradschin, in 1834 (11).
Cauchy's duties as royal tutor consumed most of his time. He took the
mission that had been entrusted to him very seriously; and, with great ardor
and great clumsiness, he set about trying to teach his high-born pupil the
rudiments of science. A short while after she arrived in Prague, AloYse Cauchy
wrote of her husband:

It would be hard to imagine a position that is more agreeable than the

one he has. However, at the same time, I must say, in order to justify
some ofthe reproaches we had voiced about him for his having failed to
write to us, that he does not have a moment to call his own. I hardly see
him except for an hour at dinner and later, for a few moments in the
evening. He is busy all day: in the morning there are lessons that he
must give; these are followed by other lessons that he assists and by
walks and strolls, which scarcely leave him enough time to add a few
words and algebraic symbols to a study that he is now writing and which
he hopes to send to the Academie just as soon as it is finished (12).

According to d'Hautpoul, Cauchy taught mathematics each day (at least in

1833) for one hour, from 8: 15 to 9: 15 a.m., and other sciences, especially
physics and chemistry, for one hour, from 5 to 6 p.m. At the end of each week,
on Saturday, there were exams, and the prince's close companions were
allowed to participate in them. Cauchy prepared his courses very carefully (13).
In fact, according to Moigno, he wrote elementary treatises on arithmetic and
geometry. Cauchy's mathematical imagination and creativeness compelled
him to look for original methods to his teaching (14). In line with this
creativity, he developed a new decimal notation which would enable mental
calculations to be performed more easily (15).
Were Cauchy's efforts successful with the prince? The results suggest a
negative answer. The young prince showed neither a taste nor talent for
mathematics and the exact sciences; and, moreover, Cauchy once again gave
proof of his own mediocre teaching abilities. We have only two pieces of
relevant evidence, and they agree completely on this point. The first of these
comes from d'Hautpoul, a man who had no liking for Cauchy personally, but
nevertheless seems worthy ofbelief(16). D'Hautpoul presents us with a picture
of Cauchy in 1833, shortly after the mathematician arrived in Prague. Here, we
see Cauchy busily working with the young prince in an attempt to alienate
him from the general's influence and from the memory of Barande.
 10. The Education of the Duke of Bordeaux 165

Indifferent to his pupil's ridicule and stoically suffering the child's

intolerable snubs and misconduct, Cauchy once more proved himself to be a
man of heroic patience. During the strolls that he took with the young prince,
the boy would frequently play tricks on him that passed the bounds of simple
jokes. 'For example', says d'Hautpoul, 'one day, when the young prince came
back from a walk, he brought with him a snowball and hit his tutor with it' (17).
Moreover, the child very often addressed Cauchy in 'stinging, disrespectful'
terms, and again Cauchy would tolerate such misconduct, showing an
extraordinary degree of understanding and compliance.
Two anecdotes illustrate the trials Cauchy endured in the service of his
pupil:
M. Cauchy, having seen that the story of my campaigns was of interest to
the prince [d'Hautpoul wrote] asked me permission to sit with him
sometimes in the evening in order that he might also tell him about his
own. He had been some time in the Ponts et Chaussees, and he called
"campaigns" the works that were executed each year during good
weather. Earlier in his career, M. Cauchy had been responsible for
repairing some of the sewers of Paris. The prince, with great malice,
seized upon this information and went about saying that M. Cauchy's
first campaign had been in a sewer (18).
On the Day of the Feast of the Kings, the Duke of Bordeaux was chosen king
by the person who found the bean. The prince then formed his council and
bestowed his favors. 'One person', d'Hautpoul tells us, 'was overlooked-
perhaps intentionally: M. Cauchy. "And I?" the latter asked, seeing that
everybody else had been given an appointment. "And", the prince replied, after
a moment of reflection, "You will be the minister of stars". This little flash of
wit caused a great deal of laughter, because everyone had observed that M.
Cauchy too often brought astronomical discussions into his conversations'
(19). But, if Cauchy failed to show firmness with the young prince outside the
classroom, pure chaos reigned in the courses. The prince had no liking or
talent for mathematics and would often put on violent temper tantrums. 'But
what good are all these numbers and figures that M. Cauchy always makes me
do?' the boy would complain to d'Hautpoul. Ignored and abused, Cauchy was
simply incapable of making the boy respect his authority. One day, the young
man took his tutor by the collar and literally pushed him out the door, 'and,'
recalled d'Hautpoul, 'M. Cauchy went along with this, muttering excuses and
asking what he could have done to displease Monseigneur'. On another
occasion, the young man was beside himself with anger and grabbed the table
cloth,jerking it so hard that the books lying on it went flying off, as did a basket
of coins that was lying on the table. Without budging, the child stood glaring
at poor Cauchy, who, on hands and knees, scurried about across the floor
gathering up the coins (20).
As for the courses, Cauchy never succeeded in developing a course at a level
suitable for the prince. According to d'Hautpoul:
166 10. The Education of the Duke of Bordeaux

[One ofthe things] that inspired the young man with such a dislike for
mathematics was that he was never able to understand what was being
taught him. One day, I attended one of the sessions and during this class,
the professor was proving one of the most simple propositions in
geometry. I watched as the prince became more and more agitated and
confused, and I swore to myself that although I remembered this
proposition very well, I myself could hardly follow Professor Cauchy's
argument (21).

In order to soothe his pupil's hostilities, Cauchy began substituting other

activities for mathematics: one day he might tell stories to the boy in order to
amuse him; and on the next day, he and his pupil would sing songs and
homilies from Saint-Sulpice en duo (22).
As time passed, the prince began to pick up on the rudiments of mathe-
matics. However, according to an account by the Marquis of Villeneuve,
who came to see the prince at Goritz, Cauchy had a conspicuous lack of
success in getting mathematics across to his pupil. Villeneuve recalled:

At Goritz, as elsewhere, I paid close attention to the exams that were

given each Saturday. Unfortunately, these always began with mathema-
tics. When questioned by Cauchy on a problem in descriptive geometry,
the prince was confused and hesitant. The king [the Duke of
Angouleme] then stood up, explained the problem, and was praised by
Cauchy for having done so. There was also material on physics and
chemistry. But, as in the case with mathematics, the prince showed very
little interest in these subjects. Cauchy became annoyed and screamed
and yelled. The queen [the Duchess of Angouleme] sometimes said to
him, soothingly, smilingly, 'too loud, not so loud!' (23)

Thus, we see that Cauchy had to have an extraordinary faith and sense of
mission to tolerate the isolation in which he found himselffor five years-first
at the Hradschin Palace in Prague and then, after May 1836, at the
Kirchberg Castle in Toeplitz, and finally at Goritz-in a petty, illiberal,
cynical court in exile. Moreover, it all seems to have been for no good reason,
because, when the Duke of Bordeaux finished his formal education in October
1838, he had acquired an abiding dislike for mathematics.
Woro out by his efforts to educate the young prince, Cauchy did not do
much research between 1834 and 1839. Until the end of 1834, it seems that he
worked on the publication of the five installments of his Resumes Analytiques,
which was being printed in Turin and which he had begun the preceding year
(24). On the other hand, resuming the work he had first started in Turin in
1833, he published in lithographic form-this time in Prague-an important
study entitled Sur l'Integration des Equations Differentielles. In it, he presented
a new application of his calculus oflimits. Thus, he carried through to fruition
a goal he had set for himself in 1831: to use the calculus oflimits to determine
 to. The Education of the Duke of Bordeaux 167

fi ·

f .. fU I :... Il~L,:to}. :.!(!':~ )

f
'taJ )
·ri·.~~·)I-,I" I[~ "1 ./~} . . . ./9'e~/~I{!!P~I
- ! r t _ _ -~ - __ - ' -
!p''.,.~''~ t
---.
.wtf-r !!.!;Ir':::.1~ .I"·l"'t)
(9'@'t..~f!.l·
1... I

1Jtt.§r~~J~
I
.~ iff#/~

J ~ f~Jl-l~!¥~ .. I-} :4{'ti;!tJ,

1uJ.) •
. ~ ... ~ ,,~
1
( , .. ~1-.I!-) j.f.'!.!:,t~ J •
I\,. " ,

f"'1~L... /~'h-- -c:..~ q~-~ V- y--.----

l..9~;·....-r T,~ J-tJ- -fo=. ~- ' ~x..

Research work on the theory oflight. Manuscript by Cauchy, January 1836, Sorbonne
Library, ms 1762. Published by permission of the Ecole Poly technique.

the solution (x(t), y(t), z(t), ... ) of a system of ordinary differential equations:
dx dy dz
Fl(X,y,Z, . .. ,t) F 2 (x,y,z, ... ,t) F3(X,y,Z, ... ,t)

dt
(10.1)
Fo(x,y,z, ... ,t)'
168 10. The Education of the Duke of Bordeaux

satisfying the initial conditions x(r) = ~, y(r) = 1'/, z(r) = (, by means of series.
In this system, the functions F 0' F l' F 2' ... where supposed to be continuous
and finite.
Hamilton's study, 'On a general method in dynamics', the first part of which
was published in 1834 (25), had led Cauchy back to the idea of reducing the
resolution of Eq. (10.1) to the equivalent problem of determining the prime
integrals X(x,y,z, ... ,t,r)=~, Y(x,y,z, ... ,t,r)=1'/, Z(x,y,z, ... ,t,r)=(, ... of Eq.
(10.1), such that X(x,y,z, ... ,t,t)=x, Y(x,y,z, ... ,t,t)=y, Z(x,y,z, ... ,t,t)=
z, ... (26). More generally, a prime integral of Eq. (10.1) can be expressed
by the equation U = D, where U = u(X(x, y, z, ... , t, r), Y(x, y, z, ... , t, r),
Z(x, y, z, ... , t, r), ... ), and u = u(~, 1'/, 2, ... ). Then, Cauchy showed that U - u
is a solution of the first-order linear partial differential equation

(10.2)

He constructed this solution by an iterative method, much like the one he

had used in 1825. If Vs is the integral

_ f (F
1 F
1 OS + 2 OS + F3 OS + ... ) dt
< F 0 OX F 0 oy F0 oz
and "1"s are the multiple integrals "1("1(. ··("1s)·· .)), the iteration of the integral
equation U - u - "1U = 0 yields the development

Therefore, L ("1nu) - v represents a solution of Eq. (10.2), if the series L ("1"u) is

convergent.
Cauchy used the calculus oflimits to establish the convergence of the series
L"1nU. Thus, he obtained not only approximate values of the solution U - D of
Eq. (10.2), but also an existence theorem for the solutions of the holomor-
phic differential system, Eq. (10.1), which differed from the existence theorem
he had established in his lessons at the Ecole Poly technique. For instance,
in the case of a single equation, dx = F(x, t) (with x(r) = ~), he found that
dt

F[X+x,t+(}(r-t)]}n
mod "1"x ~ mod { 2(r - t) A X A (x + x)

where x = re pF is on the circle C(x, r) of center x and radius rand () is a

certain number in [0, 1 [ depending on r, by applying the inequalities
modj<")(x) ~ n!r-" A f(x + x),
which he had given in his first memoir of Turin. F(x) and f(x) were supposed
to be finite and continuous inside and on the circle C(x, r). Therefore, the
 10. The Education of the Duke of Bordeaux 169

series L (\7 nx) is convergent if

mod {2(r _ t)/\ F[x + x,t: 8(r - t)J} < 1

and then the solution x(t) such that x(r) = ~ exists and can be expanded into
a convergent series in a neighborhood of r. Cauchy also gave the majorant
mod {2(r - t) /\ F[x + x,t + 8(r - t)J/x}n _
--~-----'-----------'-----'-=-----'--- /\ (x + x)
1 - mod {2(r - t) /\ F[x + x,t + 8(r - t)J/x}
for the rest of order n of the series expansion of the solution.
It was to the theory oflight that Cauchy devoted most of his efforts during
the years 1835-1836. The publication of his study, Sur la Dispersion de la
Lumiere, which had been held up by the turmoils of 1830, came at the
beginning of this period of new research. He presented this study to the
Royal Society of Sciences of Prague, and that body, in effect, accepted the
responsibility of financing the printing of this work as a separate study.
However, when it began to appear in October 1835, it was not en memoire
detache, as was originally foreseen; rather, it was included in a new series of
Exercices entitled Nouveaux Exercices de Mathematiques, which followed the
pattern of Exercices de Mathematiques of 1826-1830 and Resumes Analytiques
of 1833-1834 (27).
Cauchy was not content with publishing his 1830 manuscript. He
undertook new investigations that were only partly published at the time.
In the case of an isotropic ether, he showed that the velocity V of a plane
wave can be written
V = a 1 + a2k2 + a3e + "',
where k is the magnitude of the wave vector k. The coefficients an depend
on the medium and decrease very quickly when n increases. Cauchy deduced
from this series-expansion the value of k 2 :
P = b 1 + b 2 s2 + b 3 s4 + "',
where
s=kV and

U sing a new method for the calculation of errors which he had presented in the
lithograph 'Sur l'interpolation' (28), he proved that his formula is compatible
with the results of Frauenhofer's experiments on the indices of refraction in
different media.
One highly significant paragraph of his study on dispersion was devoted to
the physical properties of molecular ether. Cauchy considered the relation
between sand k in the case of an isotropic ether that is not dispersive. The
intermolecular forces were supposed to be attractive at a great distance in
inverse ratio to the square of the distance and repulsive in the neighborhood of
170 10. The Education of the Duke of Bordeaux

each molecule, in inverse ratio to the fourth power of the distance. With this
hypothesis, Cauchy showed that the ether is necessarily an extremely
dense system of molecules. Its density varies according to the material medium
involved, maximal in empty space, where there is no dispersion, and minimal
in the dispersive bodies.
In August 1836,just before his departure for Goritz and hard on the heels of
the publication (in installments) of his study on dispersion, Cauchy had a
lithograph of another of his works published.
Entitled Sur la Theorie de la Lumiere, this study was an extract of the
materials from the notebooks of 1836 (29).
The first part of this study was devoted to some mathematical preliminaries.
In particular, Cauchy examined a new method for determining the boundary
conditions for a body regarded as a system of molecules. The remaining
sections treated various questions concerning the theory of light. Cauchy
developed a new method for transforming the equation of motion of a system
of molecules:

uOui=£..,m
::>2 " p (!(rp ).1pUi
--+ ( rp ! ' ( !(rp))llpillpj--.
rp)- .1pUj ) (6.13)
p rp rp
Instead of representing the deplacements Ui by Fourier series expansions,
as in 1830, he used Taylor's series. He simplified the notation by means of the
symbolic operator eTII,O" which represents the symbolic series expansion
relative to 0i:
r2
1 + rll i 0i + 2(lli oif + ....
Thus, according to the analogy between the powers and the differences, we
have the symbolic equation

Cauchy then defined two other symbolic operators:

V= L mp !(rp) (1- eTpIlp,O,),
p rp
and
D= Lm !(r )((p,Pi Oi)2 _ eTPIIP'O').
p
p

p rp 2 r;
He obtained new operators Dij by derivating D with respect to 0i and OJ,
symbolically interpreted as variables; for instance, if i = 1, and j = 2,

D 12 - " !(rp) (::>

- £.., mp- - l lp1Ilp2 u2 - e
TpIl",O,)
.
p rp
Finally, he showed that Eq. (6.13) can be rewritten in the form
O~Ui = 0 ijUj + VUi. (10.3)
 10. The Education of the Duke of Bordeaux 171

If the system of molecules is isotropic, Cauchy obtained the equations

(10.4)
and
(10.5)
where Vij is the cubical dilation ajUj and the operators <>, ~,and aii ~ are
defined by the symbolic equations

and

Equation (10.4) generalizes Eq. (6.15). Cauchy also established the general
equations of motion for a uniaxial system of molecules. Finally, he investi-
gated the propagation of plane waves in isotropic and uniaxial systems of
molecules.
Because of his departure, Cauchy could not pursue the publication of the
conclusion of his two studies, Sur la Dispersion and Sur la Theorie de la
Lumiere, as well as the remainder of his 1836 notebooks. The problem
concerned research on the propagation of spherical and cylindrical waves,
on shades, on diffraction, and especially on a new theory of reflection and
refraction on a transparent or opaque body (30). From letters that he sent
to Libri and to Ampere in early 1836 (31), from an unpublished notebook
manuscript from 1836-1837 (32) and from several notes from the Comptes
Rendus Hebdomadaires des Seances de l'Academie des Sciences, which were
published after 1839 it is possible to get an idea of this research (33). We
will merely mention Cauchy's new theory of reflection and refraction here.
Considering the variation of the density of ether according to the material
medium involved, Cauchy examined the behavior of light at the boundary
of a body. By applying the methods he had developed in the beginning of his
lithographic study 'Sur la theorie de la lumiere', he derived the laws that had
been articulated earlier by Brewster and Fresnel. He now had to assume that
172 10. The Education of the Duke of Bordeaux

molecular vibrations do not act parallel to the polarization plane (as he had
supposed in his 1830 theory), but rather perpendicular to this plane in line with
Fresnel's hypothesis.
While he was in Prague, Cauchy was scientifically isolated. His relations
with the local scientific community seem to have been very tenuous. On April
20, 1836, he presented his lithographed paper Sur l'Integration des Equations
Differentielles, and as we have seen, a few months later, he obtained a subsidy
for the publication of his paper Sur la Dispersion de la Lumiere (34). It
should be noted that around 1834 a meeting between Cauchy and Bolzano
took place. There seems to have been no subsequent meetings between the two
men, and the one that did take place appears to have been sought by Bolzano,
who had sent Cauchy a tract on the problem of the rectification of curves,
which he had written entirely in French for Cauchy's benefit (35).
Charles X and his court left the Hradschin Palace and the city of Prague in
May 1836 for Toeplitz. This was necessary since the new emperor Ferdinand
had come to Prague to receive his investiture as King of Bohemia. Charles
planned to settle in Garitz, being attracted by the climate of the southern
region. En route to Garitz, Charles and his entourage stopped at Kirchberg
Castle, an estate that had been purchased by Blacas, because of a cerebral fever
that affected the Duke of Bordeaux in August 1836. There, the old king
remained until October 8, 1836, when he proceeded on to Garitz, where he
died on November 6, 1836, a few days after his arrival. The Duke of
Angouleme and the Duke of Bordeaux spent the following years traveling
between Kirchberg and Garitz, with Cauchy accompanying the prince on
these peregrinations (36).
Cauchy continued his research on the theory oflight while he was at Garitz,
and he also did work on a new study on analysis and on the calculus oflimits.
During 1837, he sent several pieces to Libri or directly to the Academie, works
that dealt with the problem of the determination of the roots of an algebraic or
transcendental equation and were to be published in the Comptes Rendus (37).
The Duke of Bordeaux reached his eighteenth birthday in September 1838,
and this marked the conclusion of his formal education. It also ended Cauchy's
duties with the exiled court. Would he now return to France or continue his life
as an exile? It is clear that his friends, Moigno for example, urged him to return.
But, it was Cauchy's mother, Marie-Madeleine Cauchy, who was now
decisive. Her golden wedding anniversary had been celebrated the preceding
year; and no doubt sensing that the end was now drawing near, she was loathe
to have her eldest son so far away (38). Indeed, she was to die on May 5, 1839,
less than seven months after Augustin-Louis had returned. At last, in October
1838, Cauchy and his family arrived back in Paris and took up residence once
again in the de Bure townhouse on the rue Serpente. Cauchy's first act upon
returning to Paris was to attend the October 22, 1838, meeting of the Academie
des Sciences. He was now approaching 50, and a new period in his life was
beginning.
 10. The Education of the Duke of Bordeaux 173

Cauchy had little to show for the eight long years he had spent in exile: the
title of Baron, which Charles X and bestowed on him for his years of faithful
service-and to which Cauchy was very attached-and several notebooks
that were filled with handwritten notes on various mathematical topics. It is
from these notes that he would draw much of his scientific inspiration in the
coming years.
 Chapter 11
The Legitimist Mathematician

When he returned to France in the autumn of 1838 after an exile that had
lasted eight years, Cauchy was firmly resolved to participate fully in
contemporary scientific life. His refusal to take the oath of allegiance in 1830
had meant the loss of his professorial chairs. But, he still had his position at the
Academie des Sciences, because it had not been required that he swear to the
July Monarchy in order to keep his seat there. But, in spite of the many
advantages that came with membership in the Academie-aside from
attendance at the meetings and the rapid publications of notes and studies in
the Comptes Rendus Hebdomadaires des Seances-Cauchy could not be
content with a single position in the scholarly world.
Opportunity presented itself on July 29, 1839, with the death of Cauthy's
old teacher, Prony, which created a vacancy in the geometry section of the
Bureau des Longitudes. The Bureau des Longitudes occupied a unique
position among the French scientific institutions of the time. Like the
Academie des Sciences, it had the right to choose its new members, and the
king merely approved its choice. In this respect, the Bureau tended to regard
itself as something of an academy of the astronomical sciences in its own right
(1). However, members of the Bureau-unlike members of the Academie-
were under an obligation to take oaths of political allegiance. Within the
Bureau, such oath taking was merely a formality to which no one paid any
great attention. Accordingly, it was assumed that if Cauchy were elected, the
matter of an oath would not assume the same importance that it assumed
when the question was one of appointment to a professorial chair.
A few days after the announcement of Pro ny's death, Cauchy entered the
competition. On August 5,1839, he presented a paper on celestial mechanics to
the Academie (2). This study was in reality only indirectly related to
astronomy. In it, Cauchy gave a proof of his famous Turin theorem of 1831,
dealing with the series expansions of functions. He pointed out the possible
applications ofthe theorem to celestial mechanics, but did not detail or explore
174
 11. The Legitimist Mathematician 175

them in any way. During the same meeting, Cauchy paid homage to Prony, 'an
illustrious colleague', he said, 'who, many years ago, seemed to have taken
such pleasure in having me as one of his students and who willingly
encouraged me with my first scientific works' (3).
The elections were to be held in the fall, and a commission was appointed on
October 30, 1839. Aside from Poisson, who, since Prony's death, was the only
member of the geometry section of the Bureau, the commission
consisted of Biot and Arago. But, these three men were unable to agree on a
common list, and on November 6,1839, addressing a meeting of the Bureau's
members, Arago declared:
The majority, consisting of MM. Biot and Arago, recommends MM.
Cauchy, Liouville, and Sturm as candidates; M. Poisson recommends
M. Lacroix (4).

It seems that Biot and Arago were especially supportive of Cauchy's

candidature. According to Valson, Arago was rather blunt and impatient with
the other applicants to whose inquiries he invariably replied 'M. Cauchy is
competing'. To those petitioners who persisted in their inquiries after having
been given the standard answer, he would repeat, making a humorous little
gesture: 'But, sir, I have just told you that M. Cauchy is competing' (5). Arago's
influence in the learned society of that time is well known, and by supporting
Cauchy'S candidature, he was, in effect, showing a deep respect for a man that
he did not like, either personally or politically, and with whom he had often
clashed over scientific questions. We should not be surprised by Poisson's
attitude, because, as we have seen, he and Cauchy had been on bad terms since
1825. On April 15, 1839, shortly after the latter's return to France, an
altercation once again flared up between these two scientists. The disagree-
ment centered around a study by Cauchy, 'Sur la quantite de lumiere reflechie
sous les diverses incidences par les surfaces des corps Opaques et specialement
des metaux'. Several notes were written, prolonging the dispute over the
following few weeks (6).
On November 14, 1839, the Bureau held the election. Definitely hostile to
Cauchy, Poisson made one last attempt to swing the election in favor of
Lacroix, his candidate. The record indicates that:
M. Poisson asked to speak regarding the nomination for which the
presentation had already been made at the last meeting. It was pointed
out that, at the Academie des Sciences, it has never been the practice to
resume a discussion on a topic once the [formal] presentation had been
made. M. Poisson replied that no such presentation was obligatory at
the Bureau des Longitudes. Nevertheless, it was observed that even at
the Bureau there is no instance on record where the discussion was
resumed, but M. Poisson's motion was not opposed at this time. M.
Poisson spoke of the great services that M. Lacroix's works had
rendered to public instruction.
176 11. The Legitimist Mathematician

But this attempt failed, and:

Following a discussion in which several members participated, a secret
ballot was held. The outcome was that M. Cauchy received the majority
and was accordingly nominated as a member of the Bureau. His
nomination will be submitted to the king for approval (7).
Thus, a long and painful story began. A few days after the election, the
Ministry of Public Instruction, which controlled the Bureau des Longitudes,
sent Arago, the president of the Bureau, a letter about the recent nomination.
Arago informed his colleagues accordingly:
Being in doubt as to whether this letter had been addressed to him
personally or to the Bureau itself, M. Arago read it to the staff and sought
the Bureau's opinion as to the proper response. Basically, the Minister
asserted that he would submit M. Cauchy's nomination to the king for
approval when M. Cauchy had taken the oath. Is it up to the Bureau to
make inquiries into M. Cauchy's frame of mind in this regard? This had
never been done before, but if the Bureau takes the letter into account,
then it would be obliged to take a position.
On this matter, the Bureau is unanimous in its opinion that political
considerations should have no part in this nomination and that the
Bureau can in no way involve itself in the matters that the minister spoke
of, nor can it concern itself with either M. Cauchy's intentions or
opinions. Thus, in his own name, M. Arago replied to the Minister (8).
The situation at the Bureau became absurd: Lacking the royal approval of his
election, Cauchy could neither receive payment from the Bureau nor
participate in its meetings, even though he remained an elected member.
Good will intervened to clarify the situation (9). Villemain, Minister of
Public Instruction, favored a compromise. This was also true of Victor Cousin,
who succeeded Villemain in the ministry after March 1840. J. B. Biot, who was
a member of the Bureau des Longitudes at that time and who perhaps played a
role in the affair, declared:
During the time that this problem was festering, the Ministry of Public
Instruction was occupied successively by two persons of great literary
renown, whose characters as well as inclinations placed them above
petty and malicious aims. These two men labored, in ingenious ways, to
reduce to the fewest possible conditions what their positions required
them to obtain. They were willing to accept the minimum that would
permit them to comply with the law. But, Cauchy shunned any such
compromises ... (10).
Still adhering to his attitude of 1830, Cauchy refused to take any oaths. Elected
but not appointed, since his election by the Bureau was not approved by the
king, he acted as though he were a full member of the Bureau, except that he
could not participate in that body's business meetings. Accordingly, starting in
 11. The Legitimist Mathematician 177

1839, he devoted an important part of his research to celestial mechanics. As

for the men in authority at the Bureau, they did nothing to curtail Cauchy's
activities in the sense that for four years they acted as though they were simply
waiting on him to swear the oath, even though they knew that this was
something he would never do.
On April 25, 1840, Poisson died, and this prompted Cauchy to devote even
more of his research efforts to mathematical astronomy. Moreover, Poisson's
death removed the last of the geometers from the staff of the Bureau des
Longitudes, even though this was precisely the time at which the Bureau was
beginning to recognize the need for staffing itself with as many theorists as the
observational staff (whose number was also to be increased) (11). Since
Laplace's death, mathematical astronomy had been a rather neglected area of
research in France-and that in spite of the works of young French scientists,
such as Liouville and Leverrier, compared to Germany, where a number of
mathematicians, such as Gauss, Bessel, and Hansen, developed the methods of
applied analysis as related to celestial mechanics.
Cauchy felt a need to prove again that he was worthy of holding the
position to which he had been elected. The fact that he was not able to submit
his papers to the Bureau des Longitudes did not hinder his research in
mathematical astronomy, so that between June and November 1840, he
presented a dozen papers to the Academie on celestial mechanics or on the
applications of mathematical analysis to celestial mechanics (12).
On November 18, 1840, Liouville was elected to fill the vacancy that
Poisson's death had created at the Bureau des Longitudes; Cauchy sub-
sequently slackened his production. Nevertheless, in 1841, he still presented
several important studies on celestial mechanics, which were inspired by
Leverrier's research on the minor planet Pallas. During the following years, he
continued to do research on mathematical astronomy, even after he was
excluded from the Bureau des Longitudes in November 1843.
Cauchy's opposition to the regime was not expressed solely in his refusal to
take the oath of loyalty to the king of the French. From the moment he
returned to France in late 1838, he had favored the Jesuits in the struggle
against the Universite and for the 'freedom of teaching'.
Since 1837, the Jesuits had been training teachers and professors for their
colleges at the Ecole Normale Ecclesiastique on the rue des Postes in the
Latin Quarter. While it has not been possible to determine the truth of
Hoeffer's assertion that Cauchy played an active role in mathematical
teaching (13), it is nonetheless clear that he regularly went to the rue des
Postes to consult, at the very least, with Abbe Fran<;ois Moigno, who was in
charge of the young Jesuits' training and instruction in mathematics and
physical sciences.
Fran<;ois Napoleon-Marie Moigno was born on April 20, 1804, at
Guemene in Brittany. His father, a member ofthe petty aristocracy ofBrittany,
had emigrated during the Great Revolution. In 1815, young Fran<;ois
Napoleon-Marie enrolled in the College of Pontivy and, the next year, entered
178 11. The Legitimist Mathematician

the Seminarie of Sainte-Anne, a small institution operated by the Jesuits. He

left for Montrouge, having decided to become a member of the Society of Jesus
in 1822, when his seminary days would be over. However, in September 1824,
under orders from his superiors with a view to studying mathematics and
physics, he entered the Ecole Normale on the rue de Sevres. There he remained
until 1829, studying most of his mathematics under Charles Leroy, a professor of
mathematics of the Ecole Poly technique. He also took classes under Cauchy
at the College de France and at the Faculte des Sciences. Even at this time,
Moigno seems to have been deeply attached to Cauchy and, in fact, in 1828, he
even presented him with a new method for obtaining the equation of a tangent
plane, a method that Cauchy set forth in his Exercices de M athematiques of
1828 (14). In October 1829, he left for Saint-Acheul, where he studied dogmatic
theology until the Revolution of July 1830. On September 6, 1830, he left
France, going first to Brigg, Switzerland, and then to Turin. In September
1833, Cauchy recommended that Moigno be appointed as his sucessor to fill
the chair in higher physics that he was now vacating at the University of Turin.
Upon returning to France, Moigno taught theology at the Scholasticat of
Vals, near Puy-en-Velay, and later, in 1835, at Saint-Acheul.
In October 1836, he returned to Paris where he had been appointed to fill a
chair in mathematics at the Ecole Normale Ecclesiastique in the rue des
Postes. Once at the Ecole Normale, he established a physics department that
soon attracted the attention oflearned society. He began to acquire a name for
himself in the scientific community, a reputation that connected him with a
number of leading scientists of quite different persuasions, such as Arago,
Liouville, Thenard, and, of course, Cauchy, with whom he kept in contact
before 1838. From 1839 on, Cauchy frequently called on Moigno, regarding
him as his disciple. The closeness of the relationship between these two men is
indicated by the fact that Moigno's courses at the Ecole Normale were
inspired by courses that he had taken under Cauchy and by the fact that
Cauchy entrusted manuscripts to him (15).
With Cauchy's full agreement and support, Moigno published the first
volume of the Le{:ons de Calcul Differentiel et Integral in 1840. In the
introduction, Moigno invoked the authority of his distinguished teacher,
declaring:
M. Cauchy, whom I am honored to have had as a teacher and who I am
even more honored to have as a friend, openly accepted me as
intermediary and echo in his scholarly communications with the public.
(16).
Several chapters of the Le{:ons were taken, almost word for word, from the
manuscripts that Cauchy had entrusted to Father Moigno, this being
particularly true of an unpublished treatise on the calculus of finite differences,
and from a manuscript notebook that would have become the third volume of
the Applications du Calcul Infinitesimal Ii la Geometrie (17). Moigno's work was
quite successful.
 11. The Legitimist Mathematician 179

Father Moigno was involved in many other activities. He directed the open
reunion of the Society of Saint Francis-Xavier at Saint-Sulpice, the parish in
which he planned to create a museum devoted to sculpture. He held retreats,
preached, and took part in charitable works as well as writing articles for
Catholic journals, particularly for the Univers and the Union Catholique. His
superiors were very suspicious of all these activities, regarding them as
agitation. Even more seriously, however, Moigno made the mistake of
becoming involved in some unfortunate speculations and financial dealings.
Along with several other persons, he had incurred debts amounting to tens of
thousands of francs in order to finance the rather shady dealings of the
Marquis de Jouffroy, the inventor of the palmipedes motor. He managed to get
Cauchy to agree to make a favorable report to the Academie on an invention
of Jouffroy's in November 1840 (18). A major financial scandal threatened to
break out, and sensing the danger Father Boulanger, Moigno's superior,
decided to send the imprudent professor to Laval so that he might teach a
course in Hebrew there (19).
Moigno refused to obey, preferring to go into hiding in Paris; and, after a
four-year battle, he withdrew from the Jesuit Order in October 1843. He
continued his rather stormy career as an ecclesiastic over the following years.
In 1844, he published the second volume of his Le~ns, which, like the first
volume, contained many chapters that were based on approaches that Cauchy
had taken in the courses he had taught during the Restoration era (20).
Cauchy seems not to have been offended by Moigno's break with the
Jesuits, and continued, at least until 1844, to visit his old friend regularly,
coming to see him during retreats and occasionally writing him. Moreover,
Cauchy agreed to participate in a commission charged with the responsibility
of evaluating the study Jouffroy submitted on June 12, 1843, entitled 'Sur un
nouveau systeme de chemin de fer'. Not only did Cauchy work on the
commission, but he also wrote a favorable report about it in 1846 (21).
In 1839, while he was lending support to the Ecole Normale Ecclesiastique
with his prestige and knowledge, Cauchy took part in the founding of the
Institut Catholique, an organization in which he played a key role (22). The
purpose of this institution was to offer philosophical, literary, and scientific
conferences and thereby soften the effects of the absence of Catholic univer-
sity education. This organization really gained impetus in 1842, after
the founding ofthe Cercle Catholique in November 1841 by Ambroise Rendu.
With its more liberal orientation, Rendu's Cercle Catholique also offered
conferences and gatherings that would rival those of the Institut's, and it
counted figures such as Montalembert and Ozanam among its participants
(23). The governing bylaws of the Institut Catholique were established during
several meetings in January 1842, Cauchy being secretary (24). According to a
notice that was issued in February 1842, the Institut Catholique was to be 'a
union that is open to both the youths who come to the capital each year in
search of instruction and to mature persons who have retained a desire for
learning'. Weekly meetings and lectures were held in the rue de Verneuil
180 11. The Legitimist Mathematician

where the Institut maintained a library and reading room; the lectures were
under the direction of 'special committees consisting of persons who are
veritable authorities in science' (25). The Committee of Law and Letters was
directed by Pardessus, a member of the Academie des Inscriptions et Belles-
Lettres, and Cauchy presided over the Committee of Sciences.
A short time later, the Arts Committee was added to the two initial ones,
and it was directed by Raoul Rochette, the Permanent Secretary of the
Academie des Beaux-Arts. Heading the Committee of Sciences, Cauchy
gathered a group of legitimist scientists around himself: Coriolis, Binet,
Freycinet, Beudant (all from the Academie des Sciences), Leroy, Cayol,
Gautier de Claubry, Auguste de Sainte-Hilaire, and the doctors Cruveilher,
Recamier, and Tessier (16).
Cauchy was also a member of the commission responsible for managing the
business affair of the Institut. A general meeting was held once each month,
and during these sessions public lectures were given. These lectures were
published in the Bulletin de l'lnstitut Catholique, a publication that regularly
appeared from 1842 until 1844 (27).
Along with Auguste de Sainte-Hilaire, Dr. Tessier, and Gautier de Claubry,
who were in charge of the conferences on botany, medicine, and physics,
Cauchy gave lectures on mathematics each Wednesday evening at 8 o'clock.
There were also regular poetry readings, as well as conferences that took place
during the course of the general meetings. The function of the scholars and
scientists associated with the Institut Catholique was, in a word:
to enlighten the youth and instruct their minds, to direct them in the
ways and habits of study and work, to instill in them a taste and love for
truth and beauty, and to act as models that they should pattern
themselves on, to share in their works and efforts whenever possible, and
to confirm them in the goodwill and affection that will encourage them
to hold fast to the path of virtue and not be influenced or tempted by
perfidious suggestions (28).
With regard to this statement of purpose, Cauchy himself wrote:

A youth who is studious and eager to learn will after a few years devote
himselfto the cultivation of science and letters without losing the point of
view of religion, which comes before science. Such a youth would want
the benefit of the experience of those who have preceded him and would
hope that the true teachers and masters of science, men who are
distinguished, with proven abilities, and who have a sincere attachment
to the Catholic faith, would serve as his guides. Such a hope cannot be
denied. The members of the two committees vie in their zeal for working
toward the success of so beneficial an undertaking. They all pray, and
indeed, have long since prayed to God that He Himself will bless their
works, which cannot fail to echo his glory since such labors as these have
the search for truth as their real and final end (29).
 11. The Legitimist Mathematician 181

In spite of the efforts of its associated scholars, the Institut Catholique had
only limited success. This was no doubt due, in large part, to the competing
efforts of the Cercle Catholique, an organization that held very prestigious
conferences and gatherings under the direction of Ozanam.
Paralleling his efforts on behalf on Catholic education, Cauchy resumed his
work with charities, the 'good works' to which he had devoted so much time
and effort prior to 1830. The Congregation was now disbanded, and a whole
new generation of charitable organizations had since come into being,
organizations in which there was an uneasy coexistence between conservative
and liberal Catholics. The most important of these new charities was the
Societe de Saint-Vincent-de-Paul. This organization was founded on the ashes
of the old Societe des Bonnes Oeuvres in 1833 by a group of students and
followers of Ozanam and Bailly. Upon his return from exile, Cauchy joined it
and actively participated in its charitable works by creating the Conference of
Sceaux (30).
In any event, the modernity of Ozanam's works and efforts, which had
issued from the more realistic charitable practices of the Restoration Era but

Augustin-Louis Cauchy, medallion by David d'Angers, 1843. Published

by permission of the Ecole Poly technique.
182 11. The Legitimist Mathematician

were free of the reactionary bigotry of the old Congregation, could hardly
satisfy a person like Cauchy. This was so because Cauchy's hopes, dreams, and
ambitions, like those of other legitimists, centered on a return to the
counterrevolutionary ideology of the Societe des Bonnes Oeuvres, which had
flourished before 1830 (31).
In April 1839, an organization, Catholicisme en Europe, was founded under
the inspiration of Monseigneur Gillis, the Bishop of Edinburgh. The purpose
of this organization was to 'aid and comfort Catholics in the Protestant
countries of Europe', and Cauchy actively participated in the group from the
start. This group was led by Ferdinand Bertier de Sauvigny and had Cauchy as
one of its most active propagandists. Entitled Annales du Catholicisme en
Europe, the group's publication enjoyed a real success for a time, having
the approval of some 40 bishops and counting some 1200 subscribers in
France as of March 1841. Nevertheless, this group met with the resistance of
the Commission of Lyon of the Association pour la Propagation de la Foi
dans les Pays Infideles (Association for the Propagation of the Faith in
Heathen Lands), a missionary group that was headed by Ozanam and of
which Cauchy was a member. Well-established in Paris and possessing a
certain credibility, which was, no doubt, attributable to Monseigneur Affre,
the Archbishop of Paris, Catholicisme en Europe seemed to be a formidable
competitor with suspect political intentions.
Two letters that Ozanam wrote to Meynis in March 1841 throw light on the
basis of the conflict. Denouncing the 'scheme of infiltration' that legitimists
such as Cauchy were engaging in as members of the charit~ble associations, he
wrote:
M. Bailly thinks that the operations of Catholicisme are the business of
certain religious legitimists who are now sorry that they let all the
charitable associations slip through their fingers, and who, in particular,
are now attempting to monopolize the Societe de Saint-Vincent-de-Paul
(32).
This remark was later followed up in a letter dated March 17, 1841, in which
it was noted:
The business of Catholicisme is being monopolized by a legitimist
faction, which, using all means imaginable, seeks to worm its way into
everything ... (33).
During the same month, March 1841, the Archbishop of Lyon, Monseig-
neur Bonald, came to Paris, where he had a meeting with Cauchy. At this
meeting, the archbishop sought to persuade Cauchy to renounce the under-
taking, and at the same time, he spoke with Monseigneur Affre about the
association (34). Paralleling the archbishops's initiatives, the Propagation de
la Foi addressed itself directly to Rome and, as a result, secured the Pope's
support. In the meantime, Cauchy continued to struggle vigorously. In order
to counter the offensive now being launched by the Propagation de la Foi, he
 11. The Legitimist Mathematician 183

wrote a letter to Father Roothaan, the Vicar General of the Jesuits, in May
1841; he defended Catholicisme en Europe against its critics, explaining that
this organization's purpose was complementary to and not in competition
with that of the Propagation de la Foi. The Jesuits had, after all, initially
supported Monseigneur Gillis' work. In any event, Cauchy's letter arrived too
late; the Pope's decision in favor ofthe Propagation de la Foi movement was
irrevocable, and Father Roothaan refused to intervene. Thus, this enterprise to
which Cauchy had, as usual, given so much of himself, ended in defeat (35).
It did not take long for Cauchy's involvement on the side of the Jesuits to
hurt his scientific career. His views and persuasions not only put him on a bad
footing with the government but with his colleagues at the Academie as well.
The scientists at the Academie were generally very much attached to the
Universite. In this regard, Cauchy suffered a cruel experience in 1843, just as
the agitation of the Catholics in favor of freedom of education reached a pitch.
This agitation was but one episode in the long educational struggle that pitted
the partisans and adversaries of the Universite's educational monopoly
against each other and extended Ol'er the entire 19th century. Catholic
secondary education grew considerably and, as matters stood, illegally. In fact,
by the end of 1843, about half of the students attending the colleges were
enrolled in those operated by Catholics, so that in departments everywhere in
France the church-operated schools vied effectively with the state-sponsored
colleges. The formation, in July 1843, of the Comite de Defense de la Liberte
d'Enseignement revealed the ambitions of the Catholics who, at this time, had
Guizot's ear, who was favorable to the notion offreedom of education, at least
under certain conditions. One particular aspect of the dispute, the aspect that
concerns us here, had to do with higher education. Some fairly limited
experiments at this level, such as the Ecole Normale Ecclesiastique of the rue
des Postes in Paris and the conferences of the Institut Catholique and the
Cercle Catholique, were grounds for a belief that Catholic university
education would also be created.
Although characterized by many conflicting tendencies, the Catholic
campaign became increasingly violent and, rightly or wrongly, the Jesuits were
accused of having inspired some of the more harmful pamphlets. One of their
main targets was the College de France. Characteristically, the students of the
Latin Quarter, often republican and always anticlerical, came to hear and
applaud the words of the three oracles ofliberal thought, Mickiewicz, Quinet,
and Michelet. In much the same way, reactionary professors were booed and
hissed and, indeed, in some cases, were kept from speaking altogether. A
number of rightwing journals decried the closing of courses thought to be
seditious. As if to crown this campaign of disparagement, a particularly
venomous work, entitled the Monopole Universitaire, suddenly appeared on
the scene in Lyon early in 1843. Although it was attributed to the Canon
Desgarets, it was, in fact, written by the Jesuits. At this point, Michelet and
Quinet decided to take the offensive. Both offered lessons and discussions on
the Jesuits, and in these presentations they each denounced the system of
184 11. The Legitimist Mathematician

Ignatius of Loyola. Michelet's course on the Jesuits was held from April 26
through the first of June, while Quinet's was held from May 10 until June 14.
On the whole, these two courses of workshops created quite a stir, and despite
a few incidents, professor Michelet and Quinet scored a triumph (36).
It was precisely at this juncture that Cauchy decided to present himself as a
candidate for a chair in mathematics at the College de France. The vacancy
had been created by Lacroix's death on May 25, 1843. By law, an assembly of
the professors of the College would meet and agree on a first candidate and this
choice would be presented; similarly, the Academie des Sciences would present
a second candidate. These presentations would then be submitted to the
Minister of Public Instruction, who would then appoint the new professor.
Generally, the Academie would confirm the choice made by the assembly of
professors, and the minister would simply ratify the decision made by the
professors and confirmed by the Academie.
On June 11, 1843, the assembly ofthe professors of the College de France
met, with three candidates to be considered for the vacant chair in
mathematics: Libri, Cauchy, and Liouville. That Cauchy would win the
election seemed a foregone conclusion, because not only had several voting
scholars assured him of their support, but Liouville had also declared in his
letter of candidacy that should the assembly of professors choose Cauchy, he
would be the 'first to applaud that choice'. Moreover, Cauchy seems to have
been convinced-quite wrongly, as it actually turned out-of Libri's good
will (37).
The assumption that Cauchy would win the election may have been
reasonable, but it was made without taking political considerations into
account, because the dispute about academic freedom and, more precisely, the
quarrel about the role of the Jesuits could not but influence the election at the
College de France. By offering himself as a candidate for the chair, Cauchy
seems to have been determined to ignore the event~ that were taking place.
Nevertheless, he was aware of the problems posed by his refusal to take a
political oath should he be elected, and, with this in mind, he initially
abandoned his goal. However, for some reason, which we have been unable to
determine (but which, quite likely, was that he did not want to give grounds for
the belief that he had withdrawn out of fear of being beaten), he finally decided
to confirm his candidacy (38).
The assembly of professors took place on June 11, and during the meeting,
there was an open discussion at which a number of professors spoke. The vote,
however, was postponed to the following Sunday, June 18, 1843. From a
standpoint of scientific scholarship, the matter was clear: Liouville, himself an
able mathematiciari, was disposed to defer to Cauchy, whose merits he
acknowledged. Libri had already given clear and certain proof of incom-
petence as a mathematician when he had replaced Lacroix. Moreover, certain
persons-the historian Michelet, in particular-were already aware of Libri's
embezzlements (39). But, the matter was essentially political: it was absolutely
necessary that Cauchy be defeated because he was the Jesuits' candidate. Libri,
 1l. The Legitimist Mathematician 185

of course, had once been Cauchy's protege, as well as a friend of the Jesuit
priest Moigno. Now, however, he presented himself as the determined enemy
of the Society of Jesus. Accordingly, at election time, he published two articles
in the Revue des Deux Mondes, that were very harsh on the Jesuits (40).
Furthermore, according to Michelet's journal, Libri never ceased pestering
and worrying him about the election, swearing that his defeat would mean a
victory for the Society of Jesus, pure and simple.
On June 17, 1843, the eve of the election, Libri wrote a letter to Letronne in
which he declared that nothing could stop him from 'keeping up his war'
against the Jesqits and if, indeed, he should be rejected for the chair in
mathematics, then 'the Jesuits would sing out their victory in their newspapers'
(41). Meanwhile, in order to allay any suspicions that might occur to his
competitor, Libri, on June 13, 1843, took part in the commission of the
Academie that was to evaluate a paper by Jacques Binet. Cauchy and Sturm
were also on this commission, and moreover, Binet, the author of the study
to be evaluated, was Cauchy's close friend and a known supporter of the
Jesuits.
On June 18, 1843, the assembly of professors proceeded to nominate a
candidate after having rejected a motion by a member of the assembly who
asked that the discussion of the candidates' qualifications be reopened in light
of the fact that Cauchy had written a letter to Letronne in which he gave
assurances that if he should be elected to the chair 'the government would have
no cause to fear any serious obstacles' and that his lectures would not be the
subject of any disruptions (42). On the first ballot, with 24 voting professors
present, Cauchy received 3 votes against 12 for Libri and 9 for Liouville; on the
second ballot, Liouville received 12 votes with 11 for Libri and 1 for Cauchy;
on the third ballot, Libri carried 13 votes as opposed to 10 for Liouville and 1
for Cauchy.
That the professors at the College had chosen Libri caused a scandal in the
mathematics community. The following day, June 19, 1843, Liouville sent a
letter to Letronne, the Administrator of the College de France, in which he
resigned his position as an adjunct professor, declaring that he was 'deeply
humiliated as a man and as a mathematician by what took place yesterday at
the College de France' (43).
Cauchy attended the Academie's meeting of June 19, 1843 and made a
statement in which he asserted that he 'would never consent to the Academie's
placing his name on any list of candidates for a chair in mathematics unless, on
the one hand, no serious obstacles to his candidacy were raised if such
[obstacles] were not related to science; and, on the other hand, the candidates
themselves, being desirous of giving a new indication of their esteem for
their former teacher, should endorse that candidacy and fully consent to it'
(44).
In any event, the matter was settled. The geometry section of the Academie,
which received the candidacies, announced to the Academie that only Libri
had been presented. Cauchy and Liouville had renounced their candidacy. So
186 11. The Legitimist Mathematician

declared to the Academie that they were formally opposed to Libri's candidacy
(45). On July 3, just before the election of the Academie's candidate, Cauchy
made a final statement in which he recalled his statement of the preceding week
and the 'report from the geometry section that alleges that I stood aside and
did not offer myself as a candidate' (46). In spite of this final statement by
Cauchy, Libri obtained only 13 votes out of the 45 that were cast, because 28
were blank, 3 were for Cauchy, and 1 was for Liouville. In any event, Libri had
an absolute majority of the votes and, accordingly, became the Academie's
official candidate. Shortly afterward, a royal ordinance ratified this
choice and appointed Libri professor of mathematics at the College de
France.
This affair left deep scars. For example, not long afterward, a bitter dispute
arose between Libri and Liouville following a report made by the latter on
Hermite's study on the division of abelian functions. At this time, September 4,
1843, Liouville announced that he planned to publish Galois' study 'Sur les
conditions de resolubilite des equations par radicaux' (47). Cauchy, careful to
do nothing that would further exacerbate the matter, kept silent throughout
the dispute, which lasted from August 14 until September 18, through six
meetings of the Academie. A few weeks later, during this same period of bitter
political feelings, the matter of Cauchy's refusal to take the oath at the Bureau
des Longitudes came to a head, with negative outcome for Cauchy. That there
was no geometer on staff at the Bureau des Longitudes presented a situation
that could no longer be tolerated. Accordingly, the Bureau voted on
November 15, 1843, to send a letter to the Minister of Public Instruction in
which it declared that the 'present state of affairs cannot continue without
doing real harm to the development of the astronomical sciences' (48). The
minister responded by inviting the Bureau to proceed to make a new
appointment as 'the previous one could not be approved and was thus null
and void because the candidate refused to fulfill an obligation imposed by law'
(49).
Cauchy promptly reacted to this decision by writing an open letter to the
president of the Bureau des Longitudes in which he rejected the minister's
arguments and asserted that for four years he had never received an official
letter enjoining him to take an oath (50). At the same time, he asked that he be
allowed to attend the Bureau's meeting on December 6, 1843, in order to give
an oral explanation of his position. However, the Bureau, no doubt angered
by Cauchy's publication of his letter to the president, refused to receive him
and, accordingly, proceeded to elect a new mathematician. The outcome
of this election was that Poinsot, Cauchy's enemy, took his place at the
Bureau.
Cauchy's double defeat, at the College de France and at the Bureau des
Longitudes, was but one sudden and unfortunate turn in the ongoing
confrontation between the partisans and adversaries of the Universite's
monopoly. Since his return from exile, Cauchy had very unwisely taken part in
the Catholic offensive. Now, deprived of all official positions except his seat in
 11. The Legitimist Mathematician 187

the Academie, he gave up active scientific research in order to throw himself

into the political battle.
The question of freedom of education had, in effect, been brought before the
Chamber of Deputies where those in favor of the Universite's monopoly were
in a majority. The stakes were high, because the real issue was whether the
state or the church would control the intellectual development of the leading
social classes. To the extent that this was so, the Jesuits could not fail to have
great concern, because, within the span of a few years, they had created some
74 academic establishments across the country, even though they were not
authorized to do so. Accordingly, those who were against the Jesuits
demanded that the government expel them once and for all.
The defenders of the Jesuits mobilized, and Cauchy was among the first to
do so. He took part in a kind of public relations campaign that was organized
by Father Ravignan, a Jesuit priest who in 1837 had replaced Lacordaire as
preacher at Notre-Dame and whose conferences and retreats at the Abbaye-
aux-Bois Cauchy had assiduously attended since 1841 (51).
All in all, Cauchy'S contribution to the Jesuit cause at this juncture was
rather modest. He addressed himself to those scholars with whom he was well
acquainted in a pamphlet entitled Considerations sur les Ordres Religieux
Adressees aux Amis des Sciences. Published in March 1844, this work touched
on the main arguments that were likely to appeal to men of science: the general
spirit of sacrifice that motivated the religious orders, the Jesuits civilizing
mission in Paraguay, the role that their colleges and students had played in the
development of science, the value of Jesuit scholars, and the blindness of their
enemies, etc. Shortly afterward, he distributed a short work entitled M emoire a
Consulter Adresse aux Membres des Deux Chambres. In this work of only a few
pages, he summarized the arguments that he had explored in the preceding
pamphlet. In 1844, Cauchy published a third tract entitled Quelques Rejlexions
sur la Liberte d'Enseignement. Here, after having recalled the promises made in
the Charter of 1830, he affirmed the necessity for religious instruction within
the framework of a good education, and then proceeded to give a comparison
between scientific and religious instruction. He condemned Lamennais'
neocatholicism and in conclusion exposed the inconveniences and shortcom-
ings of the Universite's monopoly.
The discussion of freedom of education that took place in the Chamber of
Deputies ended inconclusively. However, Guizot had to take account of the
opinions that had been expressed during the discussion. The Jesuit activities
had been denounced, and there followed a long series of negotiations with the
Pope with the outcome that a number of Jesuit institutions were closed and the
Society of Jesus was disbanded in France. None of this, however, kept the
Jesuits from discretely pursuing their activities until the Revolution of 1848.
The measures that were taken, with the approval of Pope Gregory XVI, as
well as the support of Monseigneur Affre, the Archbishop of Paris, against the
Jesuits were a stunning blow to the their supporters, who now had to keep
quiet. At the same time, however, Montalembert, operating under the broad
188 11. The Legitimist Mathematician

cover of freedom of education, was trying to gather various Catholics into a

Catholic party, in which stalwarts and traditionalists such as Cauchy would be
on the fringes.
Cauchy continued to pursue his charitable works after the eventful years of
1843 and 1844. However, he did so without any pretense of steering them in the
direction of his personal persuasions. While actively working with the Societe
de Saint-Vincent-de-Paul, he tried to use his fame as a person of science to
advance pious designs. As Valson stated:
When it happened that he had to overcome a difficulty, an obstacle, that
would have deterred a less courageous person, it was to his fellow
scholars and scientists at the Institut that he would first turn for help.
There, he would invariably find the means of exerting his influence, and it
must be said to the honor of that illustrious body, that they never faiied
or let him down. With tremendous warmth and perseverance he went
about his task. By using diplomacy, he was able-sometimes through
mere sympathy for the cause that he was so ably defending; sometimes
through his own friendship with an individual-to win over this one, to
convince that one, and so obtain first a few signatures alongside his own.
As time passed, the number of signatures would increase, and in the end,
everyone would have signed. In this way, he associated the Academie-
and often the entire Institut-with his own worthy efforts and good
works (52).
Thus, he supported the Societe de Saint-Regis, an organization to which he
belonged. Founded in 1826 by Jules Gossin, this group worked to obtain civil
and religious marriage for couples who were living out of wedlock and to
legitimize the offsprings of such unions. In 1844, Gossin had entered a
competition in statistics in which he had presented statistical data on persons
who had recourse to the Societe de Saint-Regis; and he had received an
honorable mention from the commission in May 1846 (53). This initial success
prompted Gossin and Cauchy to seek the Academie's help in obtaining an
amelioration ofthe fiscal laws relating to the acts required for the celebration
of marriage. Cauchy drafted a petition that he circulated among his colleagues
and that was designed to stir 'the concern of the government and public
authorities about a charitable organization known as the Saint-Regis' (54).
There were 174 members who signed the petition, and a committee that was
headed by Portalis and on which Cauchy served was appointed. This
committee, which included men such as Tocqueville, Villeneuve, Villerme, and
Pardessus, proposed to the Chamber of Deputies that the tax laws be
amended. This proposal was supported by Villeneuve, and on June 19, 1846, it
was enacted. Under the terms of this measure, distressed people were
exempted from the stamp and registration fees that were required for any and
all kinds of certificates pertaining to the celebration of marriages.
Encouraged by the success of this undertaking, Cauchy once again stirred
the Institut to take a position, this time regarding the Oeuvre de l'Agonie
 11. The Legitimist Mathematician 189

Irlandaise (work on the Irish agony). Here, too, the initiative had come from
Jules Gossin, the president of the general board of the Societe de Saint-
Vincent-de-Paul. The question now centered on aid for the Irish, who were
then suffering from the terrible famine of 1846. Writing in 1847, Henri de
Riancey recalled:

Early one morning, before dawn, Cauchy hastened to my home. A new

idea, a new approach had occurred to him, and he wanted to act on it
right away. A supplication would be addressed to the Father of the
Faithful. The Sovereign Pontiff would be incited to directly appeal to the
Catholics to snatch a God-fearing faithful people from the jaws of
hunger and death. Members of the Institut would be urged, without
regard to beliefs or [political] persuasions, or even to religion itself, to
sign this supplication. Once the Institut had been canvassed, attention
would be focused on the chambers, the salons, and indeed on any
persons of standing and repute in France. Once the letter had been
drafted, Cauchy would carry it from door to door, certain of success, and
nothing was going to stop him. He was not mistaken; he had not placed
too much faith in the generosity of the French people. Thus, how could
anyone fail to lend himself to such a work of spirit and charity? Within a
few days, Cauchy, weakened and exhausted with fatigue, had collected
hundreds of signatures. These he submitted to the papal nuncio who,
delighted and amazed, speedily sent them off to Rome (55).

Another issue that compelled Cauchy's attention during this time was the
problem of prisons; this question had been debated in the Chamber of
Deputies. Cauchy had served on a jury in the Court of Assizes of the Seine. In
the name of all the jurors, he drafted a study that he published under the
title Considerations sur les Moyens de Prevenir les Crimes et de Reformer les
Criminels. He proposed in this study that the conditions of prison detention be
reformed (56).
As we have seen, Cauchy's exclusion from all scientific institutions and his
inability to get a professorship were direct consequences of his political views.
Not only was he opposed to the regime that came to power on the heels of the
July Revolution, but he also struggled against the Universite's monopoly. This
latter fact, of course, meant that he soon became all but isolated in the
French scientific community; for, by and large, the members of that
community were deeply attached to their alma mater.
It will be seen that Cauchy's virtual quarantining had an altogether
negative effect on the evolution of his scientific work. Here, we note the
particularly negative effects that his exile had on mathematics in France. By
the mere force of the situation, an entire generation of French youths were
deprived of instruction by Cauchy. Moreover, if the eight years of exile are
taken into account, it is possible to take some measure of the harm that
political events inflicted on the mathematical tradition in France.
190 11. The Legitimist Mathematician

The politicizing offaculty appointments was a consequence of the fact that

the French scientific community of that time was overly concentrated in Paris,
where it was under the discrete but ongoing surveillance of the authorities, as
well as in the midst of the lively debates that stirred the political classes.
It can even be said that it was the organization of French scientific life itself
that permitted the ostracism of the country's most important mathematician.
 Chapter 12

Scientific Works from 1838 to 1848

Between October 1838 and February 1848, Cauchy's scientific accom-

plishments were as abundant as during the Restoration Era. However, the
conditions under which he worked were very different from those that had
existed before 1830. Most important, Cauchy was now no longer teaching;
and, by the very nature of things as they stood at this time, he found himself
working on the fringes ofthe French scientific community. He had been away
from that community for more than eight years, and during that time,
considerable changes had taken place. He was still a member ofthe Academie,
des Sciences, attending the weekly meetings during which he was able to
communicate the results of his research. In 1836, the Academie began to
publish its Comptes Rendus H ebdomadaires des Seances, which allowed
members to publish their works quickly. This stimulated interest in Cauchy's
communications on his research. In fact, Cauchy, more than any other
member of the Academie, took advantage of this journal, publishing a note or
memoir each week. Moreover, starting in September 1839, he resumed
publication of Exercices (which had been interrupted since 1836) under the
new title Exercices d'Analyse et de Physique Mathematique. Altogether,
Cauchy presented about 240 notes and studies to the Academie during this
period, of which most were published in the Academie's Comptes Rendus.
Aside from these works, Cauchy published 27 reports during this period.
Furthermore, there were some 40 installments of Exercices containing about
50 studies, along with several articles that appeared in reviews and two earlier
studies that appeared in the Memoires de l'Academie. The articles appearing in
the reviews, like the material contained in Exercices, were generally based
on-albeit sometimes in modified form-Cauchy's communications with the
Academie.
Did Cauchy abuse the publication privileges that were now available to
him? It is clear that he wanted to recover the position that was opening for him
in spite of the ostracism that he was being subjected to for political reasons.
This desire could well have prompted him to increase his activities at the
191
192 12. Scientific Works from 1838 to 1848

Academie, even at the risk of repeating himself (in certain papers). After all, the
Academie was now the only forum, the sole tribunal, in which he could
publicly and regularly participate. However, his exclusion from scientific
institutions and the resulting lack of an established position rather aggravated
a tendency that had been perceptible quite early on, namely, to extend his
efforts in many directions, according to the circumstances. That Cauchy was
so prolific in his research soon aroused criticism and sarcastic comments, first
expressed in the autumn of 1842. The scientific writer of the National, a
republican publication whose hostility to Cauchy probably sprang more from
political grounds than from scientific ones, first broke the silence by publishing
several articles during October 1842. In particular, on October 19, 1842, the
National remarked:

Might there really be a sickness that goes with the study and cultivation
of geometry? Should the noble kind of intelligence that is devoted to it be
subjected to chronic outbursts of a strange fever, which is nothing more
than a sort of algebraical spurt in the form of a new research
paper? In truth, though we may be little inclined to treat serious things
lightly and though M. Cauchy's very name inspires deep respect in us all,
we must ask the question, seeing that an honorable academician
continues to pile one study on top of another, and, in this way, swells the
number of mathematical works beyond all need or accounting. Thus, we
repeat today that algebra, at least as we have grown accustomed to it,
was so different that our surprise should be understandable. Instead of
this carefully nurtured source exuding its precious liquids in a careful
and measured way, we are now confronted with an inexhaustible urn
that overflows its sides, always gushing new tides on which we find there
are floating, in a pell-mell fashion, all kinds of bizarre signs and unknown
symbols. It is perhaps quite natural that we should ask if his fervor does
not alter the purity of all that he does; and if by thus increasing the
quantity of his works so prodigiously, he does not diminish the very
quality of geometry (1).

Convinced that time would vindicate him, Cauchy, as he wrote in a letter to

Moigno, refused to reply to this attack (2).
However, a short time later, Jean-Baptiste Biot published an article on the
Academie's Comptes Rendus in the November issue of the Journal des Savants.
In this article, Biot reproached Cauchy, without mentioning him specifically.
He claimed that Cauchy was threatening the very existence of the Academie's
new publication policy through his excessive prolixity. After having praised
the Comptes Rendus, he continued:

But, it is said, these detailed extracts, written at the whim of the author,
may very well be too verbose and are, in fact, no more than disguised
publications of entire memoirs, whole extensive works. This is very
 12. Scientific Works from 1838 to 1848 193

deplorable, but I shall not pretend that this habit prevails, however, in a
single instance, with one particular author, it has happened. This author,
a geometer and most assuredly a very cleaver and able one at that, has
used the publication privilege to publish whole memoirs in almost every
issue of the Comptes Rendus. These extensive works that he has
published fairly bristle with symbols that have little or no connection
with each other. Frequently, they simply restate the same result several
different times, or they may just present the same idea in several different
forms, so that today, if the author is indeed able to understand their
interrelationships and all the agreements between them-and I have no
doubts that he can indeed do so-then he is probably the only person
who can make any use of them, or can follow the lines of thought that
they follow. This is surely a highly unfortunate situation, a situation that is
regrettable even for the author himself. This said, we come to the central
issues: by what means and by what authority can this grievous situation
be set aright? These essays-and the overhasty manner in which they
were published justify the use of the term "essay" in referring to them-
do indeed contain, despite their capricious diversity, much that is
worthwhile: some very beautiful proofs and arguments, some very
powerful computational procedures, some methods that appear to be
very fruitful in terms of applications ... If only the author would take the
time to follow up on them. Would you deny him the liberty of making
these things known? We should hope that the advice of his friends-and
he does not lack friends; nor are they in disagreement on this point - will
persuade him that he ought to be more mindful of his own interests as
well as those of science and that he might very well lose this useful means
of publishing by continuing to use it in this immoderate way. If he should
agree to devote several months to the completion of the calculation of
planetary perturbations, let us say, then there is no one who would regret
the change in direction by which he will have come to focus himselffully
on this important subject (3).

One particular circumstance not mentioned by Biot in the article probably

prompted him to publicly criticize his colleague. In 1839, Biot had asked
Cauchy if he could explain the phenomenon of rotatory polarization and give
the laws governing this phenomenon by means of his theory of light. Cauchy
accepted the challenge. However, in an article that appeared in the Comptes
Rendus of November 14, 1842, he had to admit that the problem presented
difficulties that he was unable to overcome (4). Although Cauchy still claimed
that he could explain this phenomenon by a new method of analysis, it can
hardly be doubted that this defeat served to convince Biot, who remained a
strong supporter of the emission theory of light, of the sterility of the works in
mathematical optics that Cauchy presented in the Comptes Rendus during the
succeeding years.
194 12. Scientific Works from 1838 to 1848

Cauchy felt that he had to reply to Biot's criticisms; and, thus, without going
into the background of the dispute with his colleague, he expressed a wish that
he presented before the Academie in a note on December 19 and that, he said,
was 'made in the interest of science':

This wish is that, if in the future, the author of this article believes himself
qualified, either on the basis of his experience or by dent of being a friend,
to give advice to a colleague, then he should kindly confine his
observation specifically to the Academie (5).

This response was typical of the method Cauchy always used in dealing with
such attacks: simply remove polemic from the realm of personalities and place
it in a purely scientific domain. Having replied to Biot's criticisms, Cauchy did
not slacken the pace at which he published his research during the following
months (6). Indeed, until 1848-and even beyond that, until his death in
1857-he continued to publish in this way. Thus, much later, in 1869, we find
that Bertrand leveled criticisms against him that were quite similar to those
that Biot had made. In fact, Bertrand observed:

The dangerous tendency to rush into publication was an irresistible

temptation to Cauchy, and quite frequently, it was a stumbling block.
His mind was always active. Every week he would bring to the Academie
his scarcely completed works, as well as plans for memoirs and studies,
and notions that were sometimes downright unfruitful. But, then, out of
all this, a brilliant discovery would crown his efforts. He obliged his
readers to follow, trick by trick, along paths that were often sterile,
meandering, and careless (7).

In order to fully appreciate Cauchy's approach to scholarly publications, it

is necessary to take into account his relations with his fellow scientists.
Abstract research is an endeavor that requires peace and solitary work, but the
many contacts that come by way of teaching and participation in the scientific
community are also beneficial. Now, Cauchy found himself increasingly in
'splendid isolation', which he was rather fond of simply as a matter of personal
temperament. From this point of view, it is interesting to note that he
published very little in the scientific reviews, such as Liouville's Journal de
M athematiques Pures et Appliquees or any of the other publications in which
the other mathematicians of that era usually published. He preferred the
Academie's Comptes Rendus or his own Exercices.
Learned society in Paris had undergone profound changes between 1830
and 1838. The great scientists of the Restoration Era had died one by one:
Fourier in 1830, Legendre in 1833, Ampere and Navier in 1836, and after
Cauchy's return from exile in 1838, these were followed by Prony in 1839,
Poisson in 1840, and Lacroix and Coriolis in 1843. Imperceptibly, in spite of
some brilliant individuals, the center of scientific life shifted from France to
 12. Scientific Works from 1838 to 1848 195

Germany. In Paris, Cauchy, now approaching 50, represented the past.

Generally not much appreciated by his contemporaries, men such as Poinsot
(8), Biot, and Arago-and that more on the basis of his political ideas and
behavior than on scientific grounds-he was regarded, if not as the leader of a
school, then at least as the leader of a line by a new generation of French
mathematicians. An analysis of the commissions of the Academie on which he
served shows that he collaborated with members of all the various areas of
mathematics, with the exception of those in geography and navigation.
However, five mathematicians stand out in terms of the frequency with which
they collaborated with Cauchy: Sturm, Liouville, Poncelet, Binet, and Lame
(9).
Sturm and Liouville were born in 1803 and 1809, respectively (10). A native
of Geneva, Sturm had arrived in Paris toward the end of the Restoration Era
and soon became associated with Fourier and his group. Liouville had
attended the Ecole Polytechnique, where he had studied under Ampere but
not under Cauchy. During that period of time, Cauchy had become
acquainted with these two youths, who attended his courses at the Faculte des
Sciences. Later on, in 1829, when these two young scholars simultaneously
presented studies to the Academie, Cauchy acted as reporter of the evaluation
commissions for both papers. Unfortunately, because of his departure in 1830
he never submitted a report on either work. During the years when Cauchy
was in exile, research on the number of roots of an algebraic equation
contained within a given region tended to bring the three men together in spite
of the differences that separated them, and it can hardly be doubted that the
two young men hailed Cauchy's return to Paris in October 1838 as a major
event for the future of science in France.
Scientific collaboration was especially close between Cauchy and Liouville,
who appears to have been his most frequent associate on evaluation
commissions. Cauchy and Liouville dominated French mathematics during
this era, and they remained close up until the Revolution of 1848 and the
election at the College de France in 1850, which was a consequence of the
revolution. Liouville, a man of democratic leanings and a republican, was
more open than Cauchy and less concerned with questions of priority of
publication (11). Moreover, within the framework of the mathematical
sciences, he was basically an organizer and leader (12).
We should also mention the scientific collaboration between Cauchy and
Poncelet. These two mathematicians were very different and hardly cared for
each other. We have already seen the unpleasant memories that Poncelet had
of Cauchy's reception of his work on projective geometry. Nevertheless, they
frequently worked together evaluating studies on mechanics. In particular,
both participated in the evaluation of the numerous studies that Saint-Venant
presented to the Academie between 1843 and 1847. As for Lame and Binet,
they had long been acquaintances of Cauchy. Although they were late in
becoming members of the Academie-the one first taking his place on the
Academie on March 6,1843, and the other on July 10, following-they soon
196 12. Scientific Works from 1838 to 1848

Joseph Liouville (1809-1882), the best French mathematician of the 1840s, founder of
the Journal de Mathematiques Pures et AppliqUl?es.

began to collaborate regularly with Cauchy. In particular, once Binet had

became a member of the Academie, he became Cauchy's closest and most
faithful colleague on the evaluation commissions.
Cauchy's scientific pursuits enabled him to maintain contacts with a new
generation of mathematicians whose studies he had evaluated. Thus, during
 12. Scientilic Works from 1838 to 1848 197

Jacques Binet (1786~ 1856), French mathematician and Cauchy's friend since 1806.
Photograph by J. L. Charmet, permission of the Academie des Sciences.

this era, he was able to evaluate and report on the works of the younger men
who represented the best hopes of French scientific life, men such as Bertrand,
Leverrier, Bonnet, Saint-Venant, and Laurent. Some ofthese men, while firmly
acknowledging Cauchy's influence, inspired him while he was charged with the
198 12. Scientific Works from 1838 to 1848

responsibility of evaluating their works. Leverrier, for example, was able to

persuade Cauchy to apply his analytical talents and powers to the study of the
perturbations and to planetary inequalities. Meanwhile, Saint-Venant, who
was one of Cauchy's most gifted and closest disciples, perfected and simplified
his theory of elasticity; on the other hand, Cauchy used all his authority in
support of Saint-Venant's election to the Academie. Saint-Venant was
competing for election to the seat in the mechanics section that had been left
vacant by Coriolis' death in 1843. Aside from Saint-Venant, there were other
candidates for the seat: Morin, Fourneyron, and Combes; and Cauchy was
responsible for presenting the section's report to the Academie. The mechanics
section had jointly placed Saint-Venant and Morin at the top of the list of
candidates for the seat. However, Cauchy intervened in support of his favorite
by pointing out that only Saint-Venant had been given the unanimous vote
of the mechanics section. But, this little finishing touch came to nothing,
because Saint-Venant, who made only a very mediocre showing on the
final vote, which was held on December 18, 1843, was defeated by Morin
(13).
In spite of his scholarly activities and his friendly relationship with several
young mathematicians of great talent, Cauchy's position in France's scientific
community was marginal during these years. As a mathematician, he was
admired and respected while, at the same time, his exclusion from the
Universite coupled with his proverbial lack of personal warmth kept him
isolated from young students. Cauchy was completely aware of his isolation,
and it pained him. In fact, he used to define himself-of course, with a certain
bitterness-as an 'old professor to whom youth has listened to for such a
long time and with such goodwill and who still works for the young in the
quiet of his study, still seeking to be of use' (14). He deeply left the loss of
his professorial chairs and regarded it as a loss of personal standing. For
him personally, teaching mathematics was a veritable joy and, in a sense,
his own element and his life (15). He had placed a great deal of hope in the
Ecole Normale Ecclesiastique in 1839, just as he had done with regard to the
Institut Catholique in 1842. This had happened again, and particularly, so, in
1843, when he attempted to gain a chair at the College de France. After his
defeat, he became very dejected. Political requirements had definitely thrown
up an insurmountable barrier between him and those he desired to have as
students.
The sheer number and importance of the papers Cauchy published between
1838 and 1848 prohibit giving a detailed critical analysis here; this body of
work could easily be the subject of one or even several studies in its own right.
Hence, an analysis of Cauchy's methods of work and of the interplay of the
influences to which he was subjected between 1838 and 1848 will be given. By
using this approach, a certain order into this rather considerable body of
written material will be introduced.
The first period lasted from October 1838 until December 1843. When he
returned to France, Cauchy expected to resume his previous place in the
 12. Scientific Works from 1838 to 1848 199

French scientific community, and his entire scientific production was oriented
toward that goal. On the one hand, he sincerely wanted to publicize the
contents of the works he had undertaken during the time he was in Turin and
Prague. This work included investigations on series expansions of functions
and the calculus of limits and its applications to the theory of light, as well as
other topics in analysis that were developed during that time. On the other
hand, he hoped to show by his studies in higher analysis and particularly in
mathematical astronomy that he deserved a position at a scientific institution.
The problem first arose in 1839 with his nomination to the Bureau des
Longitudes. However, as the years passed, Cauchy permitted himself to be
increasingly guided in his choice of research topics by the events of the day; this
is especially true as concerns the subjects that were examined in studies by the
young scholars that he evaluated.
Cauchy's double rejection, first at the College de France and then at the
Bureau des Longitudes, was followed by an entire year in which his scientific
production was reduced. From December 1843 until December 1844, Cauchy
submitted relatively few studies to the Academie, publishingjust three issues of
Exercices in which he examined certain questions from classical analysis,
infinitesimal analysis, and the calculus of variations. Moreover, he rarely
participated on academic committees until November 1844 (16). He was
obviously greatly affected by the quarantine that had been imposed on him.
Beginning in December 1844, he resumed his normal rhythm of publishing.
Now less preoccupied by questions relating to his career, he allowed academic
life to dictate his choice of research to a large extent. Significantly enough, he
became interested in probing his earlier works more deeply; thus, in 1845-
1846, he examined the theory of permutations, while in 1846-1848, he worked
on the theory of functions of a complex variable. This latter undertaking
quickly became a very important area of interest to young mathematicians of
the time.
During the first few months after his return to Paris, Cauchy publicized the
work on light that he had undertaken while in exile and, as we have seen, had
remained unpub1ished except for one part. Thus, until August 1839, he
communicated to the Academie studies almost exclusively on mathematical
optics and on the methods used in mathematical physics. In the spring of 1839,
he published a collection ofthese papers at the publishing house Bachelier (17).
Although he continued to submit papers on these subjects to the Academie
during the following years, he did so at an increasingly reduced pace.
Paralleling this development, he devoted the greater part of the first volume
of the Exercices d'Ana/yse et de Physique Mathematique between Septem-
ber 1839 and June 1841 to problems that had to do with the theory oflight
(18).
In several studies, Cauchy resumed his research work that he had taken up
in Prague on infinitely small motions in a homogeneous system of molecules,
especially on the wave propagation initiated by a perturbation at a given
point of the system. He considered the simple displacements, of molecules
200 12. Scientific Works from 1838 to 1848

corresponding to the real parts of the plane-wave solutions,

c;eUX+vy+wz-st (c;, u, v, wand s are complex constants),
of the fundamental equation of molecular physics:

a~u; = "L., mp ( f(r p~pu;

) - - + (r pf,(r p) - f)) ~pUj) .
(r p I-lp;l-lpj-- (6.13)
p rp rp
All the small motions of the molecules can be obtained by either finite or
infinite superpositions ofthese simple displacements. Cauchy also investigated
the motions in two mutually penetrating systems of molecules, in the hope of
explaining the interactions between the light and the matter (19).
Cauchy focused special attention on the theory of reflection and refraction
that he had developed during 1836 and 1837, while he was still in exile. He
applied a general method for determining the conditions on the boundaries of
bodies to the special case of light by attempting to deduce these conditions
from Fresnel's and Brewster's laws. Cauchy maintained his confidence in the
theory that he had developed in Prague. He assumed the existence of relations
between the elasticities of the ethereal substance such that the vibration ofthe
molecules were perpendicular to the plane of polarization. However, at the end
of 1839, he proposed a new theory: he assumed that the speed of propagation
of the longitudinal vibrations was reduced to zero (20). In this way, he
managed to free optics from a constraining longitudinal wave that he, for want
of a better idea, had momentarily tried to identify with a heat wave (21). But,
this simplification could be carried out only by introducing some additional
new relations between the elasticities of the ether, a substance that had a
negative compressibility.
Cauchy did not probe his theory closely. Moreover, he increasingly
abandoned optics and focused on general methods in mathematical physics.
Still, toward the end of 1842, he once again took up the problem of diffraction,
which he had already studied in 1837. Biot's very harsh article in the Journal
des Savants made Cauchy drop his communications. It is true that he had to
admit defeat over his research on rotary polarization. Cauchy presented no
more communication on optics until the 1848 Revolution (22). The theory of
light, which, since 1838, had occupied first place in his research interests and
efforts, was now abandoned. Thus, the difficulties inherent in Cauchy's
approach to the nature of light sprang from his own stubbornness and zeal.
The biting comments from a colleague put an end to it (23).
Meanwhile, Cauchy continued to improve his developments on the
applications of analysis to mathematical physics. Specifically, the question
that had to be dealt with here was that of solving the linear ordinary
differential equations and linear partial differential equations with constants
coefficients, which occur so often and in such important contexts in
mathematical physics.
In May 1839, Cauchy presented his great study 'Sur l'integration des
equations lineaires' to the Academie (24). In this work, he developed the
 12. Scientific Works from 1838 to 1848 201

method for integrating the linear ordinary and partial differential systems with
constant coefficients that he had created during 1820. He considered first the
first-order homogeneous differential system
dX i
-=a··x·
dt 'J J
(1 ~ i ~ n, 1 ~j ~ n). (12.1)

Let S(s) be the characteristic polynomial det ((a i) - s). If a. is the arbitrary
vector ((Xi) (1 ~i~n) and V(s, a.) = (V;(s, a.)) (1 ~i~n) the linear mapping
S(s) ((aij) - s) -1 (a.), Cauchy showed that the functions

_ C' VieS, a.)e st

X· ( t ) -<..., (l~i~n) (12.2)
, ((S(s)))
are the solutions of Eq. (12.1) satisfying the initial conditions
xi(O) = (Xi (1 ~ i ~ n). Then, he introduced the function

which he called the principal function of the system. The principal function is
the solution of the differential equation of order n,

S(!!'-)0 = °
dt '
satisfying the initial conditions
d dn - 2 dn - 1
0(0) = 0, dt 0(0) = 0, ... , dt n - 2 0(0) = 0, dt n - 1 0(0) = 1.

With the principal function, Eq. (12.2) can be written in the following form:

X;(t) = Vi (:t'a. )0(t) (1 ~i~n).

Cauchy extended the use of the principal function to linear differential systems
with second members of any order. Then, he applied this method to the
integration of certain linear partial differential systems with constant coeffi-
cients. For instance, let the system be

(12.3)

of n}h order relative to t for ~j and satisfying the initial conditions

Ux,y,z, ... ,0) = CPj(x,y, z, ... ), ata ~j(X' y, z, ... , 0) = Xj(x, y, z, .. .), ... ,
(1 ~j ~ p).
 202 12. Scientific Works from 1838 to 1848

Cauchy transformed Eq. (12.3) by Fourier transforms into

f +OO f+oo f+oo

-00 -00 -00 •••
[ Fi ( u,v, w""'at a) (~l(A,p, v, ... ,t), ... ,~p(A,p, v, ... ,t))
- fiCA, p, v, ... , t) ]

x eu(x-l)+v(y-~)+w(z-v)+ .. dA du dp dv dv dw
~~2;-"'=0,

where u = uJ=1, v = v J=1, w = w J=1, . .. and thus reduced the in-

tegration of Eq. (12.3) to evaluating the integral

;: (x,y,z, ... , t) =
<"j
f+ f+ f+
00
-00
00
-C»
00 •••
-00
l'J'(" p, v, ... , t)e"(x-i. l +rl.r-I'I+n l:-rl +·"
<" IL

dAdu dpdv dvdw

(l~j~n)
2n 2n 2n
where ~ are the solutions of the linear ordinary differential system

( a) -
Fi u,v,w""'at (~l(A,p,V, ... ,t), ... ,~p(A,p,V, ... ,t))
_
- fiCA, p, v, ... ,t) = 0 (12.4)
satisfying the initial conditions

0(A, p, v, ... ,O) = ¢j(A, p, v, .. .),

a-
at ~iA, p, v, ... ,0) = Xj(A,p, v, ... ), ... ,
a"r 1 _
at"r 1 UA, p, v, . .. ,0) = t/Jj(A, p, v, ... ) (l~j~p).

In the case of an homogeneous system, Eq. (12.3) witli fi = 0, of the first order
relative to t (nj = 1, for all j), where the coefficients of at~j are constants, Eq.
(12.4) is a first-order linear system that can be solved by using its principal
function 0(t). Cauchy skillfully used this method and reduced the investig-
ation of the linear partial differential system to the determination of the
function
+ 00 f+ f+ eU1X-A)+vIY-!')+w(z-v)+.,
O(x,y,z, ... ,t)=E f -00
00 00

-00 -00 W(A,p,V, ... )

((S(s)))
dA du dp dv dv dw
x------···
2n 2n 2n '
which satisfies for t = 0 the conditions
a a"-l a"
0= 0, at 0 = 0, ... , at" -1 0 = 0, at" 0 = w(x, y, z, . .. ).
The function O(x, y, z, ... , t), which generalizes the function 0(t), was called by
 12. Scientific Works from 1838 to 1848 203

Cauchy the principal function of the homogeneous linear partial differential

system.
Following this study, Cauchy devoted several studies to the case in which it
is possible either to simplify the principal function or reduce the degree of the
characteristic equation S(s) = 0. In particular, he considered the case in which
the characteristic equation is homogeneous. In this situation which he had
considered in 1830, the principal function would reduce to a quadruple
integral; and, if the characteristic equation should be of order two, then it
would reduce to a double integral (25).
This last type of equation, the homogeneous equation, is of great impor-
tance in mathematical physics. In fact, in 1830, Cauchy showed this type of
equation allowed representation of the propagation of a wave in a system of
molecules. The mathematician P.-H. Blanchet, whom Cauchy had cited in his
1830 paper, 'Sur la theorie de la lumiere', and who in 1838 had presented two
studies on the propagation and polarization of motion in elastic media to the
Academie (26), had, on Liouville's urgings, undertaken new investigations on
the equation of the propagation of light.
On June 21,1841, and on July 5,1841, he presented two studies that were to
be evaluated by a committee consisting of Cauchy, Sturm, Liouville, and
Duhamel. In these two works, Blanchet determined the external boundaries of
the wave fronts corresponding to solutions of a given homogenous equation
by using the calculus of residues.
Cauchy immediately set about publishing weekly in the Comptes Rendus
studies and papers on principal functions satisfying homogeneous
equations and on the reduction of such functions. He also investigated
characteristic surfaces, as well as the corresponding wave surfaces (27). He
reproved certain results that had been obtained by Blanchet and published
them right away in the Comptes Rendus, while all the evaluations of Blanchet's
work have not even today been published. However, on August 9, 1841,
Blanchet presented a note in which he showed that the results that Cauchy had
obtained were equivalent to the results that he had obtained, and he was able
to publish this note in the Comptes Rendus (28).
Even afterward Cauchy continued to publish papers on the same subject at
a steady pace, and Blanchet once again requested permission from the
Academie to publish a note in the Comptes Rendus on November 15, 1841.
This time he spoke out against an incorrect statement that Cauchy had made
(29). During the course of the meeting at the Academie, Cauchy recognized
that he had, in effect, encountered a difficulty in his research. Thus, in a note
published shortly thereafter, he abandoned his earlier statement without really
disowning it. The roles of Cauchy and Blanchet were now completely reversed;
Blanchet, now evaluating his former examiner, put an end to the debate by
publishing a letter in the Comptes Rendus on December 20, 1841 (30).
Furthermore, on March 14, 1842, Cauchy, speaking on behalf of the
evaluating committee, issued a report that praised Blanchet's studies and to
which he appended a note on his own regard.
204 12. Scientific Works from 1838 to 1848

On the whole, while recognizing the merits of the studies that had been
submitted to him, Cauchy once more showed himself to be manifestly clumsy
and lacking in tact; for he had, in effect, submerged Blanchet's works-before
he had even evaluated and reported on them - under a mass of his own studies
that dealt with the same questions as the works that had been submitted to him
for evaluation. Cauchy habitually went about things in this way; later, we will
see some more instances. The point of the story is that this time it put him in a
difficult, uncomfortable position.
Following his election to the Bureau des Longitudes in 1839, Cauchy took
an interest in theoretical astronomy, an interest that paralleled his research in
mathematical physics.
We have already seen that in 1831 Cauchy had presented his Turin theorem
on series in the context of celestial mechanics. It is therefore not surprising that
he had quickly published this result in the Comptes Rendus in August 1839,
while he was seeking election to the Bureau des Longitudes. This first proof,
which was, in fact, identical to that of 1831, was followed by a second in 1840;
the 1840 proof was based on the consideration of the mean value of a function
on a circle (31).
The mean value relative to r of a function w(z), which is finite and
continuous with its derivative in the ring ro ~ r ~ R is the value of the
expression
.
11m w(r) + w(Or) + ... + w(on-lr)
, (12.5)
n-++oo n
where 0 is an nth root of unity. By using the mean-value theorem, Cauchy
proved that this expression remain constant inside the ring r ~ r ~ R. Then,
0

he stated that any finite and continuous function f(u), with its derivative in the
disc 0 ~ r ~ R, can be represented by the mean value of w(z) = (z/z - u)f(z);
relative to r if Iu I < r < R. Finally, by substituting the series expansion
"i:.f(z)(u/z)n for w(z) in Eq. (12.5), he obtained a series expansion of f(u) in the
disc 0 ~ lui < R.
Over the following years, Cauchy often had occasion to recall the theorem
of Turin. He had some difficulties in stating conditions for the applications of
his result, because he had not really clarified the concept of an analytic
function of a complex variable. In 1831, he had merely assumed such a
function to be finite and continuous but, from 1839 on, he added the additional
hypothesis of a continuous derivative, which he used in the second proof.
Then, in December 1844, on Liouville's advice, he dropped this supplementary
condition, without clearly justifying this new choice (32).
The Turin theorem could obviously be put to good use in mathematical
astronomy in the study ofthe series expansions of certain functions, such as the
perturbation function. In 1839, thanks to his calculus of limits, Cauchy was
able to deduce the first really rigorous proof of the implicit function theorem,
in the case of analytic functions (33). However, it was the problem of
applications to the theory of differential equations that now absorbed
 12. Scientific Works from 1838 to 1848 205

Cauchy's attention. In December 1840, he published his Prague study 'Sur

l'integration des equations differentielles' in Exercices d'Analyse et de Physique
Mathematique. At various times, he would return to his method of solution
by series. In particular, during the summer of 1842, he presented several
important studies in which he simplified his method of finding series solutions,
and for the first time, he applied his calculus of limits to systems of linear
partial differential equations. Thus, he obtained the first rigorously proved
existence theorem for such equations (Cauchy-Kovalevskaya theorem) (34).
Laurent's paper of August 21, 1843, 'Extension du tbeoreme de M. Cauchy
relatif it la convergence du developpement d'une fonction suivant les
puissances ascendantes de la variable', on which Cauchy wrote a favorable
report on October 30,1843, and to which he added a note with his own ideas-
extended the applications of the Turin theorem (35). Laurent showed that it is
possible to expand a function J(z) in a series

which today is known as a Laurent series, in the neighborhood of an isolated

singular point. This was the first really important result that had been
developed in the theory of the functions of a complex variable by any
mathematician except Cauchy.
Paralleling the development of these very general mathematical theories,
which were applicable to celestial mechanics, Cauchy developed new tech-
niques for the calculations then used in mathematical astronomy. He focused
special attention on the problem of determining the series expansion of the
perturbation function, a topic he had earlier examined in his Turin paper of
October 11, 1831. In several studies that he presented to the Academie in
September and October 1840, he improved Liouville's method of 1836, which
consisted of substituting simple approximate integrals for the double integrals
representing the coefficients in the series expansion of the perturbation
function (36).
Several weeks later, Cauchy sat on the committee that was charged with the
responsibility of evaluating Leverrier's study 'Sur Ie developpement de la
fonction perturbatrice.' No doubt, on this occasion, he established a rapport
with the young astronomer who was trying to calculate a large inequality in
the Pallas motion and who had sought Cauchy'S advice. Cauchy wrote:
Responding to the requests made by this young scholar and scientist, I
wrote my research results on a problem with a solution that might very
well spare astronomers a great deal of tiresome labor and an equal
amount of vexations' (37)
On August 9,1841, Cauchy presented his method of calculating inequalities
to the Academie. This was the same day on which Leverrier presented his
study on the Pallas motion to the Academie. One year later, Cauchy came
back to examine the problem of calculating the perturbation function, this
time using the calculus of residues. This investigation resulted in a new theory
206 12. Scientific Works from 1838 to 1848

of planetary motion that used an (unpublished) expansion of the perturbation

function, in which the coefficients could easier be calculated (38).
The cancellation of his election to the Bureau des Longitudes did not
dampen Cauchy's efforts. In April, July, and December of 1844, he presented
new studies on the perturbation function and the calculation of planetary
inequalities, and in March 1845, he submitted a report on a study by Leverrier,
'Sur l'inegalite du moyen mouvement de Pallas', which was essentially a
follow-up on the 1841 study mentioned earlier. Here, the young astronomer
calculated the great inequality describing the influence of Jupiter on the minor
planet. Leverrier used interpolation formulas that required certain very long
calculations. By using methods that he had developed during the preceding
months, Cauchy easily simplified the calculational problems that had
confronted Leverrier. Thus, he confirmed the correctness of Leverrier's
calculations in two different ways. Cauchy's report and the notes that he
added, and in which Cauchy, as was his habit, literally smothered the work
that he was supposed to evaluate, made as much of an impression as
Leverrier's study (39). Leverrier, encouraged by the success of his work on
Pallas motion, immediately set about investigating Uranus's unexplained
perturbations. This research effort ended in the discovery of a new planet,
Neptune, first by calculations and then by observations in September 1846. As
for Cauchy, he continued to present studies on celestial mechanics at regular
intervals during the following years.
For a while, following his defeat at the College de France and his ousting
from the Bureau des Longitudes, Cauchy slowed the pace of his mathematical
production. He was exhausted by the part he had played on the Jesuits' side in
the struggle against the Universit6's monopoly. After 1845, when he resumed
publishing at his usual rate, his research was more oriented toward abstract
areas and less toward the applied fields that had so long been the focus of his
attention. Thus, he began to follow his own inclinations and, allowing himself
to move with the tide of academic influences, he returned to investigate
problems and questions that he had long neglected.
Cauchy resumed his research on the theory of permutations, an area that,
after his first investigations in 1812, he had all but abandoned. In September
1845 and April 1846, he presented a long series of notes and studies on
permutations to the Academie; these writings were soon assembled into a
major work, 'Sur les arrangements ... ,' which was published in December 1845
and in March and April 1846 in three issues of Exercices d'Analyse et de
Physique Mathematique (40). It was clearly the evaluation of Bertrand's study
'Sur Ie nombre des valeurs que peut prendre une fonction quand on y permute
les lettres qu'elle renferme', which was presented to the Academie on March 17,
1845, and for which Cauchy was to serve as reporter, that prompted research
in this area. In his study, Bertrand used a certain postulate on prime numbers
to prove that any function of n letters assume at least n values (41). Cauchy
sought to establish this same result without using Bertrand's postulate.
As usual, Cauchy did not wait to submit his evaluation of Bertrand's study
 12. Scientific Works from 1838 to 1848 207

until he was able to communicate his research in this area. He gave his report
on Bertrand's paper on November 10, 1845, but went considerably beyond the
framework of Bertrand's theorem in his own work. With great virtuosity, he
set forth a subtle calculus on 'systems of conjugate permutations,' that is, on
the subgroups of permutations that were as yet imperfectly defined. He
obtained some powerful results, such as the theorem known today as Cauchy's
theorem: if p is a prime divisor of the order of a finite group, there is an element
of the group whose order is p (42).
It might be asked whether Bertrand's work by itself could have motivated
Cauchy's important research on permutations during 1845-1846. About the
same time, on September 4, 1843, Liouville announced that he was about to
publish Galois' papers. It can hardly be assumed that Cauchy was not aware of
the difficulties that were confronting Liouville relative to his publisher's tasks
on Galois' works (43). Liouville and Cauchy were on good terms at the
Academie and had frequent chances to exchange ideas. Moreover, Cauchy
referred twice to Charles Hermite, a young protege of Liouville who at this
time knew of Galois' works (44).
Nevertheless, Cauchy's approach to the theory of permutations was quite
different from Galois' approach. Cauchy, who was not at all interested in the
problem of the solvability of equations by radicals, developed a formal
calculus in which structural properties were scarcely clarified. Unlike Galois,
Cauchy remained a prisoner of his calculation techniques. If, indeed, Galois'
papers had prompted Cauchy to reexamine the theory of permutations, an
area that he had not bothered with for 33 years, then it would certainly seem
that Cauchy did not have a precise knowledge of their contents (45).
Starting in mid-1846, Cauchy's work was dominated by the construction of
a theory ofthe functions of a complex variable. He made fundamental progress
in the development of the theory of imaginary integrals, which he had created
in 1825 and which he used in 1831 to provide his theorem on the infinite series
expansions of functions and in determining the number and the nature of the
roots of an equation within a given contour. Cauchy had neglected this theory
for his work on applications, the calculus of residues, and the calculus oflimits;
this neglect, as we have seen, had led to certain difficulties in defining the class
of functions to which these theorems applied. As was frequently the case with
Cauchy, an outside force was needed to prompt him to reexamine the
theoretical framework in which he was operating. It seems that an article by
Anatole Lamarle, which appeared in the April 1846 issue of Liouville's Journal
de Mathematiques Pures et Appliquees, actually triggered these new research
efforts. In this article, Lamarle discussed the hypotheses of the Turin theorem;
like Liouville, he affirmed that the continuity of the derivative of the function
to be expanded was indeed a superfluous condition and that it was only
necessary to assume 'a certain periodicity for the function', that is, that the
function takes on the same value each time its argument is increased by 2n (46).
Cauchy replied with a long article in the August 1846 issue, in which he took
up the question of multiform functions (47). Stressing the principle that 'it is
208 12. Scientific Works from 1838 to 1848

obviously useful to adopt conventions that preserve the property of continuity

for the functions used in this calculus for the longest possible time' (48), he
extended the definition of imaginary logarithms-which had been defined in
1821 in his Cours d'Analyse only in the plane R > O-for the whole complex
plane with a slit, along the negative real axis: if z = re(P+2kn lR ( - n < p < + n),
then log z = log r + p F-l. He thus took up an idea of Lamarle (who, in his
article, had used the notion of a logarithm defined in the complex plane with a
slit along the positive real axis to prove the Turin theorem) by modifying the
domain of the definition in such a way as to keep the logarithmic function
continuous in the neighborhood of positive real values.
At the same time that he was replying to Lamarle, Cauchy began to publish
a series of papers in the Comptes Rendus in which he moved matters much
further along. In these papers, he arrived at the theory of imaginary integrals
by using a completely geometric point of view. In a first note, presented on
August 3, 1846, to the Academie, he gave some general theorems on curvilinear
integrals (S) taken along the oriented border s of a surface S (49). First, Cauchy
supposed that S is changing from S 1 to S2 by a continuous deformation. If the
differential X(x, y, z, ... ) dx + Y(x, y, z, ... )dy + Z(x, y, z, ... ) dz + ... is exact, he
. . ax oz
stated that the mtegral (S) ofthe functIOn k = X (x, y, z, ... ) as + Y(x, y, z, ... ) as

+ Z (x, y, z, ...) :: +... remains constant as long as k remains finite and

continuous when s is moving continuously from the border of S 1 to the border
of S2 (i.e., when the borders are homotopic). If one of the functions, X, Y, Z,
etc., fails to be finite and continuous at some points P', P", pm of S, then we
have the formula
(S) = (a) + (b) + (c) + ... (12.6)
where a, b, c, ... are elementary surfaces around P', P", pm ... , and (a), (b), (c) ...
are singular integrals taken along the border of these surfaces. If these
functions are finite and continuous at each point of S, then
(S) = o. (12.7)
In the case of the exact differential X (x, y) dx + Y(x, y) dy, Cauchy found this

If
last theorem again by means of Green's formula:

1s
ax + Y -;-ds
X:;-
uS
oy =
uS
-ax --dxdy.
saY ax
oy

Cauchy applied Eqs. (12.6) and (12.7) to the theory of complex functions in a
study of September 21, 1846. In this paper, he associated the complex
variable z = x + yJ=1" to the mobile point P in the plane whose coordinates
are x and y. In order to prove the residue theorem, he used the same procedure
that he had used in 1825 and 1826 (50). First, he established that
(S) = 2nJ=1" (12.8)
 12. Scientific Works from 1838 to 1848 209

for k = _1_. aaz, if the mobile point Q relative to Zo is inside the surface S. On
Z-Zo s
the other hand, if J(z) has the single pole Q inside S, (S) = 0 for

k = (J(Z) _ E«(f(U)))) oz.

u-z as
Consequently, this result combined with Eq. (12.8) gives

(S) = 2nJ=1 E«(f(z))) for

This last formula, which can be extended to functions with several poles, is the
residue theorem.
Cauchy took an interest in other 'multiform functions' as well as in abelian
integrals and, in particular, in hyperelliptic and elliptic integrals. In 1843, he
had presented a series of notes on elliptic functions in which his point of
departure was not the inversion of elliptic integrals but simple infinite
products, which he called 'geometric factorials' (51). He also knew that
Liouville was developing a theory of doubly periodic functions. On December
9, 1844, Liouville had stated, but did not prove, a very general theorem
according to which a doubly periodic function that is bounded in the entire
complex plane is constant. At the very next meeting, Cauchy gave the first
proof of this result, which is known today as Liouville's theorem. Using the
calculus of residues, Cauchy extended this theorem to any bounded cont-
inuous function of a complex variable (52).
In 1846, Cauchy's idea for dealing with multivalued functions was very
simple (53): he considered the different ways in which the imaginary variable
could pass continuously from one value to another. The value of an integral
thus depended not only on the endpoints but also on the path selected to
connect them. If the function J(z) to be integrated is uniform, that is, if it takes
the same value for each value of the variable, then the integral along a closed
path depends on the number of revolutions of the path around the poles of the
function. If J(z) has only one pole and if the closed path turns n times around
this pole, the value ofthe integral along this path is nI, where I is the residue of
the function at the pole multiplied by 2n J=1 (the residue theorem). Let the
differential equation
dt = J(z)dz (12.9)

satisfy the initial conditions t = 0, z = a and suppose that J(z) is uniform, finite,
and continuous except in some poles C, C, C etc. Then, t =
If
, IZ J(u)du is a
multivalued function-the complete integral of the differential equation, said
Cauchy-that depends on the number of revolutions that makes the path of
integration around the poles. For instance, for z = a, the paths of integration
210 12. Scientific Works from 1838 to 1848

are closed paths and the values of t are contained in the formula
nI + n1' + n" I" + ... , (12.10)

where n, n', n", etc. are any integers and I, 1', 1", etc. are the residues of f at the
poles C, C', C", etc., multiplied by 2n~. The constants I, 1', 1", ... are called
by Cauchy the indices of periodicity of the path. By means of Eq. (12.10), it is
possible to reduce the determination of the complete integral ofthe differential
equation, Eq. (12.9), to rectilinear integrations, that is, integrations along
rectilinear paths.
Thereafter, Cauchy began to study the case of a function u(z) that is
multivalued, for instance, if u is an algebraic function defined by F(u,z) = O.
The decisive idea, which was inspired by his research on logarithms, was to
derive the value of the function on one point from its value on another point by

f:
joining the two points by a continuous path. So, he showed that the periodicity
of the inverse function of F(z) = f(u)du can be explained if one supposes
that fez) takes the same value after a certain number of revolutions along a
closed path. But, Cauchy restricted himself to general considerations. He did
not investigate how the values of F(z) interchange, a task that he left to his
disciple Victor Puiseux (for the case of algebraic functions) (54).
Curiously enough, the geometric interpretation that Cauchy systematically
used in these notes in 1846 did not mean that he had abandoned his symbolic
theory of imaginaries. It was, in his opinion,just a matter of his simplifying and
clarifying the exposition of the theory, of describing how this calculus actually
goes (55).
During this period, he presented a paper entitled 'Memoire sur les fonctions
de variable imaginaire' in which he continued to adhere to his symbolic theory
of imaginary expressions as presented in his 1821 publication Analyse
Algebrique, while still allowing the possibility of representing an imaginary
number as a variable point in the plane (56). This timorous attitude could at
best be a temporary, provisional position, and Cauchy soon replaced his
symbolic theory by two new theories of imaginaries, one in 1847 and the other
in 1849.
The first of these two theories, the theory of algebraic equivalencies, was
presented to the Academie in June 1847 (57). In this development, imaginary
numbers were regarded as equivalence classes of polynomials with real
coefficients modulo (x 2 + 1) (58). Cauchy replaced the symbolic sign ~ by
the letter i (already used by Euler and Gauss), which he introduced into the
polynomials as an unknown. Without giving any theoretical justification, he
extended his theory of algebraic equivalencies modulo (x 2 + 1) to convergent
series in order to define imaginary exponentials (59).
Cauchy presented this theory following his research on Fermat's last
theorem. This research was begun in March 1847. At the Academie's meeting
of March 1,1847, Lame announced that he had found a proof of this theorem
 12. Scientific Works from 1838 to 1848 211

based on a complete factorization of xn + y" into n relatively prime linear

factors containing complex numbers (60). Liouville immediately expressed his
scepticism by observing, in particular, that Lame's proof assumed, but did not
show, the uniqueness of the decomposition of x" + yn into n relatively prime
complex factors. Cauchy, however, seemed to have been persuaded that Lame
had achieved a fundamental result and quickly began his own investigations.
The problem he faced was that of putting Lame's ideas into better form and
clearing up certain technical difficulties. On March 15, Wantzel stated that he
had established the uniqueness of the factorization of xn + yn into relatively
prime complex factors (61). There followed a veritable race between Cauchy
and Lame, each trying to be the first to obtain a complete proof of Fermat's
theorem. On March 17, both mathematicians registered sealed envelopes
containing the main features of their respective proofs, the customary
procedure in case a question of priority should arise later (62). The following
meetings of the Academie were taken up by their discussions and communic-
ations with each other (63). In this connection, Cauchy investigated the theory
of 'radical polynomials', that is, cyclotomic polynomials a o + alP + a 2P2 + ...
+ an_lp n- l (where P is an nth root of unity and the coefficients are integers),
without achieving any precise results.
At the next meeting, on May 17, Liouville read a letter from Kummer in
which he stated that three years ago he had shown that factorization into
relative prime complex factors was not unique (64). This development dealt a
death blow to the hopes that had been generated by Lame's paper of March 1.
In any event, while Lame recognized his error and kept quiet about the matter,
Cauchy continued to present regularly notes on radical polynomials and their

f-f" .' ;'

, f~/~-V' «'114.,....,'" -<it-y f
a~J""")i""f ,\." -6"",-r",
\t¥11~).\ P~f'p-". 1/~f'F~f.1 ~H
if
< f'"''' /..,s...tr;. ';:; -tf1j .)0.

IV.;1M1/1«"""St-.:ot~ p!...,("I1....,( 1.,1..1'.... Vi. >-;- ... J.... "" ......J.,..D'I. ~tkr')<$.
'ca"'d. .i~j 1{.~f ....I:~{1'J
1T ....~,,<f,,( ~<.-..>- /"""~"" Jorw
....."

,<...w"rt-~~ :"'1'1).(: r")1~4<-, .. ~> ......-I-Jo.-ji<';'I~.,l.. U"1I'f'<1"t1i"'·c-£,.

"'<{~''1\.- .,.~>- ""'-(-;. Ft..~~oC' Jr" ..... f,. ,r,.." .J ""{''if if' .4'r"
".'-'T'cI .~ ,.'''''{cjo,..

Sealed enveloped containing the main features of the proof of Fermat's theorem by
Cauchy, March 17, 1847. The text is written in Italian with Greek letters. Academie des
Sciences de Paris. Published by permission of the Academie de Sciences.
212 12. Scientific Works from 1838 to 1848

decomposition until August (65). Relative to Kummer's investigations, Cauchy

noted in a paper that he presented on May 24, that:
At the last meeting, M. Liouville spoke of M. Kummer's works on
complex polynomials. From the little bit that Liouville said on the
matter, I was persuaded that the conclusions reached by M. Kummer
are, at least in part, the same as the ones to which I have been led by the
preceding considerations. If M. Kummer had investigated the question a
bit further, even if he had been able to raise all the questions and
difficulties inherent in the problem, I should have been the first to
congratulate him on his success, because what we all really want more
than anything else is that the works of all friends of science should join in
determining the truth and making it known (66).
During the following meetings of the Academie, Cauchy expanded his work on
the theory of radical polynomials of ideal complex numbers, without, however,
bothering with Fermat's theorem at all (67). In June 1847, he applied
Kummer's theory of algebraic equivalencies to the theory of complex numbers
(68). This return to the theory of complex numbers clearly showed that Cauchy
was no longer directly interested in Fermat's theorem.
Spread in all directions and presented in a multitude of notes, papers,
studies, and reports, Cauchy'S work between 1838 and 1848 has suffered from
its very mode of publication. In 1857, Biot, expressing the view that Cauchy
would have been one of the 'leading lights ofthe mathematical sciences' had his
life, like the lives of Euler and Lagrange, somehow been able to flow along
without trouble, observed:
By dint of the instability and disorder that events stamped on his
personality and thinking, the influence that he exerted on them [that is,
on the mathematical sciences] will not be completely felt until sufficient
time has passed to evaluate all the consequences [of his scientific works].
Young geometers who have the courage to read his works in detail and
with care will find them to be a mine of ideas, with rich veins of
discoveries and insights to follow through on and to bring up to date
(69).
Today, Cauchy's natural intellectual prolixity, encouraged by circum-
stances, allows the historian of the sciences to follow the development of
mathematical ideas on an almost week-by-week basis.
 Chapter 13
Practices and Principles
in Cauchy's Works

In this chapter, we examine how, relative to the theme of rigor, principles and
practices were articulated in Cauchy's works. The age of Cauchy was a
historical period in which scientific practices asserted an increasing degree of
autonomy vis a vis philosophical reflections. If it is true, on the one hand, that
the philosophers of the first half of the 19th century were less and less interested
in the development ofthe sciences, with Auguste Comte as a notable exception,
then it is also true that scholars-particularly mathematicians-increasingly
insisted that they and they alone had the right to determine the content oftheir
discipline and its standards of scientificality. The divorce between a particular
scientific practice and general philosophical reflection, in a discipline whose
foundations were quite uncertain at the time, gave rise to a blossoming of what
might be termed local and spontaneous philosophies among the scholars and
scientists who had dedicated themselves to this discipline. While these
scholars' philosophies were incomplete, approximate, and often anomalous,
they nevertheless worked to the extent that they could justify current practices
and thus mitigate existing theoretical shortcomings. Thus, the principle of
rigor in mathematics seems to be an excellent example of what we earlier
referred to as local and spontaneous philosophies.
Although Cauchy was not the only mathematician at the beginning of the
19th century to call for rigor in mathematics, he is justly regarded as one of the
main initiators and supporters of the movement to attain that goal. When
Cauchy embarked on his career, France's scientific institutions, which had
been developed, for the most part, during the era ofthe Great Revolution and
First Empire, were beginning to assume a definitive form, a shape that they
would retain almost unchanged throughout the century. In particular, the
professionalization of mathematical activity, together with the gradual
appearance of a standard curriculum within this framework, could not remain
without consequences: little by little both the state and the content of the
mathematical sciences changed.
213
214 13. Practices and Principles in Cauchy's Works

Certainly, mathematics was a great deal more than a means of making a

living for Cauchy: it was a reason for living, as well as a vocation. Working, he
resembled a monk cloistered away in his study:
His tastes and ways were simple and peaceable in the extreme. His study
was small and modestly furnished. Most often, he did his writing at a
small desk-a table, really, without pigeon holes-which, in the
evenings and at night, was lighted by two simple wax candles with
shades. Everything, whether in this small room or in his library, was in
perfect order. He never went out without raising the wood in the
fireplace and extinguishing the fire. He never wanted anything for
himself, being oblivious to himself, as it were, and having no desires to
satisfy (1).
From time to time, in frustration, he would utter a complaint:
There is no let up! No end to it! Accursed problems! Innumerable
calculations. Endless fighting. Signs. Formulas. Theorems besetting me
from dawn to dusk (2)!
Far from living from his work as a mathematician, Cauchy used a
considerable amount of money-especially during the 16 years that he was
kept on the fringes of the scientific community-to guarantee the publication
of his works. Institutional practices, both pedagogical and theoretical, had
been rapidly evolving since the end of the 18th century, and it would be wrong
to overlook the influence that these practices had on a mathematician such as
Cauchy. The advancement and expansion of technological education, a
development that civil and military engineers pushed to new heights, as well as
the development of mathematical physics to a stage that was infinitely more
ambitious than the rational mechanics of previous ages, revived the old
theoretical and practical problem of the relationship between mathematics
and the applications of mathematics. The creation and early success of the
Ecole Poly technique, an institution at which Cauchy was first a student and
later a professor, reveal the will and desire on the part of the political
authorities and the scientific community to solve this problem. Within the
span of a few years, the Ecole was a center producing the best engineers for the
public services, as well as exceptional scholars and scientists who quickly
blazed a path of progress in mathematical physics.
But, even at the Ecole, the problem of defining the relationship between
mathematics and the applications of mathematics was sharply debated. From
the very beginning, the Ecole Poly technique had accorded a special place to
analysis in its educational program, and during the opening years of the 19th
century, this special position for analysis was strengthened, as is witnessed by
the fact that between 1801 and 1806 the time allotted to analysis in the
instructional program, in the first and second years, grew from 16 to 29% and
from 11 to 18% of the total instructional time (3). In spite of a slight decrease in
these percentages during the following years, the leading place of analysis in
 13. Practices and Principles in Cauchy's Works 215

the Ecole's educational program was never seriously questioned. Thus, it

would appear that analysis was the most general and most widely taught
subject at the Ecole, required of all the engineering students, regardless oflater
areas of specialization. Placed at the very head of the official instructional
program, analysis was the heart of the curriculum. Several courses in the area
of applications, such as analysis applied to the geometry of three dimensions
and mechanics depended on it very closely, and other areas, such as machines
and physics, did so to a lesser degree, at least after Petit joined the faculty.
This leading role had at least two important implications for the content of
the analysis courses. On the one hand, it largely determined the structure and
organization of the courses to the extent that the basic idea was always to
proceed from the general to the particular, from algebraic analysis to analysis
applied to the geometry of three dimensions. On the other hand, it had a direct
influence on methodology. Two notions come together at this juncture: some
who were associated with the Ecole-and indeed they were the majority-
thought that analysis should be presented in the most simple and straightfor-
ward way possible, always keeping in mind the possible applications in the
engineering sciences. This approach encouraged systematic appeals to
intuition and to geometric representations. Others-and they were supported
by Laplace and Lagrange-held the opposite view. They felt that it was
necessary to present analysis in a way that was sufficiently abstract so as to
permit its methods to be easily used in quite diverse situations. Cauchy, who, of
course, shared this latter view, went so far as to propose in 1816 that
instruction in mechanics, a subject for which he was responsible, be relegated
to the second-year program in order that the entire first year could be devoted
to infinitesimal calculus (4).
In his reasearch, Cauchy, despite his training as an engineer, rarely
bothered with applications per se. Being little inclined to the physicists' view,
his efforts were really focused, except, of course, for his fundamental
contributions to the theory of elasticity, on the development of mathematical
tools that would be applicable to physics and mathematical astronomy, as well
as on the development of the calculus of residues, the calculus oflimits, and on
characteristics, i.e., on differential operators, etc.
These works illustrate the taste and flair for formalism that is characteristic
of Cauchy's creative works. This is particularly so of his research in
mathematical physics. Cauchy tried to develop various formal calculi with
universal claims, not only in analysis but in algebra also, with his work on the
calculus of permutations, his work in geometric calculus, and his theory of
algebraic keys. On that basis, then, he was an 18th-century mathematician, a
worthy successor of Euler and Lagrange.
Nevertheless, as far as the history of mathematics is concerned, Cauchy
remains firmly connected with the development of rigor. This trend, as we have
seen, legitimized the dominant position that analysis enjoyed at the Ecole
Polytechnique, as well as that generally enjoyed by mathematics at scientific
institutions relative to the applied, empirical sciences. Cauchy went further in
216 13. Practices and Principles in Cauchy's Works

this respect than most. He justified the requirement of rigor in mathematics

by a philosophical and religious theory of knowledge that did not change
significantly during the 46 years of his scientific life.
To understand this, we will examine several texts of a philosophical and
epistemological nature that he wrote at one time or another-introductions
to books and papers, as well as nonscientific discussions, lectures, and
publications. The fact that Cauchy's epistemological concepts did not really
change will facilitate our work.
The philosophical and religious ideas to which Cauchy remained faithful all
his life were well in line with traditional Catholicism and extreme conservatism
as espoused by the Jesuits and, during the Restoration Era, by the Congreg-
ation, to which he then belonged.
Rightfully anointed kings have regained the throne [wrote Haller in
1816]. We will similarly enthrone a rightful, lawful science, a scie,nce
which will serve the Sovereign Lord and whose truth is confirmed by the
entire Universe (5).
Here we have it. Truth: truth was an essential term to Cauchy:
Truth is a treasure beyond value, and no remorse or anguish in the soul
ever stems from its acquisition. The mere contemplation of these
heavenly wonders, oftheir divine beauty, suffices to compensate us for all
that we may sacrifice in discovering it, and the joy of heaven itself is no
more than the full and complete possession of immortal truth (6).
Cauchy, of course, did not define truth:
Here, on earth, truth will never be complete, it will never be wholly
revealed (7).
That could only be encompassed by God. Nevertheless, he did distinguish two
orders of truth.
The first order consisted of philosophical and moral truths. These were
revealed truths, verities 'too sublime for our thoughts and understanding ever
to attain them' (8). In 1811, he stated:
The most submissive [obedient] persons are also the wisest; and by force
of various sophistries, an individual might very well come to doubt the
truths that have been taught to him, but he will not learn any new ones
(9).
Accordingly, it would be far better to leave the teaching of these truths to those
whom the creator of the universe has entrusted with that mission and
responsibility. This mission, according to Cauchy, was entrusted to the priests,
a group whom, J ullien wrote, Cauchy revered (10). All the doctrines other than
the revealed truths as contained in the Holy Scriptures, and as interpreted and
taught by the Church, were false and dangerous. For this reason, then,
religious education was necessary. The Universite's monopoly was culpable
 13. Practices and Principles in Cauchy's Works 217

not because it was a monopoly, but because it planted 'chaos and anarchy' in
beliefs by not accepting religious precepts as the foundation of education (11).
This kind of reasoning, which was quite widespread among the supporters of
the free school movement, provoked mockery and raillery from their
opponents who accused the Catholics, and especially the Jesuits, of wanting to
use the cover of religious education to reestablish the Church's control over
the Universite.
With Cauchy, as with all the counterrevolutionaries, religion was necessary
for the maintenance of the established order, because it served to 'hold man's
passions in check and make him practice virtues' (12). The vain and pernicious
philosophy' of the last century, 'after having overrun the higher classes in
society, then descended into the huts of the poor and there turned the lower
classes into its toys, making these classes the authors of its misery and the
instruments of its crimes' (13). From this stemmed all the evils that beset
the 19th Century. Cauchy was thus being quite specific in 1844 when he
declared:
Unless it be accompanied by a good education, instruction can become
more troublesome than useful ... Of what use is it for the child of a poor
man to learn how to write, if he only takes up the pen to snare the
innocent, to deceive and undermine the good faith of others ... (14).
The second order of truth was composed of scientific truths; these were
'conquered' verities as opposed to 'revealed' truths, which constituted the first
order. The pursuit of truth should be the sole aim of any science' (15). In 1811,
Cauchy paid homage to the efforts that generations had made in increasing the
scope of human knowledge (16), and he regarded the times he lived in as 'an
extraordinary era in which a renascent, ceaseless activity devours all thought'
(17). In spite of remarks such as these, Cauchy did not set forth any distinctive
defining criteria for scientific truths. Indeed, if he regarded exactness as an
'essential and necessary feature of any true science', then he also saw exactness
as a crucial feature of 'the most beautiful creations of the human mind, even in
literature, even in poetry' (18).
It is necessary to contrast Cauchy's religious dogmatism with his epistem-
ological relativism. Moreover, the classification scheme that he used in the
sciences is imprecise. He regarded the sciences only from the point of view that
was important to him, namely, in light of his own religious faith and his
practices and perceptions as a scholar and scientist. In the introduction to his
Analyse Algebrique of 1821, he contemplated what he referred to as the
sciences of reason which, aside from mathematics, included 'those sciences that
are called "natural" sciences in which the only method that can be successfully
used consists of observing the fact and then subjecting these facts to
[mathematical] calculations' (19). In more modern words, he said nearly the
same in 1842 but replaced the term sciences of reason by exact sciences, a term
that included the physical and mathematical sciences (20). In 1833, he set forth
'the [sequence of] steps that one should follow in order to arrive at the
218 13. Practices and Principles in Cauchy's Works

knowledge of truth' in the sciences. The method he proposed was a positive

one:
First, it is necessary to study the facts, to multiply the number of
observations made, and then later to search for formulas that connect
them so as thus to discern the particular laws governing a certain class of
phenomena. In general, it is not until after these particular laws have
been established that one can expect to discover and articulate the more
general laws that complete theories by bringing a multitude of appa-
rently very diverse phenomena together under a single governing
principle (21).
In 1842, the natural sciences founded on observation only were not regarded
as exact sciences; the term no longer carried the same meaning as in 1821.
Mathematics was of no help in these cases.
Scientific truths, being of second order, were inferior to first-order truths. In
fact, a knowledge of the basic essence of things or an understanding of what
created objects were actually composed of was, in Cauchy's view, a divine
privilege. Confined as they are to the here and now, human beings would never
be able 'to penetrate the innermost nature of beings, of things, and would never
clearly discern the secret mechanisms that underly the motion of things' (22).
Being of an inferior nature, second-order truths were to be subjected to first-
order truths; and a scholar or scientist 'ought not hesitate to reject any
hypothesis that contradicts revealed truth. This point is of crucial importance',
Cauchy emphasized (23). Second-order truths simply lay the groundwork for
first-order truths. For this very reason, then, the Christian religion is so highly
favorable to the advancement of the sciences and to the development of the
most noble faculties of our intelligence' (24). Thus, for example, astronomy
brings us closer to God. In fact, at the end of his 'Leyon d'astronomie', Cauchy
gave a description of this 'spiritual elevation', which proceeds from the
observation of the stars rising to the contemplation of God:
We go on. Always keeping a close eye on that brilliantly lighted path,
that great belt that is painted on the heavens even on a dark and cloudy
night. From the purest milk, a whiteness shines forth. In this dazzling
trance, we can see still other suns. We perceive suns in each bank of
clouds and haze. My very soul is smitten by such spectacles, and I am
hushed with awe, adoring that Being whose glorious name is read by
such wonderful traits in the fires of the aurora and on the pavillon of the
skies (25). .
But, a danger always threatens scholars and scientists: the desire to extend
second-order verities at the expense of those of the first order, as the
philosophy of the Enlightenment had sought to do. On this, Cauchy
declared:
The great crime of the last century was that of wanting to raise nature
itself up against its very Author, of desiring to set creatures in a state of
 13. Practices and Principles in Cauchy's Works 219

permanent revolt against the Creator and even to arm the sciences
against God himself, the sciences whose only real aim must be the search
for truth (26).

Mathematicians were not spared the risks, because they might be tempted to
place the propositions of their particular science over first-order truths.
Cauchy not only rejected this platonic interpretation of mathematics, but, in
a famous passage in the introduction to his Analyse Algebrique of 1821, he
asserted that mathematics cannot be applied to historical, political, or moral
questions, insofar as these relate to first-order truths:
Let us cultivate the sciences with true ardor, but without falling prey to
the desire to extend them beyond their proper domains. Let us not think
that we can come to grips with history by means of formulas, nor should
we attempt to base morality on theorems from algebra or from integral
calculus (27).
On the basis of the above remarks, it is clear that Cauchy was taking a stand in
opposition to the use of statistics and the calculus of probabilities in the human
sciences, which had been developing quite rapidly since Laplace's pioneering
works on probability.
Slightly modifying his views, Cauchy later came back to this indictment. In
January 1827, he used Dupin's statistical studies to vindicate the works of the
Brothers of the Christian Schools (28). In particular, in 1845, he presented the
paper 'Memoire sur les secours que les sciences du calcul peuvent fournir aux
sciences physiques et meme aux sciences morales et sur l'accord des theories
mathematiques et physiques avec la philosophie' in which he rehabilitated
statistics (29), declaring:

Geometers [mathematicians] and physicists have sometimes been

accused of wanting to apply the methods and procedures of calculus and
mathematical analysis to the search for all truths. It is beyond question
that there have been exaggerations, just as it is not to be doubted that
there have been abuses of other things, even of numbers. However, it is
only fair to state that in many instances the science of numbers and
analytical methods cap help us to discover the truth or, at the very least,
how to recognize it (30).
According to Cauchy, statistics, in effect, presented 'a rather secure means of
determining whether a given doctrine is true or false, safe or depraved, and
whether a given institution advances or hinders a people and their welfare',
This is so because statistics enables rigorous estimates to be made of facts and
effects.
The diagram below gives a schematic representation of the types of human
knowledge relative to revealed truth. From this representation, it is clear that
in Cauchy'S view the mathematical sciences were to be placed at the opposite
end of the 'table of knowledge' from revealed truth.
220 13. Practices and Principles in Cauchy's Works

A .
~c-o.\'v.'"
0\ c'l!. Observation
,\\.o-r-.

,?-~~\. .
c'l!.
Physical
~~:~~ History,
~
science m~r~
Demonstration politlcs Revelation
/ Mathematics ( \ Religion
~
Truths of second order Truths of first order

Cauchy's table of knowledge (interpretation scheme).

Thus, science reigns supreme in the domain of second-order truths, and

mathematics could not claim that it could be extended to areas beyond the
exact sciences. Yet, on the other hand, the position of mathematics at the other
extreme of the table assured this discipline of its autonomy with regard to both
religion and metaphysics. Moreover, the autonomy enjoyed by mathematics
was greater than that accorded any other science. Strangely enough, on the
basis of his religious philosophy, Cauchy completely isolated the mathe-
matical sciences in the sense that, by his philosophy, mathematics could not
find justification for its propositions and methods in any appeal to meta-
physics. So, after having taken the position in 1811 that mathematics had now
reached its 'highest stage of development and could thus be regarded as
complete' (31), he did not hesitate to change this point of view and to subject it
to serious internal analysis and criticism.
Mathematicians of the preceding era had developed the procedures of
analysis, the method of infinitesimals, the expansions of series, the passage
from the reals to the imaginaries by appeal to reasons based on the generality
of algebra. Only a belief in the indefinite progress of human knowledge and the
unusual outpouring of fundamental results in mathematics could possibly
have justified the use of so metaphysical a principle. Mathematicians at the end
ofthe 18th century and the beginning of the 19th century were acutely aware of
the fact that this principle was highly questionable from the standpoint of
mathematical rigor, although they did not doubt the validity of the various
theories that had been obtained by appeal to that dubious principle. Lagrange
continued to use the method of infinitesimals in mechanics, even though he
would have replaced it by the method of derivatives (offunctions), because he
regarded the latter method as being more satisfactory from the standpoint of
rigor. Similarly, Laplace, though searching for more orthodox methods of
calculating integrals whose values he had determined by passing from the reals
to the imaginaries, did not doubt the correctness of the basic results that he
had obtained by means of the latter method. Cauchy's position, however, was
quite different. It was not enough simply to replace old methods that were
based on a kind of principle of continuity by new methods that might be more
 13. Practices and Principles in Cauchy's Works 221

mathematically rigorous. It was, in his view, necessary to determine precisely

the domain of validity for methods and results that the mathematicians of the
Age of Enlightenment had improperly (and, quite frequently, erroneously)
extended. Cauchy, a man whose philosophical ideas were formed in reaction to
those held by 18th-century thinkers, the philosophes, essentially had no
reason to place any confidence in metaphysics, an area from which he
completely detached the mathematical sciences. He similarly had good reason
for restricting the domain of validity of mathematical propositions, which he
regarded as being second-order truths, and he struggled forcefully against any
tendency to extend mathematics into sciences where it was inapplicable.
Thus, it was that the confident age of indefinite extension was followed, in
analysis, by a more rigorous age of voluntary restriction. It was, then, in this
spirit that Cauchy undertook a work on the foundations of analysis. This
proved to be very fruitful, leading not only to the articulation of many
existence theorems on series, integrals, differential equations, etc., but also to
the creation of completely new concepts and methods. By strictly defining the
boundaries beyond which the traditional mathematical methods and proce-
dures could not be applied, he opened up a vast and new domain for research,
which he explored by inventing his theory offunctions. Starting in 1814, he
approached the theory of integration from this perspective. We recall that in
connection with his investigation of the rigorous passage from the reals to the
imaginaries in integral calculus he studied the behavior of a simple integral
when the function (to be integrated) was discontinuous at a point within the
interval of integration. In such a case, the theorem that states that the value of
the definite integral is equal to the difference between the values of the
primitive at the endpoints of the interval no longer holds. For this simple
reason, from 1815 on, Cauchy preferred to use Leibniz's definition of the
integral as the sum of the elements that correspond to different values of the
variable rather than the definition that was derived from the primitive function
and had been generally adopted by mathematicians of his day. In the same
way, the restrictions on the theorem concerning the substitution of the order of
integration in a double integral led to the definition of a completely new type of
integral, the singular integrals, and later to the development of the calculus of
residues.
But, it was in Analyse Algebrique of 1821 that Cauchy first elaborated on
rigorous methods, which he defined in the following terms:

As to the methods [used here], I have sought to endow them with all the
rigor that is required in geometry and in such a way that I have not had
to have recourse to the generality of algebra. Reasons of this kind,
although commonly accepted-particularly in the passage from con-
vergent to divergent series, and from real quantities to imaginary
expressions-cannot be considered, it seems to me, as anything other
than proper inductions to be used sometimes in guessing the truth.
Such reasons, however, ill agree with the mathematical sciences' much-
222 13. Practices and Principles in Cauchy's Works

vaunted claims of exactitude. It should also be observed that they tend to

attribute an indefinite extent to algebraic formulas when, in fact, most of
these formulas hold only under certain conditions and for only certain
values of the variables involved. In determining these conditions and
these values and in settling in a precise manner the sense of the notation
and symbols I use, I eliminate all uncertainty. In this way, the different
formulas only describe certain relationships that are always easy to
verify by merely substituting specific numbers for the quantities
themselves. It is true that in order to remain faithful to these principles, I
sometimes find myself forced to depend on several propositions that
perhaps seem a little hard on the first encounter .... But, those who will
read them will find, I hope, that such propositions, implying the pleasant
necessity of endowing the theorems with a greater degree of precision
and of restricting statements that have become too broadly extended,
will actually benefit analysis and will also provide a number of topics for
research, which are surely not without importance (32).
Thus, for the first time, Cauchy gave a foreword on the requirement of rigor in
analysis. If he was, in fact, unable to do all for analysis what Euclid had done
for geometry, then it would be well to keep in mind the stubborn opposition
that he encountered from the conseils at the Ecole Poly technique, as well as
the difficulties involved in the tasks he had undertaken. This aside, he was able
to make the theme of rigor a fruitful one, one whose effects it is easy to note in
the sequence of works that he would produce during the coming years. From
this standpoint, then, Cauchy, along with Gauss, should be regarded as the
leading mathematician of the 19th century.
 Chapter 14

The Final Years 1848-1857

Cauchy was not unhappy about the Revolution of February 1848. The fall of
Charles X in July 1830 had left him dejected, but that of Louis-Philippe revived
his hopes. Would not his former student, the Duc de Bordeaux, now Comte de
Chambord, and for the legitimist party, Henri V, soon take his place on the
throne that had been vacated by the Orleans? That hope was quickly dashed.
However, Cauchy had other reasons, more personal ones, for being happy
about the change in government. His scientific career would now be free and
unfettered; it was no longer necessary that he take a loyalty oath, because it
had been eliminated by the provisionary government of the Republic during
its first days in power.
He felt a kind of jubilation, a deeply felt happiness, during the weeks
following the fall of the July monarchy. Certain of his rights, and without
informing anyone of what he was about to do, he appeared at the very first
meeting that the Bureau des Longitudes held after the Revolution. Now, he
could claim the place that had been denied him on political grounds. This act
was, of course, unprecedented and was highly displeasing to the members of
the Bureau.
The session had not yet begun when M. Cauchy came into the meeting
room, without having informed anyone of his intentions, and signed the
attendance roster. Since M. Cauchy was not a member of the Bureau, the
president, having the unanimous support of the Bureau, requested that
he withdraw and struck his name from the attendance roster. After M.
Cauchy departed, the meeting began (1).
He also played an active part in the preparation for the elections to the
Constituent Assembly, which the provisionary government had announced on
February 24 and were to take place on April 23. Discounting the special case of
the Convention, this would be the first time that there would be elections by
universal sufferage, and this raised the problem of recording the vote. Cauchy
agreed to serve on a commission that was to examine various proposals that
223
224 14. The Final Years 1848-1857

were made to the Academie des Sciences regarding methods of counting the
vote; during April and May 1848, he gave several reports on studies
concerning this question (2).
The Revolution of February 1848 was not an anticlerical uprising as that of
1830 had been. This point was obviously of paramount importance to Cauchy
(3). It should be observed, however, that Cauchy's family-his father and
brothers-did not take the same sympathetic view of these political events; for
this revolution had cost them their positions at the Palais du Luxembourg,
which they had held since the beginning of the century (4}.1t seems that Louis-
Fran<,:ois was unable to adjust to this turn of events and withdrew to Arcueil,
where he died a few months later, on December 28, 1848. He was 88 years old.
Several weeks after his father's death, Cauchy was able to get himself
reinstated in higher education and at the Universite. The oath of loyalty no
longer presented an obstacle to his appointment, and it was, then, merely a
matter of waiting until there was a position available. Leverrier, who occupied
a chair for mathematical astronomy that had been specifically created for him,
switched on October 23, 1848, to the vacant chair for physical astronomy. Was
this done merely to create a position for Cauchy? That would seem to have
been the case, for everything appears to have been well prepared in advance.
Even before the vacancy was declared, the Faculte des Sciences submitted a list
of candidates, which the Minister of Public Instruction rejected as irregular.
However, on February 11, 1849, a second list was submitted; Cauchy was first
choice and Delaunay second. On February 17, 1849, the Academic Committee
approved the choices that the Faculte des Sciences had made, and on March 1,
1849, Falloux, Minister of Public Instruction, signed the decree appointing
Cauchy (5).
Cauchy's reinstatement at the Universite, though something that he had
heartily desired, did not mean that he had renounced his views in any way;
with the enactment of Falloux's law, he took two opportunities to express his
views publicly. In February 1850, he published two essays in EAmi de la
Religion, as well as in separate abstracts, entitled 'Lettres sur la compagnie de
Jesus, adressees a un representant du peuple al'occasion de la discussion de la
nouvelle loi sur l'enseignement'(6). This was a reply to the attacks that had
been leveled against the Jesuits during the debates on the educational law.
Enacted on March 15, 1850, this law established freedom of secondary
education and teaching and instituted a Superior Committee of Public
Instruction. Bishops would serve on this committee as a matter of law, as
would three members of the Institut. Obviously quite pleased with these
developments, Cauchy made a speech at the meeting ofthe general assembly of
the five Academies, which met on May 29 and June 12 to select three persons
who would represent the Institut on the Superior Committee of Public
Instruction. Following a clumsy, ill-conceived attack on the principles and
ideas that had characterized the preceding century and some passionate words
in praise of truth, Cauchy concluded his speech with some remarks that
encapsulated the spirit of Falloux's law:
 14. The Final Years 1848-1857 225

Everybody feels that we have now reached one of those solemn periods
in which we see a society that, though severely shaken from top to
bottom, can nevertheless be saved if good and decent citizens hasten to
contribute to the well-being of their homeland and unite in their
common efforts to attain their goal. It is especially by a wise policy
based on public education and enlightenment that a social order that
has been shaken to its very foundation shall be strengthened and
consolidated (7).

Such a decisively political statement could hardly be expected to find much

support in university circles, where, of course, there was a rather general
animosity to Falloux's law, as this was seen as something that had placed the
Universite under the Church's thumb. Cauchy's position probably had a
negative influence on the election at the College de France several months
later, and we must take this into account if we are to understand the second
defeat that he suffered in the elections for the College.
Libri, the encumbent in the mathematics chair at the College de France,
gave up his position in the wake of the Revolution of February 1848, having
occupied that post since his election in 1843, when he had been selected against
both Cauchy and Liouville. Having renounced his professorial chair, Libri fled
France, afraid of judicial proceedings for his theft of precious books from the
public libraries. Until the outbreak of the Revolution of February 1848,
Guizot, Libri's protector, had always intervened to suppress the proceedings.
But, with the revolution, Guizotfell, and Libri was forced to flee. In April 1848,
Hermite was appointed as special replacement to teach the courses that had
been assigned to the hastily departed Libri.
In June 1850, following the judgement against Libri, who, from a safe
distance abroad, continued to insist on his innocence, the Minister of Public
Instruction called a meeting of the Assembly of Professors of the College de
France in order that this body might express its opinion relative to the
'deplorable event' which had just taken place. The Assembly decided to allow a
'grace period' of several months during which time Libri could return to
France and clear himself of the charges that had been brought against him.
However, the decision would be sustained if he did not return by December 1,
1850. The College de France would regard him as having officially resigned his
chair and then seek a replacement. The Minister of Public Instruction,
however, did not consider himself as being bound by the College's resolution,
and the vacancy was announced. On November 15, 1850, Cauchy and
Liouville offered themselves as candidates for the chair. Thus, it was that the
two most outstanding French mathematicians of that era found themselves
competing with each other. It should be recalled that earlier, in June 1843, the
same situation had arisen and that at that time Liouville had publicly declared
his intention to withdraw his own candidacy in favor of Cauchy. But, it was
Libri who, in fact, had won the appointment, thus creating a great scandal in
the mathematical community (8). It seems that Cauchy declared his intentions
226 14. The Final Years 1848-1857

of seeking this position quite a while before Liouville, who only announced his
candidacy shortly before the election following certain changes that were
made in the program of the chair in question (9). The professors met on
November 18, and, following a reading of a letter of candidacy that Cauchy
had submitted, a letter in which he spoke of some of his works, the assembled
professors discussed the two candidates' qualifications (10). At this meeting, it
was decided that the ballot would be held at the next meeting, which took
place on November 25. Ofthe 23 votes cast, Cauchy received 11 as opposed to
10 for Liouville, with two abstentions (11). Accordingly, a second ballot was
held. This time, Liouville took 12 votes against Cauchy's 11, with a single
abstention. Thus, without any objections, the College selected Liouville as its
candidate.
Cauchy and his supporters were quick to react. The fact that there had been
two abstentions on the first balloting raised some questions about the validity
of the election. Should these abstentions count as having established a
majority, as it was in the view of the faculty assembly? If that was the case,
then clearly Cauchy had failed to get a majority of the votes, since he had only
managed to get 11. However, if the blank votes were disregarded, then
obviously Cauchy had won on the first ballot, since he had received 11 of the 21
non abstaining votes that were cast. At the meeting of December 1, 1850, one
of the professors arose and demanded that there be a clarification as regards
the conditions under which the results of the first ballot had been set aside.
Once this had been placed in the minutes of the meeting, Cauchy's friends were
able to start their offensive.
In their search for material that would support their position that Cauchy
had indeed won the election on the first ballot, his friends were able to find,
with the help of Sedillot, the Secretary of the College de France an example
that dated from 1837. In that case, Letronne had been elected to a position on
the first ballot, and no attention had been paid to the abstentions (12). On
December 6, Desgranges, Quatremere, and Binet, acting with Cauchy's full
consent, sent a letter of protest to Parieu, the Minister of Public Instruction;
Quatremere, the most determined of the three, took the floor in the Assembly
of Professors and demanded 'that the College settle once and for all the
jurisdictional question of whether or not the abstentions should be counted in
the voting'. The real point was, of course, to open a debate on the election of
November 25. This discussion, which set Quatremere and Barthelemy Saint-
Hilaire, the Administrator of the College, against each other, ended inconclus-
ively. The meeting, on Duvernoy's motion, decided to adjourn without making
a decision (13).
Cauchy's three supporters had not mentioned their letter of protest during
the discussions; however, Barthelemy Saint-Hilaire had heard of it, and he
promptly wrote a letter to the minister in which he set forth his own views on
the matter. In the meantime, the minister had received a third letter, this one
sent by Cauchy himself on December 14. Embarrassed, the minister replied to
the administrator that 'it seemed to him necessary that a formal decision was
 14. The Final Years 1848-1857 227

now called for and that the College de France should apprise him definitively
of which candidate had been properly chosen by the professors. Although
Barthelemy Saint-Hilaire confirmed on December 23 that Liouville was the
candidate that had been properly elected, Parieu requested on December 30
that another election be held. He easily justified his decision by requiring an
absolute majority of 13 votes. This number was not based on the number of
professors present at the balloting but on the number of incumbent professors.
That much so, he asserted, neither Cauchy nor Liouville could really claim to
have been properly elected (14). That same day the geometry section submitted
the names of two candidates to the Academie: Cauchy and Liouville.
The administrator now called a meeting of the professors in order that a
new election could be organized. However, Cauchy's attitude remained firm,
and he refused to accept what he called the minister's 'Solomonic judgment'.
He was utterly convinced that he had carried the election of November 25 on
the first ballot. Accordingly, as he explained in a letter to the professors dated
January 4, he refused to go through the election process again (15). In any
event, the election took place as scheduled, and of the 23 votes cast, Liouville
received 16 to Cauchy's 7. Cauchy had, of course, declared that he would
withdraw his candidacy should an election be held. Depressed and humiliated
by what had taken place at the College de France, Cauchy now read a note to
the Academie. Entitled 'Sur l'influence souvent exercee par des circonstances
etrangeres a la science dans la solution des questions qui paraissaient
purement scientifiques, et sur Ie pouvoir attribue, dans une election recente, a
un bulletin blanc,' this note also appeared in a lithographic version (16). The
Academie did not even have to proceed with the selection of a candidate,
since Liouville was the only person to present his candidacy.
In the wake of this affair, relations between Cauchy and Liouville
deteriorated. On March 31,1851, when Cauchy presented a paper by Hermite
on doubly periodic functions, an incident flared up between the two
mathematicians. In his report, Cauchy presented the theory of doubly periodic
functions within the general framework of his own theory of functions. At this
juncture, Liouville referred to the fact that quite a while ago he had given a
general theory on this topic as could be substantiated by notes taken by
Joachimsthal and Borchardt, and the manuscripts of these notes, Liouville
went on, had been registered with the Secretariat of the Academie. Cauchy
replied by citing his own research on the topic, which dated back to 1843 and
by observing that Liouville's theorem, a result that provided the foundations
(for the general theory now under discussion) could be easily derived from
his calculus of residues (17). Eight years had now passed, and Hermite
found himself placed, by pure chance, in the same situation that he had been in
1843. As we have already seen, his first study had given rise to a lively quarrel
between Liouville and Libri shortly after the latter's election to a position at
the College de France (18). After this incident, Liouville and Cauchy continued
to work on the same evaluation committees at the Academie, but there was
now little if any real scholarly collaboration between them. In fact, relations
228 14. The Final Years 1848-1857

between these two mathematicians became so strained that in July 1856

Liouville complained to Cauchy about his rather cavalier attitude. This
complaint stemmed from the fact that Cauchy submitted a report on a study
entitled 'Sur l'integration des equations difihentielles au moyen des fonctions
elliptiques' by Briot and Bouquet, even though Liouville, who was also a
member of the committee responsible for evaluating this work had neither
seen nor read either the study itself or the report (19).
Cauchy's uncompromising views in the political sphere caused him one last
problem shortly after the coup d'etat of December 2. The new regime, like the
one that had been in power before February 1848, quickly came around to
requiring that a loyalty oath be taken by all state functionaries, including
professors at the Faculte des Sciences. However, on May 14, 1852, Cauchy sent
a letter to the dean of his faculty in which he recalled his loyalty to the
Bourbons and declared that he would give up his chair in mathematical
astronomy if he were required to take on oath (20).
Cauchy, in effect, gave up teaching in 1853 and was replaced by Faye. In
December 1853, Leverrier was appointed to replace Cauchy in 1854 (21). But,
as in 1839, there were attempts to find an arrangement. Toward the end of
1853, Marshall Vaillant, who had only recently been made a member of the
Academie des Sciences, wrote a letter favoring Cauchy to Minister Fortoul:
You are aware,just as I am, of M. Cauchy's reputation as a scholar. It is
commonly said by even the most learned persons that he is the greatest
mathematician in Europe. I should also point out that he is a man of
tremendous integrity and exemplary character. He is a very religious
man, alert, learned, and sincere. Everybody admires and respects him.
The reserve that M. Cauchy has been said to show should not be
regarded as hostility toward the Emperor's government. Indeed, anyone
who considers it as such would only be fooling himself. Working
alongside M. Cauchy at the Academie as I do, I can assure you that he
holds His Majesty in the highest esteem and is deeply respectful of the
protection that the Emperor has accorded the sciences and the arts (22).
Finally, Fortoul was able to get Emperor Napoleon III to exempt Cauchy
from the loyalty oath requirement. In this affair, at least, the Emperor showed
himself to be a more generous and intelligent man than his predecessors had
been. This was the first time that Cauchy had ever seen his political
intransigence payoff, and he promptly resumed teaching his classes at the
Faculte des Sciences, where he remained until his death.
After the Revolution of February 1848, Cauchy continued his old habit of
regularly presenting notes and studies to the Academie; however, the pace at
which he published slowly declined. At the end of 1853, he ceased publication
of Exercices, though he continued teaching a course at the Faculte. His
scientific interests, research, and production had always been very broad-
mathematical physics, the theory of light, celestial mechanics, the theory of
 14. The Final Years 1848-1857 229

functions, and so on. Aware ofthe need to assemble his works, which were now
scattered about in a multitude of studies, papers, and memoirs, into a coherent
collection, Cauchy contemplated a synthesis, a kind of central theme, around
which his works could be organized. The question, of course, was not merely
one of making his works more accessible and better known but also of putting
them on a more solid foundation. This project, an undertaking that would
haunt Cauchy until he died, was never finished. Still, it did provide, at least
in a certain sense, the starting point for the definitive Oeuvres completes,
which was decided on nine years after his death, but not actually begun until
1882.
In 1855 or 1856, Cauchy requested help from the Father Superior of the
institution in the rue des Postes, asking that Michel Jullien, who was a young
Jesuit student, be assigned to help him in putting his papers and scientific
notes in order. Jullien later gave the following picturesque account:
I spent several days with him actively engaged in this task. A good
portion of my time was spent searching for the kindly old gentleman's
eyeglasses, beneath the stacks of papers where he would always put
them. His papers were written in such a scrawl that I do not know how
the compositors at the printers will ever be able to decipher them (23).
This work concluded in the development of a manuscipt consisting of several
large notebooks that were compiled by Jullien and that he later entrusted to
Valson for use in his biography on Cauchy (which was written in 1868) and the
Oeuvres completes (24).
From 1848 to 1850, Cauchy presented a considerable number of studies on
mathematical physics to the Academie. A particularly large portion of these
works dealt with theory oflight, a subject that he had neglected since the end of
1842. The research carried out by Laurent, Jamin, Bravais, La Provostaye, and
Desains was praised by Cauchy, as was that of Saint-Venant. These works
probably account for Cauchy's revived interest in that topic (25). As had been
true of his investigations during the period 1829-1842, he approached the
theory oflightfrom two different points of view. On the one hand, during 1848,
he resumed the study of the homogeneous equations of motion in an isotropic
system of molecules. These investigations were the subject of many notes;
however, only the titles of these papers were published in the Comptes Rendus
(26). On the other hand, the preparation of a note on Jamin's study 'Sur la
reflexion de la lumiere a la surface des corps transparents' inspired Cauchy
to write several papers on the simple motions in a system of molecules. These
studies were presented to the Academie at the end of 1848. In these
in~estigations, Cauchy showed that when the modulus of the simple motion
was less than 1, then the ray of light grew progressively weaker; the ray was
evanescent (27). In addition, Cauchy advanced the hypothesis that the
obstructing longitudinal vibrations in the case of reflection of simple motion
gave rise to an evanescent ray that moves along the boundary (28).
230 14. The Final Years 1848-1857

On January 2, 1849, a few days after Louis-Fran'Yois, death, Cauchy

announced he was going to write a treatise on molecular mechanics. This
work, he explained, would be written not only in response to the expressed
wishes of a number of illustrious scholars and scientists but also as 'a kind of
filial piety, since this had also been the wish of a loving father, who all during
his life had combined a real love of knowledge and cultivation of letters with
every virtue and who now is sleeping with the just, removing himself to a better
world.'
In this treatise, Cauchy wanted to set forth the general principles governing
molecular mechanics and then successively apply them to problems on the
theory of light, sound, heat, elastic bodies, etc. (29). Unfortunately, he never
realized this project, but rather contented himself with the publication of
several articles in Memoires de l'Academie des Sciences of 1850. As Valson
noted:
Hardly would he have taken his pen and begun to write before a new idea
would come to the fore, and with an irresistible charm, overcome what
he had been working onjust moments before and lead him to pursue an
important result that he, like a hunter in pursuit of a most exciting prize,
would enjoy in advance (30).
Cauchy continued to work on calculation methods that would be applicable
to celestial mechanics. The Revolution of February 1848 had interrupted a
long series of works that he had begun in September 1847 that dealt with the
determination of the orbits of the planets and comets (31). During the
following years, indeed until his death, he would occasionally present studies
on mathematical astronomy: in July 1849, he discussed the problem of
calculating the errors in the study of planetary motion (32); in December 1851
and in January 1852 he took up the question of expansions of functions into
convergent or even divergent series and their application to astronomy (33); in
May and June 1854, he discussed the problem of transformations of implicit
functions by 'isotropic means' and the use of the logarithmic spiral in
determining the orbits of planets and comets (34); in August 1854, he discussed
the solution ofthe differential equations governing the motion of the stars (35);
and, finally, in April and May 1857, on the very eve of his death, he discussed
the use of regulators in astronomy (36).
However interesting these studies are, they occupy a place of lesser
importance in any ranking ofthe works he produced between 1848 and 1857.
This might seem somewhat surprising because during this period Cauchy held
a chair in mathematical astronomy at the Faculte des Sciences. However, in
reality, he taught quite a few other things in addition to calculational methods
in celestial mechanics.
In an article written in 1869, Bertrand spoke of how Cauchy's first lectures
at the Sorbonne in 1849 progressed:
His first lectures, it must be admitted, completely disappointed the
expectations of his audience, who were more surprised than charmed by
 14. The Final Years 1848-1857 231

the slightly confused variety of subjects discussed. I distinctly remember

the third lecture: it was completely devoted to the problem of extracting
square roots, and the number 17 was taken as an example. Calculations
were performed up to the tenth decimal place, calculational methods
which were well known to the audience, but which Cauchy thought were
new, no doubt because he had spontaneously thought them up the
evening before.
This, of course, was typical of Cauchy's sometimes strange pedagogical style.
However, Bertrand added:
I did not attend any further lectures, and I was quite wrong in that
decision, because, ten years later, these lectures could have provided me
with an introduction to some of this illustrious teacher's most exciting
discoveries (37).
Bertrand's remarks could only have been an allusion to the theory of complex
functions, a theory that Bertrand characterized as having a 'main theme and
appeal that was as subtle and as valuable as his most exciting discoveries'.
Finally, it should be observed that teaching gave Cauchy the opportunity to
impose at least some order on the vast theoretical structure that was begun in
1814 and was finally ready in 1846.
In effect, Cauchy's lectures at the Sorbonne primarily consisted of an
exposition of this theory of functions. The fact is confirmed by two partial
transcripts: the first was written by G. Lespiault (38), probably in the academic
year 1855-1856, and the second by the Norvegian C. R. Bjerknes in the
academic year 1856-1857 (39). The venerable old mathematician could now
exert an influence proportional to his ability: there was a small audience
indeed, between three and five persons in 1856-1857 according to Bjerknes
(40); among those who attended his lectures, however, were some talented
young analysts, such as Hermite, Puiseux, Briot, Bouquet, and Meray, who
would later develop the ideas and concepts that Cauchy had presented to
them. Cauchy's courses at the Faculte des Sciences should be placed on the
same level as those given by Liouville at the College de France, because these
two teachers exerted a parallel influence on the new generation of French
mathematicians. Briot and Bouquet, in their treatise Theorie des Fonctions
Doublement Periodiques, et en Particulier des Fonctions Elliptiques synthesized
the concepts that Cauchy and Liouville had developed in their courses (41).
Thus, until 1857, Cauchy'S work with the theory that he had initiated
advanced in step with the works of his disciples. Valson noted:
When he taught the methods he had developed, he presented them in a
neater, more precise way, carefully coordinating method with basic
principle. It seems that once he had taken chalk in hand, he could explain
his formulas more easily on the board than he was able to do when he
wrote them out in his notebooks. This was so because his mind was so
alert that once it was in contact with his audience, a new liveliness would
232 14. The Final Years 1848-1857

take place. A beam of joy would light up his whole being when the proof
that he was trying to get his listeners to understand suddenly became
clear to them (42).

On the basis of his work as a teacher, Cauchy sensed that it was necessary that
his complex variable theory be put on a clear and rigorous footing. A direct
consequence of this realization was his adaption of the 'theory of geometric
quantities' as a new theory of complex numbers. In the preceding chapter, we
saw that Cauchy had been rather indecisive on this point since, until 1848, he
had made use of geometric representations in order to develop his theory of
curvilinear integrals while, at the same time, holding on to what he referred to
as a 'symbolical theory of imaginaries,' which he replaced in 1847 by a 'theory
of algebraic equivalences'. Although this latter theory may have been
aesthetically appealing, it was inadequate as a basic principle for complex
variable theory. The question had to be settled, because it was now no longer
possible to continue in this indecisive way in an official course (43). In
September 1849, he articulated his new theory of imaginaries (44). Aside from
the works of Bude and Argand, he cited a study by Saint-Venant, 'Sur
les sommes et les differences geometriques et sur leur usage pour simplifier la
mecanique', which had been presented to the Academie on September 15,
1845. As a member of the commission charged with the responsibility of
evaluating this study, Cauchy had read this work closely and discussed it with
the author. In it, Saint-Venant had exhibited a kind of vector calculus that was
very similar to the one that had been published a year earlier by the German
mathematician Grassmann in his Ausdehnungslehre. Thus, Cauchy knew the
principles on which his new theory rested a full two years before his paper
'Memoire sur les quantites g60metriques' appeared. We thus get an idea of
Cauchy's reluctance to use a geometric approach, a reluctance that was so
strong that it could only be overcome by the requirements of teaching (45). As
he wrote, 'some new and mature thought was needed if advantage was to be
obtained by replacing imaginary expressions by the geometrical quantities
whose use gives algebra not only a clarity and new precision, but also a greater
generality' (46).
The evaluation of Hermite's note 'Sur la theorie des fonctions elliptiques' in
March and April of 1851, a work that had been presented to the Academie in
November 1849, and even more importantly, two studies by Puiseux, 'Sur les
fonctions algebriques', which were presented on January 13, 1851, and March
17,1851, prompted Cauchy to focus his attention once again on the theoretical
foundations of his theory.
In his study, Hermite applied Cauchy's methods to the study of doubly
periodic functions for the very first time. It is possible-indeed, quite
probable-that Hermite's study had been inspired by Cauchy's lectures at the
Faculte des Sciences in 1849. In any event, Hermite acknowledged his debt to
the great mathematician by presenting his study to the Academie as an
application of the principles of Cauchy's calculus of residues (47). He
 14. The Final Years 1848-1857 233

integrated around the periphery of a period parallelogram and thus showed

that the sum of the residues of a doubly periodic function reduces to O. From
this, he obtained a decomposition of these functions into a sum of terms that
were proportional to transcendental functions of the form 8(z - Zi) and their
derivatives. He then examined the function 8(x), which he expanded in a series,
and its derivative <p(z)(48). Hermite discussed the contents of his paper in his
courses at the College de France during the academic year of 1849-1850 (49).
But, only on March 31,1851, did Cauchy finally present a report on this work
to the Academie; and, although the report was very favorable, it, as we have
seen, gave rise to a quarrel between Cauchy and Liouville
Thus, it took Cauchy more than a year to submit his report on Hermite's
paper. Cauchy decided to act in this manner once he had made up his mind to
resubmit the report on a study that Puiseux had just presented and that he
obviously liked. He could not bring himself to favor one of these young
mathematicians over the other insofar as they both had applied his theory to
the study of 'multiform functions.' One week after he had submitted the report
on Hermite's paper, he submitted the report on Puiseux's study, 'Sur les
fonctions algebriques'; the paper itself had been presented to the Academie on
January 13.
Puiseux, who had been a young maitre de conferences at the Ecole Normale
Superieure since 1849, had already published several important works by the
time he met Cauchy. He took courses under Cauchy at the Faculte des
Sciences and became his best disciple. Moreover, these two men also quickly
became good friends. No one better fits Valson's description of the 'young
people at the schools' of whom Cauchy was so fond, 'young scholars who were
invited into his study as well as into his parlor, and with whom he would talk
and chat amiably, more like a friend than like a teacher' (50). Puiseux professed
to holding the same religious views as Cauchy and, according to Jullien,
was an excellent and fervent Christian who assisted Cauchy with various
undertakings on behalf of the faith (51).
Using the still somewhat confused ideas of Cauchy's 1846 papers, which
were, perhaps, discussed in lectures given at the Sorbonne, Puiseux showed
how, in the neighborhood of a branch point, the values of an algebraic function
change according to the closed path along which the function is evaluated,
thus forming several circular systems. He then used this result to calculate the
value ofthe different periods of elliptic integrals. Puiseux's study was written in
the careful and rigorous style that characterizes Cauchy's best works (52).
Cauchy knew the essentials of what was contained in this study even before
it had been submitted to the Academie. On January 20, he presented a study in
which he investigated questions that dated back to 1846 and had now been
raised again by Puiseux, and he introduced what he termed 'lignes d'arret'
(lines of stoppage), these being straight lines that cut the complex plane in such
a way that the values of an algebraic function could be separated (53). However,
Cauchy, though tempted, as usual, to append his own work to the work that
he was evaluating, directed his efforts in another direction. Instead of following
234 14. The Final Years 1848-1857

the study of elliptic functions along the lines that Puiseux and Hermite had
taken, he tried, in effect, to make his theory of functions more precise.
In a short paper bearing the very general and nondescript title 'sur les
fonctions de variable imaginaire,' he first set forth the definition of the
derivative or differential coefficient of a function of a complex variable and
then followed up on this definition by giving the conditions for the differentiabi-
lity of a function u = v + wi of a complex variable z = x + yi: Duv = Dyw,
Dyv = - Dxw (54).
Later, on April 7,1851, he defined what he termed a monotypical function,
which is continuous and uniform, and a monogenic function, which has a
derivative at each point. Cauchy's intention here was clear: he was setting
concepts and vocabulary that were needed to define the class of functions to
which his calculus of residues and calculus oflimits could be applied (55). In a
new paper of May 12, he showed in substance that a mono typical and
monogenic function can be expanded into a Laurent series, and its integrals
can be evaluated by the theorem of residues if its singular points are poles
(56).
In a February 1852 paper in which he sought to give the appropriate
hypothesis for implicit functions and especially for those defined by differential
equations in order that they should have series expansions, he also clarified,
connected, and completed his terminology. He replaced monotypical by
monodromic, he further asserted that the derivative of a monogenic function is
continuous and proposed to refer to a monodromic function that is monogenic
and finite as a synectical function. Today, these functions are called
holomorphic (57). He then showed that functions that can be expanded in
Taylor series must be synectical.
As we have seen, Cauchy stopped teaching a few weeks after this last paper
was published, because he would not take a loyalty oath. At the end of 1853,
when he resumed teaching, he published several articles in Exercices in which
he explored elements of the theory offunctions that probably constituted the
basic material for his lectures. We specifically mention here the articles 'Sur les
fonction des quantites geometriques', in which Cauchy gave a very general
definition of the notion of function (58); 'Sur les fonctions continues de
quantites algebriques ou geometriques', in which he restated the definition of
lines of stoppage and monodromic functions (59); and, finally, 'Sur les
differentielles des quantites algebriques ou geometriques et sur les derivees des
fonctions de ces quantites', in which he introduced the concept of directional
derivative (60).
During the following years, Cauchy closely followed the work of young
mathematicians as they went about developing his theory, because he was
frequently charged with the responsibility of examining their works at the
Academie. The most important contributions were undoubtedly those by
Briot and Bouquet on the integration of differential equations, works that
prompted Cauchy to undertake new research on differential equations as a
branch of the theory of functions (61). Accordingly, on April 2, 1855, he
 14. The Final Years 1848- 1857 235

( ::S44 )
Si, d'ailleurs, on pose
( 5) dy
iG = tangl'J,
au, ce qlli revient au meme,
(6) ~ =..!)L,
CO! g StU g

I'equation (4) donnera

D Z= D,ZC05"+.~IZsin,,;
~ COSu + 1 SID cr

puis, en ayant egard :lUX formules

la=cos.,.+isin"',I",_,,=r,
00 tirera de I'equation (7),
(8) D.Z = L"{D,,,Zcos!Of + DyZsio!Of).
II est bon d'observer' que, dans les formules (4), (7), (8), Ie derivees
patJielles DzZ, DyZ varient geoeralement avec la position du point mo-
bife A, et se reduisent a des fOllctions des deux coordonnees rectangulaires
x, y de ce mrhne pOint, ou, ce qui revient au meme, II des fonctions de
l'affixe z. D'autre part, tandis que les coordonnees x, y variel'ont par de-
gres insensibles, Ie point A decrira generalement une courbe continue; et
i par ce point on mime une droite qui forme, avec I'a. e polaire, un
angle.,. propl'e a verifier la formule (5), la direction de certe droite,
appelee tangente, sera ce qU'OD peut nommer la direcLioll de la courbe au
point dont iI s'agit. Si la ligne Mcrite par Ie point A se reduir it ulle droite,
In tangente ne ditIerera pas de cette merne droite. Cela pose, il suit imme-
diatement de la formule (4), (7) ou (8) que la derivee de Z, consideree
comme fonction de z, depend en general, non-seulement de la position d.u
point.A SUI' la ligne qu'i! decrit, mais encore de la direction de cette ligne_
Si cette direction devient parallele it l'axe des x ou a I'axe des y, on
allra
dy=o au dx=o,
et la derivee de Z, prise par rapport 1I:z, sera, dans la premiere hypothese,
(9) J),rZ;
dans la seconc1e hypothese,

(10 ) D~Z = _ iDyZ;

I

Differentiability ofa complex function. 'Sur les differentielles des quantites algebriques
ou geometriques, et sur les derivees des fonctions de ces quantites', Exercise d'Analyse
et de Physique Mathematique, 4, p. 336- 347. Published by
permission of the Ecole Poly technique.
236 14. The Final Years 1848-1857

presented Meray's paper 'Sur les fonctions periodiques monogenes et

monodromes' to the Academie (62). Meray took Cauchy's courses at the
Faculte des Sciences; and, later on, at the request of Saint-Venant and the
Cauchy family, he would do the editorial work on this material for publication
in Cours d'Analyse (63).
Parallel to the development of his theory offunctions, which had now come
to be based on geometric quantities, he developed a calculus of keys, which he
thought would have a certain universality, because, in relation to this calculus,
he noted:
Their use not only allows us to easily introduce certain quantities that
come as a natural result [of the study of this calculus], and to which they
'open the door', so to speak, but they also enable us to resolve efficiently a
considerable number of other questions and problems of a rather diverse
nature (64).
He made particular use of this calculus of keys in presenting the theory of
determinants. He also applied, without any great success, this formalism to the
calculus of geometric quantities. Thus, it was that Cauchy returned to a
formalistic approach at the very time that he was increasingly geometrizing his
theory of functions. This attitude, which is ostensibly so contradictory, is
reminiscent of the attitude that, beginning in 1825, led to the simultaneous
development of the formal calculus of residues and the geometric theory of
imaginary integrals.
When, with the help of Mobius, Grassman discovered Cauchy's works on
algebraic keys and those of Saint-Venant on the same topic, he felt that his own
research had been plagiarized. At the beginning of the 1840s, he had, in fact,
created a certain geometrical analysis involving vector concepts, and it
strongly resembled Saint-Venant's geometric calculus, as well as the calculus
of algebraical keys. Accordingly, after reading a paper by Saint-Venant that
was published in an 1845 issue of the Comptes Rendus, Grassman sent copies
of his Ausdehnungslehre to Saint-Venant and Cauchy in 1847. This was
followed by an exchange of letters between Grassmann and Saint-Venant at
the end of 1847 and the beginning of 1848 (65). Grassmann thus had good
grounds for believing that Saint-Venant, and perhaps even Cauchy, knew of
his geometric analysis. He also sent a protest to the Academie des Sciences,
which was received on April 17, 1854 (66). Accordingly, a commission
consisting of Cauchy, Lame, and Binet was appointed to investigate this
matter; however, it never submitted its findings. Nevertheless, Cauchy was
occupied by this affair until the end of 1856, just a few months before his death.
On December 8, 1856, he asked Saint-Venant to write a note for him on
Grassmann's works and on the correspondence that had taken place during
1847-1848. Saint-Venant submitted this note on December 15 following, and
at the same time, he wrote a letter to Grassmann as a token of his good faith
(67). Based on Saint-Venant's note and letter, it seems that the copies of the
Ausdehnungslehre, which Grassmann sent to the two French mathematicians
 14. The Final Years 1848-1857 237

in 1847, were mislaid by the postal service and that, even ifSaint-Venant knew
anything about the studies and papers that Grassmann had sent in 1848,
Cauchy himself had read nothing by Grassmann. The great similarity between
Cauchy's theory as based on algebraic keys and the theory that Grassmann
had developed thus cannot be explained as a gross plagiarism, an act that
would have been out of character for a person like Cauchy. However, Cauchy,
through Saint-Venant, who did have a vague idea about Grassmann's work,
may have been indirectly influenced at the beginning of his research by
Grassmann's research.
We have seen that Cauchy exercised a great influence on the young
generation of scholars and that, therefore, many of them, Puiseux, Hermite,
Briot, and Bouquet, shared his ideas. Furthermore, he was very active at the
Academie where, until his death, he was busy evaluating a considerable
number of studies, which the secretariat had to demand that the Cauchy family
return after his death (68). He often spoke out at meetings of the Academie,
sometimes doing so to support an opinion of a colleague, to support a scientific
project, or to criticize a study. Thus, in July and August of 1853, he engaged in
a long argument with Bienayme on his interpolation method of 1835 as
compared with the method of least squares (69).
He also engaged in quite a bit of active proselytizing among his colleagues
at the Academie, at least among those who seemed to him to be capable of
being brought back to the faith of their childhood. It seems that he won
Duhamel over. However, as Jullien observed, Duhamel was content to merely
perform the obligatory religious practices, such as attending mass and taking
the sacraments. We will later see that at the Academie relations between these
two scholars grew steadily worse. Cauchy also tried to lead Lame, who had
become 'disillusioned with the world and saddened by his declining state of
health', back to religion. In order to accomplish this, Cauchy asked Jesuit
students to attend Lame's courses in mathematical physics, which were very
poorly attended (70).
On another level, right up until the end of his life, Cauchy continued to
devote himself to activities and projects that were nonscientific in nature and
had now come to absorb his attention to a greater degree than science. He
worked tirelessly, visiting people at all hours of the day or night, seeking to
advance his charitable works (71).
In particular, he was the guiding light behind the work of the Ecoles
d'Orient, which was created in 1856 (72). The Crimean War was concluded by
the Treaty of Paris in March 1856. In exchange for a French-British guarantee
of the independence and integrity of his empire, the sultan had to grant equal
rights to his Christian subjects. The new perspectives that were opened up by
the Hatti-i-Humayoun lighted fires of hope in Cauchy's breast; he saw a
possibility for a new crusade (73). The problem was less one of converting
Muslems to Christianity than of bringing Orthodox Christians, who were now
under the sultan's rule and to whom the Tsar of Russia had sought to extend
his protection, back to the Catholic faith. Tsar Nicholas I's amibition had been
238 14. The Final Years 1848-1857

to pressure the sultan into granting him the right to protect the holy places for
the Greek monks that lay within the sultan's domains. These places were under
the protection of the Latin faiths and the tsar's scheme had made the French
keenly aware of the whole question since 1853.
Cauchy's first idea on this point was to work toward creating an
educational institution along the lines of the Frt!res des Bcoles Chretiennes to
win the people over to Catholicism. Accordingly, he collected a mass of
documents and papers on this question and, after having discussed it with his
friends, sought basic support 'from persons belonging to the highest levels of
society, and particularly from among the ranks of scholars and scientists as
well as from his colleagues at the Institut' (74). A provisional committee held
meetings at his home in the de Bure townhouse in Paris. Cauchy, along with
the archaeologist Charles Lenormant and Father Gagarin, a Jesuit priest,
played a leading role at these meetings. The organizational meeting for the
project was held on April 4, 1856, at the home of Mandaroux-Vertamy (75), and
Marshall Bosquet was selected as honorary president of this project, with
Mathieu from the Bureau des Longitudes as acting president. Cauchy and
Lenormant became vice presidents. Of the many persons who claimed
membership in this group, we mention Armand de Melun, President of the
Administrative Committee; Montalembert; and Falloux; and certain academi-
cians, such as Wallon, a historian who served as Secretary General; de Wailly,
a paleontologist; de Rouge, an Egyptologist, and the mathematicians, Hermite
and Binet; Abbe Lavigerie was director of the project (76). Of all the charitable
works that Cauchy created, the work of the Bcoles d'Orient, which still exists
today, was by far the most successful.
In spite of the time and effort devoted to the sciences and to the Bcoles
d'Orient project, Cauchy continued to work actively with the Society of Saint-
Vincent-de-Paul, an organization from which he had founded the Conference
of Sceaux. He used the income from his chair in mathematical astronomy to
support charitable works in the community, particularly the activities of the
Sisters of Saint-Andre and a young boys' orphanage. Shortly before his death,
he was also concerned about the difficulties that were encountered in setting
up a school of the Freres des Bcoles Chretiennes at Sceaux, and in order to
assure its success, he made some rather 'considerable pecuniary sacrifices' (77).
The last few months of Cauchy's life were painful for him. A rather silly and
utterly useless quarrel took place between him and Duhamel, one of his so-
called converts. Centering on a question of priority, this disagreement clouded
his final scientific activity. On December 22, 1856, in the wake of certain
observations that Bertrand had made with regard to a study that Ostrogradski
had done in 1854, Cauchy claimed that in 1818 he had generalized Carnot's
theorem on inelastic shocks in arbitrary bodies. At the next meeting of the
Academie, Duhamel questioned Cauchy's conclusion by referring to a note
that had been presented in 1832. The quarrel continued well into the following
weeks, until finally Poncelet, supported by Morin, exhibited the general
principles governing the shock of inelastic bodies and recalled studies of his
 14. The Final Years 1848- 1857 239

own that had been written in 1826 and the objections he had raised in 1829 to
Cauchy's results (78). Cauchy, of course, lost the argument; in fact, this quarrel
did nothing to increase his standing, because during the discussion, he
stubbornly refused to concede priority, even though he admitted the
inadequacy of his study of 1828. The attacks mounted against him during these
discussions and 'gave the final days of his life a basic sadness and bitterness
that only his friends were aware of' (79).
According to Valson, during this time, he experienced a feeling of emptiness
(80). In May 1856, his old friend Jacques Binet passed away. Speaking for the
Academie des Saiences at the burial, Cauchy declared:
More fortunate than we are, Binet has now gone to the source of all
Light, to learn the secrets that we ourselves shall one day know by
traveling the road that he has already treaded. Lost in thoughts such as
these, thoughts of higher things, I know I will be forgiven, Messieurs, for
abridging these remarks (81).
Less than a year later, on March 30, 1857, his younger brother Alexandre died,
and Cauchy was profoundly affected by his death. Now aillicted with what he
called 'great rheumatism', he left Paris for Sceaux on May 12, 1857, where, on
his doctor's advice, he was to spend the summer. Little by little, the malady
seems to have weakened its hold, and then suddenly, on Tuesday, May 21,
1857, the symptoms became sharply aggravated. On the following day, he

Cauchy's country house, at Sceaux, near Paris. Cauchy died in this house on May 23,
1857. Published by permission of the Ecole Poly technique.
240 14. The Final Years 1848-1857

received the Last Sacraments, and at about 4 o'clock a.m. during the night of
May 23, 1857, Cauchy died.
Our fears have now come to pass [his daughter Alicia announced later
that same day]. Having remained fully alert, in complete control of his
mental powers, until 3:30 a.m., my father suddenly uttered the blessed
names of Jesus, Mary, and Joseph. For the first time, he seemed to be
aware ofthe gravity of his condition. At about 4 o'clock, his soul went to
God. He met death with such a calm that made us ashamed of our
unhappiness (82).
Thus ended the life of the greatest French mathematician of his times-
scarcely two years had passed since Gauss had died in Germany. A new age
was now opening in the long history of mathematics, an age in which the
leading figures in the mathematical sciences would be Germans. Between 1854
and 1859, Riemann, Weierstrass, and Kronecker came onto the scene on the
other side of the Rhine. Meanwhile, however, in France, there was a
blossoming of works on Cauchy's theory.
Concerned about his works, Cauchy had requested that someone be
appointed to have his unpublished papers edited and published. To the
greatest extent possible, his family followed this request and, accordingly,
entrusted the project to Jullien, and later to Meray, who had taken Cauchy's
courses at the Faculte des Sciences, and then finally to Valson, who in 1876
prepared Cauchy's Oeuvres Completes under the auspices of the Academie des
Sciences.
 Notes

Notes to Chapter 1
1. From the marriage contract between Louis-Franyois Cauchy and Marie-
Madeleine Desestre. Arch. Nat., Minutier central des notaires, Etude CXVIII, 640,
October 13, 1787.
2. Th. Lebreton, Biographie Normande, 1, Rouen, 1857, article on Louis-Franyois
Cauchy.
3. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, Paris, 1868, p. 3.
4. See. Dr. Delaunay, 'Le Parnasse du temps de Napoleon: L. F. Cauchy, correspond-
ant de la Societe des Arts du Mans'. Bulletin de la Societe d'Agriculture, Sciences, et
Arts de la Sarthe, 1949-1950, 2nd fascicule, p. 140.
5. Almanach Royal, 1788.
6. From the marriage contract, Arch. Nat., Minutier central des ntoaires, Etude
CXVIII, 640.
7. The property included a country house, a farm house, and an enclosed area of 12
acres (Archives de la Seine, Mutations apres deces, Ivry-Villejuif, DQ 14858,
December 25, 1848, inheritance of Louis-Franyois Cauchy).
8. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 3.
9. Undated letter from Louis-Franyois Cauchy to his mother residing at Rouen, cited
in C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 13.
10. Archives E. P., VI, 2, a, Cauchy file, medical certificate.
11. Arch. Nat. FIb h 531. Organisation des Bureaux du Ministere de l'Interieur par
ordre chronologique, 1792-1811.
12. The expression is from C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1,
p.14.
13. Archives de la Societe d' Agriculture, Sciences, et Arts de la Sarthe, XIV, B8, quoted
by Dr. Delaunay, 'Le Parnasse du temps de Napoleon ... ' p. 143.
14. During the Consulate, Louis-Franyois Cauchy published Ode latine adressee au
Premier Consul de la Repub/ique Fran(:aise Napo/eon Bonaparte in 1802. During
the period of the First Empire, Louis-Franyois published at least nine pieces of
poetry in honor of Napoleon (see the bibliography in Dr. Delaunay, 'Le Parnasse
du temps de Napoleon ... ').

241
242 Notes

15. See. C. A. Val son, La Vie et les Travaux du Baron Cauchy, 1, p. to. It should be
noted that Vitry, one of Louis-Fran90is' colleagues at the Ministry of the Interior
and Chief of the Bureau of Agriculture, was the uncle of the Marquis de Fontanes.
16. Ibid., 1, p. 17.
17. Ibid., 1, pp. 16-18, and 1. B. Biot, 'Lettre a Monsieur de Falloux', Melanges
Scientifiques et Litteraires, 3, p. 144.
18. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 18.
19. See 1. B. Biot, Melanges ... , 3, p. 144, and C. A. Val son, La Vie et les Travaux du
Baron Cauchy, 1, p. 18. According to Valson, ibid., 1, p. 18, one day, Lagrange told
Louis-Fran90is:

Unless you hasten to give Augustin-Louis a solid literary education, his tastes (for it)
will get swept aside; he will be a great mathematician, for sure; but, he won't be able to
write his native language.

20. Presently the Lycee Henri IV.

21. See the Almanach National, 1802.
22. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 20-21.
23. Ibid., 1, p. 21.
24. Ibid., 1, p. 22.
25. Following is the letter that 1. F. Cauchy wrote to Coulomb, Secretary General of
the Ministry of the Interior, on that occasion [Arch. Nat. AA 63 (167)].

Paris, 14 Fructidor, Year XI of the Republic.

I should be very much obliged to you, Citizen Secretary General, if you would kindly
obtain for me two or three tickets for the general awarding of prizes, which is to take
place at the Institut the day after tomorrow, Sunday, 16 Fructidor. My eldest son, who
successfully participated in the competition between the students of the three Ecoles
centrales, has received a letter from the director of his school summoning him to the
[general] awards ceremonies, and I should like his mother and his brother, by their
presence, to be able to partake in a small part of this honorable ceremony.

If it pleases you, Citizen Secretary General, I assure you of my sincerest and utmost
devotion.

1. F. Cauchy

P. S. Ten tickets have just been sent to the Senate; but guid hoc inter tantos? You can
well imagine that there will not be one left that I can get.

26. Letter from Marie-Madeleine Cauchy to her granddaughter, cited by C. A. Valson,

La Vie et les Travaux du Baron Cauchy, 1, p. 21.
27. Alexandre-Laurent Cauchy was born on March 12, 1792, in Paris. He became
assistant to his father at the Senate in 1810 and, after 1815, at the Chamber of Peers.
In 1816, he entered the Congregation. After the retirement of his father, he was
named by the king, Keeper of the Archives of the Chamber of Peers in 1825. Still,
his father, Louis-Fran90is, remained at the Palais du Luxembourg as Honorary
Keeper. Meanwhile, Alexandre-Laurent was raised to the Bench: first judge-
auditor in 1815, he was soon appointed judge at the Court of Paris and, in 1825,
president of the division (president de chambre). He successively became president
of the division at the Court of Appeals of Paris in 1847 and judge at the Cour de
Cassation in 1849 and at the High Court in 1851. He died on March 3,1857, a few
 Notes 243

weeks before his brother Augustin-Louis. Alexandre-Laurent had married Clem-

entine Blanchet de la Saliere in 1825 and had three sons and four daughters.
Eugene-Fran~ois Cauchy was born on October 16, 1802, in Paris. He entered
the Congregation in 1822. In 1835, he replaced his brother Alexandre-Laurent,
whose assistant he had been since 1815, as Keeper of the Archives ofthe Chamber
of Peers. At the same time, he became maitre des requetes at the Conseil d'Etat. He
lost both of these offices in 1848. Eugene Cauchy was a famed jurist, who was an
expert in international law. He was elected to the Academie des Sciences Morales
et Politiques in 1866. He died on April 2, 1877. Eugene had married Marie
Anthelmine Richerand in 1833 and had one son and two daughters.
Louis- Fran~ois Cauchy's daughter Marie-Madeleine was married to a relative,
Pierre-Louis Guignon, an official of the Cour des Comptes in Paris. Thus,
connections were not formed outside of the central administration in Paris. His
second daughter, Marie-Therese, was born in 1796. She died at an early age.
28. An expression by A. L. Cauchy, taken from a letter dated August 17, 1835, and
cited by C. A. Valson La Vie et les Travaux du Baron Cauchy, 1, p. 90.
29. Histoire de [,Academie Royale des Inscriptions et Belles-Lettres, 2, 14, p. 85.
30. On this occasion, Dinet became the friend of the Cauchy family, his neighbors at
Arcueil (C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 22).
31. A. Fourcy, Histoire de ['Ecole Poly technique, Paris, 1828, pp. 376-379..
32. Ibid., pp. 255-257. In Arch. Ac. Sci., Ampere File, carton 5, folder 101, there is a
detailed table of the analysis course taught to the second division by Lacroix in
1805-1806. This table shows that Lacroix closely followed his Traite Elementaire
de Calcul DifJerentiel et Integral for his teaching.
33. According to the Tableau presentant Ie merite des ell!ves sur chaque partie de
['instruction d'apres les examens interieurs, Cauchy earned the following merits in
the first division: analysis, 20; statics, 18; geometry, 13; map drawing, 5; figure
drawing, 12; general comments: hardworking and outstanding student.
34. Correspondance sur ['Ecole Poly technique, 1, no 6, July 1806, p.227, note. The
geometrical solution of a problem proposed by Malus for the examination during
this same period and published in the same issue of the Correspondance, p. 227, is
simply signed c.; it was undoubtedly done by Cauchy.
35. Ibid., 1, no. 6; July 1806, pp. 188-191. Cited by R. Taton, J;Deuvre Scientifique de
Monge, Paris, 1951, p. 262.
36. Correspondance sur ['Ecole Poly technique, 1, no 6, pp. 193-195 (D.C., 2, 2, pp. 399-
401).
37. Here is the description of A. L. Cauchy when he entered the Ecole des Ponts et
Chaussees: light brown hair and eyebrows, low forehead, Roman nose, grey eyes,
middle mouth, round chin, long face, size: 1,63 m (Archives E. P., VI, 2, a, Cauchy
file).
38. The Cauchy file at the Ecole Polytechnique contains two letters from Louis-
Fran~ois Cauchy to the governor of the Ecole regarding this matter. These letters
are given here. The first letter, dated January 26, 1807, and addressed to Devernon,
the Assistant Governor of the Ecole, states:
Sir

My son has been helped in regaining his health by the special permission that you were
so kind as to have granted him. However, in spite of the very best of care that we have
given him, the illness continues, and I find that he cannot resume his work. I have had
244 Notes

him remain in bed this morning, and I request that you allow him to remain here at
home until he has recovered. I respectfully recognize this new obligation that I have to
you, and I am, sir,

Your very humble and obedient servant,

Cauchy

The second letter, though undated was probably written early in October 1807 and
was addressed to General Lacuee, governor of the school:
My Dear Governor:

My son has completed his examinations and thus his presence at the school is pointless.
I would like to request that he be allowed to take advantage of these last days of the
season to come to Arcueil and improve his health while awaiting news from the schools
to which he applied. I therefore humbly request that you grant him a leave of absence,
sir. This, I know, is a new request, and I shall not be less grateful to you for granting it
than I have been for those that you have granted in the past. Please be assured of my
deepest respects,

Most sincerely,

Cauchy

39. See A. Debauve, Les Travaux Publics et les lngenieurs des Ponts et Chaussees
depuis Ie XVllIeme Siecle, Paris, 1893.
40. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 24. According to a letter
that Louis-Fran90is wrote to Count Mole, Director-General of the Ponts et
Chaussees service, in 1811 (Arch. Nat. F14 2187 2 , Cauchy file) Cauchy won the
competition for wooden bridge engineering.
41. ENPC library, Ms. 1845. The autographic manuscript, dated February 15, 1808, in
Paris, and signed Aug. Louis Cauchy consists of 27 pages with 25lines per page.
42. See Arch. Nat. F14 2234 5 , Girard file and Archives Ac. Sci. Girard file. P. S. Girard
(1765-1836) entered the Ecole des Ponts et Chaussees in 1784. He was appointed
engineer-ordinary in 1789. In 1792, he won the Academie's double prize for his
study on canal locks, and in 1798, he presented his Traite Analytique de la
Resistance des Solides et des Solides d'Egale Resistance, auquel on a joint une Suite
de Nouvelles Experiences sur la Force et l'Elasticite Specifiques des Bois de Chene et
de Sap in. He was appointed member of the Institut d'Egypte in 1798 and was
elected to the First Class of the Institut in June 1815. Because of a disagreement
with Ponts et Chaussees inspector Cahouet, he was removed from his position
directing the Ourcq canal in 1817.
43. In 1806, P. T. M. Egault (1777-1829) invented a new water level that was adopted
by the Ponts et Chaussees service. Under his direction, Cauchy worked on the
surveying of the circle aqueduct, which ran along the inner walls of Paris from la
Villette to Mousseaux and from which large underground conducts or galleries ran
to the center of the city. Following this, he worked on the raising ofthe land along
which the Galerie de Saint-Laurent would pass. Located between the Faubourg
Saint-Laurent and the Faubourg Saint-Denis, this gallery's length measured some
600m.
 Notes 245

44. Report from Girard to Count Mole, December 20, 1808, Arch. Nat. F14 2187 2 ,
Cauchy file. This report was published in A. Brunot and R. Coquand, Le Corps des
Pants et Chaussees, Paris, 1982, p. 95, note 2.
45. See C. A. Val son, La Vie et les Travaux du Baron Cauchy, 1, 25. Cauchy's study of
navigation on the Marne gives support to the idea that he worked also on the
Ourcq Canal in 1809.
46. Arch. Nat. F14 2148. An anonymous two-page note entitled 'Sur la solution de M.
Cauchy relative au probleme propose pour Ie concours de 1809' is in the ENPC
library (Ms 1982).
47. From a letter written by Louis-Franr;ois Cauchy to Count Mole in May 1811
(Arch. Nat. F14 21872, Cauchy file). Louis-Franr;ois heard of the loss of these
studies from Monge several days before the awarding of the prize for the
competition of 1809. Concerning these studies, Louis-Franr;ois observed that his
son 'had spent considerable time on this work' and, according to his professors,
was successful. C. A. Valson (La Vie et les Travaux du Baron Cauchy, 1, p.43)
indicated that Cauchy's first work at Cherbourg was a study 'on the theory of stone
bridges and of arches in general'. The question obviously had to do with one ofthe
studies of 1809. Val son's errors came from a letter that Cauchy wrote his father. In
it, he asked for 'this study and that you make an effort to find it and send it to me, if
not completely, then at least the main formulas that I need in order to continue my
research'. From what was said, the date of this letter from Cauchy to his father can
reasonably be set as May 1811.
48. 'Memoire sur les moyens de perfectionner la navigation des rivieres en general et
celie de la Marne en particulier', handwritten manuscript of 51 pages, ENPC
library, Ms 1982.
49. 'Memoire sur les ponts en pierre, par A. L. Cauchy, eieve des Ponts et Chaussees',
handwritten manuscript of 32 pages, ENPC library, Ms 1982.
50. 'Second memoire sur les ponts en pierre, theorie des voutes en berceau, par A. L.
Cauchy, eleve ingenieur des Ponts et Chaussees', handwritten manuscript of 52
pages, ENPC library, Ms. 1982.
51. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p.43.
52. See C. A. Geoffroy de Grandmaison, La Congregation (1801-1830), Paris, 1889.

Notes to Chapter 2
1. Arch. Nat. F14 2187 2 , Cauchy file. Letter from L. F. Cauchy to Count Mole,
September 15, 1812.
2. On the history of the site of Cherbourg, see A. Demangeon and B. Fortier, Les
Vaisseaux et les Villes. L'Arsenal de Cherbourg, Bruxelles, 1978.
3. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 27.
4. Arch. Nat. DD. 2144, 'Rapport sur l'etablissement maritime projete aCherbourg'
by Cachin, 23 Germinal year XI (April 14, 1803), published in A. Demangeon and
B. Fortier, Les Vaisseaux et les Villes ... , pp. 106-117
5. Arrete of 25 Germinal Year XI (April 16, 1803), published by A. Demangeon and
B. Fortier, Les Vaisseaux et les Villes ... , pp. 44-45.
6. Ibid., p. 154.
7. It is estimated that there were at least 200 steam engines in France around 1810.
8. Cited by C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 30.
9. Arch. Nat. F14 2187 2 , Cauchy file. Letter from Louis-Franr;ois dated July 9, 1810.
On July 16, the director of personnel replied that he had just requested that the
246 Notes

Minister of the Marine agree that Augustin-Louis Cauchy retain his rank of
aspirant while assigned to work at Cherbourg.
10. 'The enormous basins that ceaselessly resound under the blows of steel' is part of a
verse that Augustin-Louis Cauchy included in a letter to his mother in August
1810, cited by C. A. Valson, La Vie et les Travaux du Baron Cauchy, t, p. 3.
11. Letter from A. L. Cauchy to his family, written on July 3,1811; ibid., 1, p. 28.
12. Letter from J. F. Cachin to A. L. Cauchy, undated; ibid., 1, p. 32.
13. Letter from A. L. Cauchy to his mother, written in 1810; ibid., 1, pp. 37-41. The
people mentioned by Cauchy have not all been identified. Hippolyte Franqueville
was the chief commissioner of the Port of Cherbourg. L. B. Fouques-Duparc
(1772-1848), engineer of the Ponts et Chaussees, entered the Marine Service in
1803 and succeeded Cachin as the director of the site in 1813. S. Vallot (1772-1847),
engineer-geographer and then engineer of the Ponts et Chaussees, taught
construction at the Ecole des Ponts et Chaussees. As to Monsieur L. .. , the
reference is probably to Count of Latour-Maubourg.
14. Ibid., pp. 38-39.
15. Letter from A. L. Cauchy to his father, dated June 8, 1810; ibid., 1, p. 31.
16. Letter from A. L. Cauchy to his family, undated; ibid., 1, p. 31.
17. 'Sur les limites des connaissances humaines', Bibliotheque de l'Institut, Ms. 2038;
F O83r-86r, (O.c., 2, 15, pp. 5-7).
18. Letter from Augustin-Louis Cauchy to his family, dated December 10, 1810, cited
by C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 29.
19. Ibid., 1, p. 45. No source is given as a basis for this statement. It seems quite likely,
although Valson apparently confused the 1811 study with the study of January 20,
1812, entitled 'Sur les polygones et les polyedres'.
20. Proces-verbaux, Ac. Sci., 4, p. 449. The manuscript, entitled 'Recherches sur les
polyedres', is in the meeting's packet of February 11, 1811. A resume was published
in the Correspondance sur l'Ecole Poly technique, 2, nO 3, January 1811, pp. 253-
256, (O.C., 2, 2, pp. 402-405). The complete study appeared in the J.E.P., 9, 16th
cahier, May 1813, pp. 68-86, O.c., 2, 1, pp. 7-25). On Cauchy's proof of the Euler
formula, see J. C. Pont, La Topologie Algebrique des Origines a Poincare, Paris,
1974, pp. 21-24, and N. Briggs, E. K. Lloyd, and R. J. Wilson, Graph Theory
(1736-1936), Oxford, 1976, pp. 74-83.
21. Proces-verbaux, Ac. Sci., 4, pp. 467-477, and Correspondance sur l'Ecole Poly tech-
nique, 2, nO 4, July 1812, p. 361 (O.c., 2, 2, pp. 406-408).
22. Proces-verbaux, Ac. Sci., 5, p. 6.
23. The manuscript is in the meeting's packet of January 20, 1812. A resume was
published in the Journal des Mines, 31, nO 184, April, 1812, pp. 314-318. (O.C., 2, 15,
pp. 8-10). The complete study appeared in the J.E.P., 9, 16th cahier, May 1812,
pp. 87-98. (O.c., 2, 1, pp. 26-38).
24. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 47-48.
25. Legendre report, Proces-verbaux, Ac. Sci., 5, pp. 17-18, and Correspondance sur
['Ecole Poly technique, 2, nO 4, July 1812 (O.C., 2, 2, pp. 408-413).
26. See. O.c., 2, 15, p. 6.
27. Relative to this point, see, for example, J. B. J. Delambre, Rapport Historique sur
les Progres des Sciences Mathematiques depuis 1789 et sur leur Etat Actuel, Paris,
1810. According to a letter mentioned by C. A. Valson, La Vie et les Travaux du
Baron Cauchy, 1, p. 43, it seems that in 1811 Cauchy contemplated pursuing the
investigations on the theory of vaulted arches that he had undertaken at the Ecole
 Notes 247

des Ponts et Chaussees in 1809. This would have been an abandonment of pure
mathematics.
28. From J. Mandelbaum, La Societe Philomatique de Paris de 1788 a 1835, 3rd cycle,
thesis, typewritten, Paris, EHESS, 1980, 1, pp. 199-200.
29. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 54.
30. Letter from Louis-Fran.;ois Cauchy to his son, undated; cited by C. A. Valson, La
Vie et les Travaux du Baron Cauchy, 4, p. 54.
31. Ibid., 1, pp. 48-49.
32. Ibid., 1, p. 29.
33. Ibid., 1, p. 244 and p. 245.
34. Arch. Nat. F14 2187 2 , Cauchy file.
35. See the letters from Augustin-Louis, Louis-Fran.;ois, and Marie-Madeleine
Cauchy, September and October 1812, in the private correspondence of the family
Cauchy, kept by M. de Leudeville, Paris.

Notes to Chapter 3
1. Proces-verbaux, Ac. Sci., 5, p. 121. The manuscript is lost. The memoir was
published in two articles, 'Sur Ie nombre des valeurs qu'une fonction peut acquerir
lorsqu'on y permute de toutes les manieres possibles les quantites qu'elle renferme'
and 'Sur les fonctions qui ne peuvent obtenir que deux valeurs egales et de signes
contraires par suite des transpositions operees entre les variables qu'elles
renferment', J.E.P., 10, 17th cahier, January 1815, pp. 1-28 and pp. 29-112, (D.C.,
2,1, pp. 64-90 and pp. 91-169). For a critical study ofthis paper, see A. Dahan, Les
Recherches Algebriques de Cauchy, 3rd cycle thesis, typewritten, Paris, EHESS,
1979, pp. 11-34, and 'Les travaux de Cauchy sur les substitutions. Etude de son
approche du concept de groupe', Archive for History of Exact Sciences, 23, 1980,
pp.279-319.
2. On March 30 and April 3, Cauchy read a paper entitled 'Sur Ie nombre des
polygones que i'on peut former en prenant pour sommets les points de division
d'une circonference divisee en plusieurs parties egales' at the Societe Philomatique.
The problem had been proposed in Gergonne's Amiales de Mathematiques, 3,
pp. 231-232. This unpublished work was clearly an application of this study in
which Cauchy represented a p-cycle as a polygon.
3. Cauchy obtained his theorem on the product of two determinants during the
summer of 1812, while in Cherbourg. He was attempting 'to generalize the
formulas of M. Gauss'. Jacques Binet, who was CauchY's friend, independently
obtained a similar set offormulas, but they were much more difficult to work with.
4. Proces-verbaux des Seances of the Committee ofthe Ecole des Ponts et Chaussees,
meeting of February 3, 1811, cited by Andre Lorion, 'L'Ecole des Ponts et
Chausees sous Ie Premier Empire (documents inedits)" Revue de l'Institut
Napoleon, nO 66,1958, pp. 81-86.
5. Arch. Nat. F14 2187 2 , Cauchy file, letterfrom the personnel director to the Minister
of the Marine and Colonies, dated February 16, 1813:
Monsieur Cauchy, an engineer-ordinary who is assigned to the port of Cherbourg, has
informed me that his health will not easily allow him to take his post and has asked to
be assigned to service in the interior. Accordingly, I now ask your excellency to inform
me by May 1 next whether or not, in the event that it is found that his presence is not
essential, I can replace him at Cherbourg.
248 Notes

On March 18, 1813, Cauchy sent a letter of thanks to Count Mole, expressing
gratitude for his (Cauchy's) reassignment to Paris. Five years later, under rather
similar circumstances, the physicist Augustin Fresnel was appointed to a post with
the Ourcq Canal Project.
6. Later, by a decree of December 24, 1816, Lehot became a repetiteur in physics at
the Ecole Polytechnique, where he and Cauchy once again crossed paths.
7. See Arch. Nat. F14 2263 2 , Lehot file and letter from Cauchy to Lehot, Paris, April
5, 1813, in Archives Ac. Sci., Bertrand autographs' collection, carton 1:

My dear friend:

I have just visited M. Picard, with whom I have an appointment for tomorrow at the
Saint-Martin Canal. I will visit you at 11: 30, and together we will be able to check the
measurements and rule definitively on M. Picard's report. I am going to write to M.
Prosper and to M. Delozanne, so that the former will assist us, while the latter will be
able to support his interests. I assure you of my continuing sincere devotion.
Picard, Prosper, and Delozanne were probably master builders. According to the
report made by a commission of the Ponts et Chaussees on January 1, 1816 (Arch.
Nat. F14 7012), concerning the Ourcq Canal Project and related works, the work
had not progressed very far. 'Meanwhile, earth had been removed from the slope at
the bend ofthe Villette' and in continuation of the foundations of the Fontaine de
l'Elephant, a stretch of canal in stone, 90 m in length, a part of which is already
vaulted, has been constructed. During his brief time with the Ourcq Canal and
Paris Waterworks Project, Cauchy probably worked at these construction sites.
Lehot did not leave until the autumn of 1813.
8. Arch. Nat. F14 2263 2 , Lehot file.
9. See G. Bigourdan, 'Le Bureau des Longitudes; son histoire et ses travaux de
l'origine (1795) a ce jour', Annuaire du Bureau des Longitudes, 1928; A30-31 and
A36.
10. Proces-verbaux, Ac. Sci., 5, p.216, (O.c., 2, 15, p.16). The commission was
composed of Legendre, Carnot, and Poisson.
11. Rapports sur divers memoires Ius ala premiere classe de l'Institut imperial par A. L.
Cauchy, ingenieur des Ponts et Chaussees; Paris, 1813, 15p. Copy in the
Bibliotheque de l'Institut; 8° HR 25, I (10), (O.c., 2, 15, pp. 11-16).
12. 'Recherche sur les nombres', J.E.P., 9, 16th cahier; pp. 99-123 (O.c., 2, 1, pp. 39-
63).
13. The meeting's packet of May 17, 1813, contains two manuscripts on this memoir:

1. A 21-page manuscript entitled 'Theorie des equations: methode pour determiner a

priori Ie nombre de racines reelles positives et Ie nombre des racines reelles negatives
d'une equation d'un degre quelconque par A. L. Cauchy ingenieur des Ponts et
Chaussees. Memoire'.
2. A 7-page manuscript entitled 'Theorie des equations: methode pour determiner it
priori Ie nombre des racines reelles positives, et Ie nombre des racines reelles
negatives d'une equation de degre quelconque par A. L. Cauchy, ingenieur des
Ponts et Chaussees employe it Paris aux travaux du Canal de I'Qurcq. Expose
sommaire de cette methode'.

A note signed by De1ambre on the first page of the latter memoir suggests that it
was presented on May 13, 1813, and that Laplace, Biot, and Poisson were
responsible for evaluating it. It is the manuscript that begins with 'Messieurs, ... , ,
 Notes 249

which was to be presented at the session. It was printed by Veuve Courcier in 1813
(see O. c., 2, 15, pp. 11-16).
14. 'Memoire sur la determination du nombre des racines reelles dans les equations
algebriques', J.E.P., 10, 17th cahier, pp. 457-558, (O.c., 2, 1, pp. 170-257). This
memoir consists of extracts of the papers Cauchy had delivered at the Institut
during 1813.
15. Proces-verbaux, Ac. Sci., 5, p. 217 and p. 218.
16. Arch. Nat. F14 2187 2 , Cauchy file, letter of thanks from A. L. Cauchy to Count
Mole, dated June 8, 1813. See also Arch. Nat. F14 22101, DenoeI file.
17. Arch. Nat. F14 2187 2 , Cauchy file.
18. The meeting's packet of October 18, 1813, includes the manuscript of a study
entitled 'Methode pour determiner a priori Ie nombre de solutions positives et Ie
nombre de solutions negatives d'une equation de degre queiconque'. The file of the
meeting of November 22, 1813, contains several other manuscripts:
1. A small study, probably written shortly after October 18, entitled 'Theoreme sur la
difference entre Ie nombre de racines positives et Ie nombre de racines negatives
d'une equation de degre n';
2. A short note in which he explains how 'the general method, which would seem very
complicated at first, actually has an unexpected degree of simplicity'.
3. A long paper that was presented on November 22 and is entitled 'Sur un moyen
d'eviter I'emploi des indeterminees dans la formation des equations auxiliaires qui
servent afixer Ie nombre des solutions positives et Ie nombre des solutions negatives
d'une equation quelconque'.
Each of these papers has remained unpublished, but the basic ideas contained
in them are found in the long article in J.E.P., cited in note 14.
19. Arch. Nat. F14 2187 2 , Cauchy file, letter from Louis-Franf;ois Cauchy to Baron
Costaz, dated January 12, 1814.
20. A note written in pencil in the margin of the letter of January 12, 1814, which is cited
in the preceding note, points out [that] 'it is necessary to keep M. Cauchy on the
Ourcq Canal Project'. Moreover, in the Livre du Mouvement du Personnel des
Ponts et Chaussees for the year 1816, we find, that among the engineers assigned to
the OUTCq Canal Project: 'Augustin-Louis Cauchy, engineer 2nd class, attached to
the Institut, received no salary from Ponts et Chaussees'. No document, however,
reveals when Cauchy effectively ceased working with the Ponts et Chaussees.
21. 'Sur Ie systeme des valeurs qu'il faut attribuer adivers elements determines par un
grand nombre d'observations pour que la plus grande des erreurs, abstraction faite
du signe, devienne un minimum', Bulletin Phil., June 1814, pp. 92-99, (O.c., 2, 2,
pp. 312-322).
22. The Great Referendary, under whose order Louis-Franf;ois worked, was the peer
of France who was responsible for putting the seal of the assembly on enacted
measures and bills and taking care of the archives.
23. Letter cited by C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 56-57.
24. Indication sommaire des memoires presentes a la premiere classe de l'Institut par
A. L. Cauchy, ingenieur des Ponts et Chaussees. Conclusions des rapports faits a la
classe sur ces memoires, Paris, 1814. Copy in the Bibliotheque de l'Institut, Rec.
H. R. 26 (t. I, nO 9). The titles of the papers and studies presented to the Institut in
1811, 1812, 1813, and 1814 are formed here (Ml to M8 from the nomenclature of
the O. C.'s bibliography, O. c., 2, 15, pp. 589-595).
250 Notes

25. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 55-56.

26. Letter from A. M. Ampere to Bredin, dated October 13, 1814, Correspondance du
Grand Ampere, 2, Paris, 1936, p. 486.
27. Proces-verbaux, Ac. Sci., 5, p. 431 and p. 434.
28. J. Mandelbaum, La Societe Philomatique de Paris de 1788 a1835, 3rd cycle thesis,
typewritten, Paris, EHESS, 1980, 1, pp. 199-200.
29. Proces-verbaux, Ac. Sci., 5, pp. 495-496 and p. 502.

Notes to Chapter 4
1. Louis-Franyois got through this troubled period without undue difficulties, being
Keeper of the Archives in the Chamber of Peers during the First Restoration;
Secretary of the Archives of the Senate during the Hundred Days; and, again,
Keeper of the Archives of the Chamber of Peers during the Second Restoration.
Fontanes explained Louis-Franyois' staying power in the following terms: 'He was
appreciated at the Palais du Luxembourg for his Latin verses as well as for his
perfect integrity and his capacity to work in a very useful and unobtrusive,
inoffensive way'. (Cited by A. F. Villemain in his Souvenirs Contemporains
d'Histoire et de Litterature, Paris, 1855, p. 42).
2. Letter from A. L. Cauchy to the Abbe Jean-Baptiste-Armand Auger, dated
September 3, 1815, published by J. Pelseneer in the Archives Internationales
d'Histoire des Sciences, 4, 1951, pp. 631-633. Born in 1784, Auger was a professor
of mathematics, a member of the Congregation, and from 1814, the Vicar of the
Saint-Franyois Parish in Le Havre. (See Th. Lebreton, Biographie Normande,
Rouen, 1857-1861, J. B. A. Auger, article.)
3. See P. Lafitte, 'Relations d'Auguste Comte avec Poinsot', Revue occidentale,
March 1, 1886, p. 147.
4. Archives E. P., Poinsot file, Letter of November 15, 1815, from Durivau to the
governor, the Count Dejean:
My General,
Since it is urgent to temporarily replace M. Poinsot, in order to provide an unbroken
progress of teaching, I beg to subject to your examination the names of two persons
who can be equally chosen. These are MM. Cauchy and Lefebvre. Both of them are
known by the favorable reviews the Conseil de Perfectionnement, and especially
several first-class scientists, has given. Thus, in respect to ability, it is unnecessary to
collect more information than you already have. But, another kind of consideration can
guide your choice: M. Lefebvre is already busy at the Ecole as repetiteur. We cannot
take him from his position without replacing him temporarily. Such a temporary
appointment is not attractive, and we cannot propose it, I believe, to M. Cauchy. Thus,
we could not seize the opportunity to make this eminent person attached to the Ecole.
However, it is beyond doubt that he would gladly accept to replace a professor, even
temporarily. Moreover, M. Lefebvre has just been promoted as repetiteur. This first
advantage would be followed very closely by another, ifhe was chosen this time. As for
M. Cauchy, he should have obtained nothing but promises. I think, therefore, General,
we must appoint him to the temporary post. If you will engage him, I shall beg him to
come to the Ecole in order to be questioned for his new job and to open the analysis
course tomorrow, November 16, if possible. We will see which decision M. Poinsot
takes. By that time, the teaching will not be broken.

Durivau
 Notes 251

5. The Governor Dejean answered the treasurer ofthe Ecole, Marielle, who proposed
increasing Cauchy's pay (Archives E. P., Poinsot file, letter from Dejean to
Marielle):
M. Cauchy is a beginner. He has, therefore, to reduce his claims for his pay. Obtaining
the post will compensate him later. I persist. December 8.

Count Dejean
6. Joseph Bertrand wrongly stated in his Efoges Academiques, Paris, 1902, 2, p. 109,
that 'a ~ew months after his return [from Cherbourg], Cauchy became a repetiteur
at the Ecole Polytechnique'.
7. In the draft copy of a letter to te minister of the Interior, dated December 22 1815
the Governor of the Ecole himself avowed that: ' ,

In political terms, the committee's choices are not less satisfactory. Among the persons
who have been nominated there are only true and faithful Frenchmen who are quite
likely to inspire young people to be serious about their duties and obligations and to
seek to perfect their education. I add that this situation excludes from our institution
certain persons whose names became, unfortunately, too notorious during the
Revolution and admits no one who, at this time, is not acceptable to the Court.
This last phrase is crossed off (in the draft). Dejean added in the margin:
A person might well state this by word of mouth-but it ought not be put in writing.
8. P.v. Ac. Sci., 5, December 26,1815, p. 738.
9. Proces-verbaux, Ac. Sci., 5, November 13, 1815, p. 576. An abstract appeared in the
Bull. Phil., Dec. 1815, pp. 196-197,(O.C.,2,2,pp. 204-206). Legendre, who was the
reporter ofthe study, gives Cauchy's proof in a supplement to the second edition of
his Essai sur fa Theorie des Nombres in February 1816. Cauchy's paper was
published in the Mem. Institut, 1,14 (1813-1815),1818, pp. 172-220, and in the Ex.
Math., 2, Nov. 1826, pp. 265-296, (O.C., 2,6, pp. 320-353).
10. Cauchy presented to the Societe Philomatique on January 28, 1816, a solution to a
similar but more general problem than the one he had dealt with in his paper on
Fermat's theorem: to decompose a given integer into several squares whose roots
shall form a given sum. His solution remains unpublished (see J. Mandelbaum, La
Socite Philomatique de Paris ... 1, pp. 199-200).
11. Proces-verbaux, Ac. Sci., 5, December 26, 1815, p. 596.
12. J. Bertrand, Eloges Academiques, Paris, 1902,2, p. 112.
13. On Cauchy's teaching course of 1815-1816, we have the Registre d'instruction (See
Appendix I) and the notes of the student Auguste Comte (See T. Guitard, 'La
querelle des infiniment petits a l'Ecole Poly technique au XIXe siecIe' Historia
scientarium, nO 30, pp. 1-61, especially pp. 28-31).
14. For a discussion of these events, see H. Gouhier, La Jeunesse d'Auguste Comte et la
Formation du Positivisme, Paris, 1933-1941, 1, pp. 116-122.
15. See C. Marechal La Dispute de l'Essai sur l'IndifJerence, Paris, 1924, pp. 87-107.
16. Archives ofthe College de France, Registre des deliberations prises aux Assemblees
des lecteurs et professeurs du Roi au College Royal de France, Vol. III; Meeting of
November 10, 1816.
17. See Chapter 7, p. 113.
18. Archives of the College de France, vol. IV; meeting of November 21, 1824:
252 Notes

A letter from M. Biot to the administrator was read. In this letter, M. Biot, who was on a
trip that was to last for approximately one year, asked the professors to decide in favor
of M. Cauchy teaching the course in mathematical physics during M. Biot's absence.
M. Cauchy is a member of the Academie Royale des Sciences and a professor at the
Ecole Poly technique. This proposition was supported by the assembly, and M.
Cauchy's classes became part of the curriculum.
Cauchy's courses effectively began at the end of November.
19. Archives of the College de France, Biot file:

Finding it impossible to attend the next meeting, I ask the honorable administrator to
please request in my name from the professors that they agree that M. Cauchy (who had
already substituted several times for me) should assume my duties and receive, in
payment, 2000 francs of my annual pay, effective January 1, 1826 (in agreement with the
guidelines submitted by our school to the Minister of the interior). I nevertheless desire
to preserve the possibility and the right to reclaim my whole salary, if new
circumstances in my present situation should come about so as to oblige me to
reconsider and resume the duties of my chair.
Paris, November 21, 1825
J. B. Biot.

20. For a discussion on the method of solving linear differential equations, see C.R. Ac.
Sci., 8, p. 829, (O.c., 1, 7, p. 369). In 1827-1828, according to a program of the
courses, which is preserved in the Archives of the College de France, Cauchy
expounded on The general methods by which it is possible to solve the main
equations of mathematical physics'. Finally, in 1829-1830, Cauchy presented his
theory of light (see Chapter 6, pp. 104-105).
21. Arch. Nat. AJ 1625 (file 1822). The salary amounted to 2000 francs, to be deducted
from the titular's salary.
22. Arch. Nat. AJ 165126.
23. According to a letter from Ampere to Bredin, dated March 16, 1821 (La
Correspondance du Grand Ampere, 3, Paris, 1943, p. 908):
I suffer a good deal from chest and lung ailments, and it has been necessary for me to
suspend my course at the Faculte. M. Cauchy will replace me for a month or six weeks.
24. There is little material bearing on these courses. From Cauchy, there is the note
given in the appendix to his September 16, 1822, paper and the note given in C.R.
Ac. Sci., 16, February 20, 1843, p. 413, (O.c., 1, 7, p. 261):
In December 1821, in my course on mechanics at the Faculte de Sciences, I presented a
general method by means of which I have been able to obtain certain formulas
pertaining to surfaces, volumes, masses, etc., relative to an arbitrary system of
curvilinear coordinates...
In the preface of his Le(:ons de Mecanique Analytique, Paris, 1867, Moigno wrote:
This material was first given quite some time ago, because it is essentially the course
that my illustrious and venerated teacher. A. L. Cauchy, taught at the Ecole
Poly technique during the years from 1820 to 1830. I faithfully recorded and edited this
material; and, for his own part, Cauchy, in Exercices de M athematiques and in
Nouveaux Exercices de Geometrie et de Physique Analytique (sic), published the most
complete theorems, such as those on linear moments and on the investigation of the
general equations of equilibrium, which constituted the basic material for his courses.
 Notes 253

From 1838 to 1843, I used this handwritten material as the text for courses I taught at
the Ecole Normale Ecclesiastique de la rue des Postes. Moreover, if the unfortunate
incident I discussed in the preface to my lessons on integral calculus had not taken
place, they would have appeared a long time ago.
As Moigno stated later in his preface, Cauchy was the real author of Le(:ons,
especially of the first 9 and the 12th. They provide an idea of the material (at least in
part) that Cauchy presented in his courses at the Faculte des Sciences. Also, see the
10 articles contained in Exercices de Mathematiques (issues for the 1st and 2nd
years) in which the principles of mechanics are discussed. In particular. there is a
discussion of the theory of linear moments (see p. 259, note 56).
25. Arch. Nat. AJ 16207. In 1824, Cauchy seems to have been somewhat hesitant to
assume the vacant chair in astronomy. This is according to a letter Ampere wrote
to Monseigneur Frayssinous (Correspondance du Grand Ampere, 2, p. 667):
I have learned that M. Cauchy, who has so brilliantly filled the chain in mechanics at
the Faculte des Sciences, in fact, prefers the now vacant chair in astronomy. If your
excellency has not already chosen another professor of mechanics, I should like very
much to be considered for the vacancy in mechanics. I have taught mechanics for seven
years at the Ecole Royale Poly technique. At the Faculte, M. Cauchy was teaching
precisely the same material as that which constitutes the course at this school. Since I
have been continuously engaged in teaching this material, perhaps it is I who might do
the best of any of those who might be selected, in terms of presenting a course that
assumes special study and in-depth investigation of this branch of the mathematical
sciences.
26. Cauchy specified his criticisms during the following year. As to generating
functions, see the paper that was presented to the Academie on December 27,1824,
and was published under the title 'Memoire sur Ie calcul integral' in M emoires Ac.
Sci., 22, 1850, pp. 39-130; (O.c., 1, 2, pp. 195-281). For Poisson's study. see
Chapter 7, p. 122.
27. In a letter to Paolo Ruffini, dated September 20, 1821, which was published in P.
Ruffini, Opere Mathematiche, 3, Rome, 1954, pp. 88-89, Cauchy indicated the
following on this point:
I have long since been bound to the author, whom you have refuted and lowe him
much, [still] I have never hidden my feelings nor my principles from him. In the
introduction to my course on analysis, in which I otherwise gave him all possible credit,
I formally stated-as you can read for yourself-that history ought not be investigated
from a standpoint offormulas; nor should sanction be sought for ethics and morals in
the theorems of algebra or integral calculus.
As to Cauchy's views on the calculus of probabilities, see Chapter 13, p. 219.
28. See the list of studies and papers for which Cauchy acted as reporter in O.c., 2,15,
pp. 518-526.
29. The study was not published and the manuscript disappeared. The title of this
work is uncertain: 'Integration des equations lineaires aux differences finies ou
infiniment petit' or 'Recherche sur Ie calcul integral aux differences partielles.' See
Chapter 7, p. 122.
30. On the dispute of April 1829, see the minutes of the meetings of the Academie of
April 6, 1829, and April 27, 1829. Cauchy's arguments were published in Ex. Math.,
4, November 29, pp. 214-216. (O.c., 2, 9, pp. 254-258). Cauchy's observations on a
question of priority were kept in the meetings's packet for June 14, 1830.
254 Notes

31. For example, in the introduction to Analyse Algebrique of 1821 and in the foreword
of Lefons sur Ie Calcul DifJerentiel.
32. Poinsot said of his successor at the Ecole Polytechnique: 'Cauchy is affected by a
diarrhea of x'. He told Auguste Comte of his dislike for Cauchy. See P. Lafitte,
'Relations d'Auguste Comte avec Poinsot', and H. Gouhier, La Formation du
Positivisme: La Jeunesse d'Auguste Comte, 1, Paris, 1933, p. 129.
33. See Poinsot, note in Bull. Fer., 7, April 1827, pp. 224-226, and Cauchy's reply in the
same journal, Bull. Fer., 7, May 1827, pp. 333-337 (D.C., 2, 15, pp. 138-140). The
following remark by Cauchy should be noted:
The time that scholars and scientists spend in 'making war' on each other I regard as
nothing short of a loss for science. I believe that it is far better to solve problems and
investigate questions than to get involved in disputes.
It is unfortunate that Cauchy himself did not always adhere to such a wise attitude!
34. The bonds of friendship that existed between Cauchy and Binet were strong,
durable, and old. Cauchy mentioned Binet, his close friend since 1812, in his paper
on determinants. On May 1, 1821, he officiated at the reception for Binet when he
(Binet) became a knight of the Legion d'Honneur, and he spoke for the Academie at
Binet's funeral in 1856.
35. See, for example, the article on Cauchy in Galerie Historique des Contemporains,
2nd supplement, Mons, undated, signed D.M. (1828-1829?):
He has a dry, rigid personality and his lack of tolerance for or indulgence with youag
people who would carve out careers for themselves in science has made him one of the
least likeable-and certainly one of the least liked-scholar-scientists.
36. For these 32 papers that were examined outside the commission, Cauchy prepared
9 written evaluative reports and 17 verbal reports.
37. On this academic work, see the documentary appendix established by R. Taton in
D.C., 2, 15, pp. 518-526 and pp. 579-580.
38. J. V. Poncelet, Applications d'Analyse et de Geometrie, 2, Paris, 1864, p. 564.
39. Translated into French in Niels-Henrik Abel, Memorial Publie a ['Occasion du
Centenaire de sa N aissance, Christiania, 1902, 3rd pagination, p. 45.
40. Letter to Holmboe in Berlin, dated January 20, 1827; ibid., 3rd pagination, p. 57.
41. See D.C., 2, 15, pp. 572-573.
42. See R. Taton, 'Sur les relations scientifiques d'Augustin Cauchy et d'Evariste
Galois', Rev. Rist. Sci., 24,1971, pp. 123-148, esp. p. 138.
43. See A. Dahan, us Recherches algebriques de Cauchy, 3rd cycle, Thesis, typewritten,
1979, pp. 80-83.
44. A. Iushkevich, Michel Dstrogradski et Ie Progres de la Science au XIXeme Siecle,
Coriference donnee au Palais de la Decouverte, Paris, 1967, p. 13.
45. See the chronological table of the studies presented to the Academie between 1816
and 1830 and the chronological table of publications for the same period. Arranged
by R. Taton in the documentary appendix of Oeuvres Completes, D.C., 2, 15,
pp. 590-594 and pp. 598-601.

Notes to Chapter 5
1. See the introduction to Analyse Algebrique and the foreward to Lefons sur les
Applications du Calcul 11!finitesimal ala Geometrie. Cauchy also consulted Coriolis
on his first investigations on the theory of light (see Chapter 6, p. 105).
 Notes 255

2. Archives E. P., Registre des proces-verbaux des seances du Conseil d'[nstruction,

November 9, 1816.
3. Archives Ac. Sci., Ampere's papers, box 4, chapter 4, file 75, 'Programme du cours
d'analyse par Augustin Cauchy'. See Appendix II pp. 303-305.
4. Falling under the traditional heading of algebraic analysis were basic notions of
analysis, real and complex numbers, functions, infinitesimals, series, etc.
5. See especially Memoires Sav. Etr., I, 1827, p. 687, (O.c., 1,1, pp. 402-403). In this
passage, Cauchy asked how can the value of the definite integral

J:' 4>'(z)dz

be derived from the value of the primitive function 4>(z) (see Chapter 7, p. 110):

If the function <jJ(z) increases or decreases in a continuous manner between the limits
z = b and z = b', the value of the integral will be represented, ordinarily, by <jJ(b")-
<jJ(b'). However, if for one value of z represented by Z and lying between the limits of
integration, the function <jJ(z) passes suddenly from one fixed value to a value sen-
sibly different from the first, in such a way that for a very small quantitiy " we have
<jJ(Z + 0 - <jJ(Z - 0 = d, then the ordinary value of the definite integral given by
<jJ(b") - <jJ(b') should be diminished by the quantity d, as can be easily shown.

The implicit notion of continuity (.1 = 0) and the explicit definition of a jump
discontinuity (.1,= 0) in this passage are consistent with the definition in Analyse
Algebrique.
I. Grattan-Guinness assumed that Cauchy had plagiarized Bolzano's definition
of continuity (see I. Grattan-Guinness, 'Bolzano, Cauchy, and the "New analysis"
of the early nineteenth century', Arch. Rist. Ex. Sci., 6, pp. 372-400). In fact, this
thesis is not convincing. By an internal study, H. Freudenthal has clarified the
differences of the approach between the two mathematicians (see H. Freudenthal,
'Did Cauchy plagiarize Bolzano?', Arch. Rist. Ex. Sci., 7,1971, pp. 375-392, and
also H. Sinaceur, 'Cauchy et Bolzano', Rev. Rist. Sci., 26,1973, pp. 97-112). The
1816 instructional plan and the Registre d'instruction of 1816-1817 (see note 6)
corroborate the anteriority of Cauchy's definition of continuity.
6. Archives E. P., XII C7, Registres d'instruction. These registers, kept daily, give the
title and occasionally a brief summary of these lessons, according to the indications
of the teachers. See Appendix II, pp. 305-307, and C. Gilain 'Cauchy et Ie cours
d'analyse de I'Ecole Polytechnique', Bulletin de la Societe des Amis de la
Bibliotheque de [,Ecole Poly technique, 5, pp. 2-145.
In 'La querelle des infiniment petits it I'Ecole Polytechnique au XIXe siecle',
Ristoria Scientiarum, 30, 1986, pp. 1-61, Thierry Guitard inferred a chronology of
the genesis of Analyse Algebrique from the Registres d'instruction. He defined the
year 1817, the 'annus mirabilis', as the turning point in the elaboration of Cauchy's
course, especially with the 'invention of the continuity' on March 1, 1817. I do not
believe that this inference is correct, since the Registres give only a terminus ad
quem: for example, according to the Registres, Cauchy knew the modern concept of
continuity as far back as March 1817, but the 'invention' was anterior, as shown by
the instructional program of December 1816.
7. 'Sur les raeines imaginaires des equations', Bull. Phil., January 1817, pp. 5-9 (O.c.,
2,2, pp. 210-216); this article resumes the memoir that Cauchy presented to the
256 Notes

Academie on December 23, 1816 (Proces-verbaux, Ac. Sci., 6, p. 132). Cauchy

presented a new proof of the fundamental theorem of algebra to the Academie on
October 13, 1817 ('Sur la decomposition des polynomes en facteurs reels du
deuxieme degr6'; see Proces-verbaux, Ac. Sci., 6, p. 228) and published it
immediately: 'Seconde note sur les racines imaginaires des equations', Bull. Phil.,
October 1817, pp. 161-164 [O.C., 2, 2, pp. 217-222; see also J.E.P., 11, 18th cahier,
January 1820, pp. 411-416 (O.c., 2, 1, pp.258-263) and Chapter X, §1, of the
Analyse Algebrique, pp. 331-339 (O.c., 9,3, pp. 274-288)]. In the second proof of
October 1817, Cauchy explicitly used the property of real continuous functions to
be bounded and to reach their boundaries on any closed set; he gaves no proof of
this statement, however.
8. Proces-verbaux, Ac. Sci., 6, p. 395. Cauchy published a resume in the Bull. Phi/,
January and February 1819, pp. 10-21 (O.c., 2, 2, pp. 238-252) (the manuscript is
kept in the meeting's packet), and the whole paper only in 1842 [see the 'Memoire
sur l'integration des equations aux derivees partielles du premier ordre', § I, Ex. An.
Phys. Math., 2, November 1842, pp. 241-260 (O.c., 2, 12, pp. 275-295)].
9. Quoted by C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 62-63.
10. Archives E. P., Registre des proces-verbaux des seances du Conseil d'Instruction,
March 4, 1819.
11. Ibid.
12. Ibid.
13. Archives E. P., Registre des proces-verbaux des seances du Conseil d'Instruction,
June 15, 1820.
14. See Chapter 4, p. 55.
15. In the introduction to Analyse algebrique, Cauchy recalled, in a few words, how he
was led to restructure this course:

Since several persons who were kind enough to guide me in the early stages of my
scientific career-and in this connection, I am particularly pleased to mention MM.
Laplace and Poisson-have expressed a desire to see me publish the Cours d'Analyse
de rEcole Polytechnique, I have decided to develop this course in such a way that would
be useful to my students. Accordingly, I am now submitting the first part [of the
course], which is called algebraic analysis. In this development, I have successively
treated the various kinds of functions, both real and imaginary; divergent and
convergent series; the solutions of equations, and the decomposition of rational
fractions.

16. Analyse algebrique, Introduction, pp. ii-iii (O.C., 2, 3, pp. ii-iii).

17. Ibid., Preliminaires, p. 4 (O.c., 2, 3, p. 19).
18. Ibid., Preliminaires, p. 4 (O.c., 2, 3, p. 19).
19. Ibid., Chapter II, § 2, pp. 34-35 (O.c., 2, 3, p. 43).
20. Ibid., Chapter II, § 2, pp. 37-39 (O.c., 2, 3, pp. 45-47).
21. Ibid., Chapter II, § 2, pp. 43-44) (O.c., 2, 3, pp. 50-51). In note III, pp. 460-462
(O.c., 2, 3, pp.378-380), Cauchy presented an analytic proof; in this proof,
however, he missed the difficulty by implicitly assuming the existence of a limit for
an upper-bounded (resp.lower-bounded) increasing (resp. decreasing) sequence of
real numbers.
22. See P. E. B. Jourdain, 'The origins of Cauchy's conception of the definite integral
and ofthe continuity of a function', Isis, 1, 1913, pp. 661-703; J. V. Grabiner, The
Origins of Cauchy's Rigorous Calculus, Cambridge: The MIT Press, 1981; and F.
 Notes 257

Smithies, 'Cauchy's conception of rigor in analysis', Arch. Hist. Ex. Sci., 36, 1986,
pp.41-61.
23. By 'analytical expression', Euler and his successors extended algebraic and
transcendental operations as well as infinite processes; that is, sums of series,
infinite products, and infinite continued fractions (see C. Houzel, 'Euler et
l'apparition due formalisme', in C. Houzel, J. L. Ovaert, P. Raymond, J. J. Sansuc,
Philosophie et Calcul de l'Infini, Paris, Maspero, 1976, pp. 130-135, and A. P.
Iushkevich, 'The concept of function', Arch. Hist. Ex. Sci., 16, 1976, 37-85).
24. Calcul Infinitesimal, Avertissement, p. 1. (O.C., 2, 4, p. 1).
25. Analyse Algebrique, Chapter VI, §2. pp. 125-126 (O.c., 2, 3, pp. 115-116).
26. Cauchy's statement is true only if the series of continuous functions is uniformly
convergent. Cauchy corrected his theorem and introduced the concept of uniform
convergence in 1853 [see 'Note sur les series convergentes dont les divers termes
sout des fonctions continues d'une variable reelle ou imaginaire entre des limites
donnees', C. R. Ac. Sci., 26, March 14, 1853, pp. 454-459 (O.c., 1, 12, pp. 30-36)].
27. Cauchy simply wrote log(z) to denote logarithm, defined in the half-plane R(z) > O.
mx m(m -1)x 2 m(m -1)(m - 2)x 3
28. N. H. Abel, 'Recherche sur la serie 1 + - + +- - - - - - -
1 1·2 1·2·3
+ .. ", Journal fur die reine und angew. Math., 1, pp. 311-339, especially, p. 313
(N. H. Abel, Oeuvres Completes, 1, 2nd ed., Christiania, 1881, pp.221-250,
especially, p. 222).
29. Archives E. P., Reports of April 13 and 14, 1821. A significant extract was
published by I. Grattan-Guinness, Annals of Science, 38, 1981, p. 680.
30. Archives E. P. Registre des proces-verbaux des seances du Conseil d'instruction,
April 17, 1821.
31. Archives E. P., VI 2a2-1805, report of April 21, 1821. An important extract was
published in the Annals of Science, 38, 1981, p. 681.
32. Ch. Dupin, Discours auxfunerailles de M. Augustin Cauchy, May 25,1857, Paris,
1857.
33. J. Bertrand, 'Notice sur Louis Poinsot', in L. Poinsot, E:lements de Statique, 11th
edition, Paris, 1873, pp. XXII-XXIV.
34. Archives E. P., Registre d'instruction, 1821-1822.
35. Archives E. P., Registre des proces-verbaux des Seances du Canseil d' instruction,
January 30, 1823.
36. In the foreword of Calcul Infinitesimal, Cauchy wrote:
This work, which was undertaken on the request of the Conseil de Perfectionnement de
I'Ecole Poly technique, gives the basics of the lectures that were presented at that
institution on the infinitesimal calculus. It will be composed of two volumes
corresponding to the two-year course. Today, I am publishing the first volume, which
consists of forty lessons, the first twenty of which have to do with the differential
calculus, while the last twenty deal with a portion of the integral calculus.

37. See Lagrange, Theorie des Fonctions Analytiques, Paris, 1797, in which Lagrange
uses the term derivative functions and employs the notation 1', j", f m,... , pn).
38. Nevertheless, Cauchy's treatment was confusing by the use of a notation borrowed
from Lacroix:
. ~u [u(x + IXh) - u(x)]
du= hm-= ,
a-+O Ll ct
 258 Notes

where h is an arbitrary finite constant and IX. an infinitesimal (see Lacroix: Traite
Elementaire de Calcul DijJerentiel et Integral, appendixe no 345).
39. On the theory of singular integrals and its use in the Calcul Infinitesimal, see
Chapter 7, pp. 115-118.
40. Cauchy explicitly criticized Lagrange's point of view in the foreword to Le(:ons sur
Ie Calcul DijJerentiel of 1829 and in the first of the articles 'Sui metodi analitici',
which was published in December 1830 (see p. 274, note 19).
41. See also Ex. Math., 1, May 1826, pp. 25-28 (O.c., 2, 6, pp. 38-43).
42. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, November 29, 1823,
One member [Laplace?] stated these pamphlets are too complicated and are beyond
the comprehension of the students. The lectures, which these pamphlets should clarify,
are helped but little; only a very few students can benefit by a study ofthis material. If
this serious problem is not quickly remedied, it is to be feared that in the following
classes the training in mathematics, which has contributed so much in elevating the
Ecole to the eminent rank it now occupies, will be weakened.
43. A. M. Ampere, Precis du Calcul DijJerentiel et Integral, an unpublished work that
was never completed.
44. The study is the memoir 'Sur la Theorie des Ondes' of 1815. Cauchy deposited it,
along with some supplementary notes, with the secretariat of the Academie on
May 17, 1824, for final publication.
45. These lectures were published in 1981 by C. Gilain under the title Equations
DifJerentielles Ordinaires. They were taken from copies that Cauchy himself had
deposited at the Bibliotheque de l'Institut.
46. For more on the lectures, see the introduction by C. Gilain, pp. XIII-XX.
47. Archives E. P., Registre des proces-verbaux des seances du Conseil d'Instruction,
November 24,1825. Remarks reported by the governor ofthe Ecole and confirmed
by Baron d'Hautpoul, member of the Conseil de Perfectionnement. See
A. d'Hautpoul, Quatre Mois a la Cour de Prague, Paris, 1912, p. 242.
48. Archives E. P., Registre des proces-verbaux des seances du Conseil d'Instruction,
March 10, 1825, and November 24, 1825. He stated, for example, that:

It is quite necessary to say it: the students come into the various schools of applications
with no knowledge of the integral calculus-or they very soon forget the little that they
may have learned. At Metz, they say only one student could do integrations. The Ecole
Polytechnique was not founded for the training of mathematicians, but rather it was
founded for the purpose of training students to enter certain public services. It is
therefore necessary that their education should be in those fields that are applicable to
the areas that fall within the scope of the [public] services concerned.
49. Archives E. P., Registre des proces-verbaux des seances du Conseil d'Instruction,
March 10, 1825.
50. Archives E. P., Registre des proces-verbaux des seances du Conseil de
Perfectionnement, November 21, 1825.
51. Here, Cauchy means the professor of applied analysis, Arago.
52. Archives E. P., Registre des Proces-verbaux des seances du Conseil d'Instruction,
January 12, 1826.
53. A supplementary note, 'Sur la decomposition des fractions', was printed as a
supplement to Calcul Infinitesimal of 1823 and distributed to the students during
the academic year 1824-1825. It is paged from page 177 to page 182, following the
 Notes 259

note 'Sur les formules de Taylor et de MacLaurin'. One copy of this unpublished
note is kept in the library of the Ecole Poly technique (Cote A3 a 57).
54. Cauchy stated in the foreword to the first volume that:
This work, which is destined to follow the Resume des Le,ons sur Ie Calcul I rifinitesimal,
will present the applications of calculus to geometry. It will be divided into three
volumes. The first two will examine the geometric applications of the differential and
the integral calculus that are related to the first-year analysis course at the Ecole Royale
Polytechnique. The present volume covers the main applications of differential
calculus.
Three articles of Exercices de Mathematiques completed the first volume of
Applications du Calcul biflnitesimal a la Geometrie:
1. 'Sur les centres, les plans principaux et les axes principaux des surfaces du second
degre', Ex. Math., 3, February 1828, pp. 1-22 (O.c., 2, 8, pp. 9-35).
2. 'Des surfaces que peuvent engendrer en se mouvant dans I'espace des lignes droites
ou courbes de forme constante ou variable', Ex. Math., 3, April 1828, pp. 23-64,
(O.c., 2, 8, pp. 36-82).
3. 'Discussion des lignes et des surfaces du second degre', Ex. Math., 3, June 1828,
pp. 65-120 (O.c. 2, 8, pp. 83-149).
55. These investigations connected some of the most general and abstract concerns
that had commanded Cauchy's attention since his 1813 work on determinants: the
transformation of homogeneous quadratic forms with arbitrarily many variables.
Stimulated by Sturm's works, he presented to the Academie the study 'Sur
l'equation al'aide de laquelle on determine les inegalites seculaires des mouvements
celestes' on July 27, 1829 [Ex. Math., 4, August 1829, pp. 140-160, (O.c., 2, 9,
pp. 174-195)]. In this study, he proved that for any symmetrical linear mapping
there exists a basis consisting of proper vectors, a theorem that he had already
stated without proof in a note presented to the Academie on November 20, 1826
['Sur l'equation qui a pour racines les moments d'inertie principaux d'un corps
solide et sur diverses equations du meme genre', Memoires Ac. Sci., 9, (1826),1830,
pp. 111-113, (O.C., 1,2, pp. 79-81)].
For more on this question, see Th, Hawkins, 'Cauchy, and the Spectral Theory of
Matrices', Historia Mathematica, 2, 1975, pp. 1-29 and A. Dahan, Les Recherches
Algebriques de Cauchy, 3rd cycle thesis, typewritten, Paris, 1974, pp. 44-50.
56. During this period, 10 articles treating classical mechanics appeared in Exercices
de Mathematiques, 1st and 2nd year; they were:

1. 'Sur la resultante et les projections de plusieurs forces appliquees en un seul point',

Ex. Math., 1, May 1826, pp. 29-43 (O.C., 2, 6, pp. 23-37).
2. 'Sur les moments lineaires', Ex. Math., 1, May 1826, pp. 65-84 (O.c., 2, 6, 89-112).
3. 'Sur les moments Iineaires de plusieurs forces appliquees a difTerents points', Ex.
Math., 1, June 1826, pp. 117-124 (O.c., 2, 6,149-158).
4. 'Usage des moments lineaires dans la recherche des equations d'equilibre d'un
systeme invariable entierement libre dans I'espace', Ex. Math., 1, June 1826, pp. 125-
132 (O.c., 2, 6, pp. 59-168).
5. 'Sur les conditions d'equivalence de deux systemes de forces appliquees ades points
lies invariablement les uns aux autres', Ex. Math., 1, July 1826, pp. 151-154 (O.c., 2,
6, pp. 191-195).
6. 'Usage des moments Iineaires dans la recherche des equations d'equilibre d'un
systeme invariable assujetti acertaines conditions', Ex. Math., 1, July 1826, pp. 155-
159 (O.C., 2, 6, pp. 196-201).
260 Notes

7. 'Recherche des equations generales d'equilibre pour un systeme de points materiels

assujettis a des liaisons quelconques', Ex. Math., 2, March 1827, pp. 1-22 (O.c., 7,
pp.11-36).
8. 'Sur les mouvements que peut prendre un systeme invariable libre ou assujetti a
certaines conditions', Ex. Math., 2, March 1827, pp. 70-90 (O.c., 2, 7, pp. 94-120).
9. 'Sur les moments d'inertie', Ex. Math., 2, April 1827, pp. 93-103 (O.c., 2, 7, pp. 124-
136).
10. 'Sur la force vive d'un corps solide ou d'un systeme invariable en mouvement', Ex.
Math., 2, April 1827, pp. 104-107 (O.c., 2, 7, pp. 137-140).
The library ofthe Ecole Poly technique has the first of Notes sur quelques Parties de
la M ecanique, which deals with the composition offorces and linear moments (A3 a
57). This note, entitled 'Sur la resultante de plusiers forces appliquees a un seul
points', was printed for distribution to the students, probably in the academic year
1826-1827. Its content corresponds almost exactly to the two articles, 1. and 2., of
the Exercices de Mathematiques.
57. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, December 1, 1826.
58. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, November 17, 1826.
59. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, December 14, 1827.
60. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, May 28, 1828.
61. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, December 26, 1828.
62. In the foreword to Lefons sur Ie Calcul Differentiel, Cauchy wrote:

The edition of Resume des Lefons sur Ie Calcul Infinitesimal, which appeared in 1823,
being found to be limited, I decided to replace it by two separate works, one covering
differential calculus and the other integral calculus. The present work, which treats
differential calculus, is the first of the volumes.
63. Archives E. P., Registre des proces-verbaux des seances du Conseil de Perfectionne-
ment, December 11, 1829.
64. See B. Belhoste, 'Le cours d'analyse de Cauchy al'Ecole Polytechnique en seconde
annee', Sciences et Techniques en Perspective, 9, 1984-1985, pp. 101-178.
65. Cauchy had written the manuscript of the third volume of Lefons sur les
Applications du Calcul Infinitesimal ala Geometrie. Moigno used this manuscript
in preparing Lefons de calcul differentiel, published in 1840 (see the introduction to
that work, pp. XVIII-XIX).

Notes to Chapter 6
1. On the history of wave theory, see H. Burkhardt, 'Entwicklungen nach oscill-
ierenden Funktionen und Integration der Differentialgleichungen der mathematis-
chen physik', lahresbericht der Deutschen Mathematiker-Vereinigung, 10, (1901-
1908) paragraph 43ff, and H. Lamb, Hydrodynamics, Cambridge, 1932, esp.
p.373ff.
2. J. L. Lagrange, M ecanique analytique, part two, Dynamique, sect. XI, paragraph II,
'Applications au mouvement d'un fluide contenu dans un canal peu profond et
presque horizontal, et en particulier au mouvement des ondes'.
 Notes 261

3. Proces-verbaux Ac. Sci., 5, p. 530, and Mem. Sav. Etr., 1, 1827, p. 188 (O.c., 1, 1,
p. 190).
4. These manuscript supplements are contained in Cauchy's Cahier sur la Theorie des
Ondes, which belongs to Mrs. de Pomyers:

pp. 90-97: 'Du cas ou l'on a egard a la profondeur'.

pp. 98-114: 'Observations sur Ie memoire nO 2 presente au concours de 1815 relatira la
theorie des ondes'.

5. Poisson read his first paper on October 2,1815, the closing day of the competition,
and his second paper on December 18, 1815, i.e., a week before the awarding of the
prize. In the second paper, he established the existence ofa wave propagation with
a constant velocity. Poisson published his papers in Mem Ac. Sci., 1 (1816), 1818,
pp. 69-186. Cauchy's paper came out in the Mem. Sav. Etr., 1, 1827, pp. 3-312
(O.c., 1,1, pp. 5-318). The Cahier sur la Theorie des Ondes, belonging to Mrs. de
Pomyers, contains the manuscript text of the 13 notes of the prize-winning paper
and the text of notes XIV and XV, which Cauchy added in 1821. The manuscript text
of undated note XVI is in another cahier, kept at the Sorbonne Library, Ms 2057.
Note XVII is an improved version of a 1815's note (see Cahier sur la Theorie des
Ondes, pp. 142-152).
6. Proces-verbaux, Ac. Sci., 7, p. 370.
7. On the history ofthe theory of elasticity, see A. Barre de Saint-Venant, 'Historique
abrege des recherches sur la resistance des materiaux et sur l'elasticite des corps
solides', in his edition of Navier's Resume des Le(:ons sur l'Application de la
Mecanique, 1, Paris, 1858; 1. Todhunter and K. Pearson, A History of Elasticity and
of the Strength of Materials from Galilei to the Present Time, 1, Cambridge, 1886;
A. E. H. Love. A Treatise on the Mathematical Theory of Elasticity, Oxford, 1927,
Historical Introduction; pp. 1-31; and S. P. Timoshenko, History of Strength of
Materials, with a Brief Account of the History of Theory of Elasticity and Theory of
Structures, New York, McGraw-Hill, 1953. On Cauchy's research works, see A.
Dahan-Dalmedico, 'La mathematisation des theories de l'eJasticite par A. L.
Cauchy et les debats dans la physique mathematique franc;aise (1800-1840)"
Sciences et Techniques en Perspective, 9, 1984-1985, pp. 1-100.
8. See Ch. 1, p. 14.
9. Proces-verbaux, Ac. Sci., 6, pp. 474-477 (O.c., 2, 15, pp. 539-544).
10. Mem. Institut, 2, (1812), 1816, pp. 167-226.
11. S. Germain, Recherches sur la Theorie des Plaques Elastiques, Paris, 1821. Short
letter of acknowledgment from Cauchy to Sophie Germain under the date of July
24,1821, BN, Ms ffr 9118.
12. The paper 'Sur la flexion des plans elastique', has never been printed, except as
an abstract, published in the Bull. Phil., 1823, pp. 95-102. The manuscript is not
kept in the archives of the Academie, but the EN PC Library has a lithogra-
phic copy of the paper (Ms Navier 1820; see also Navier file Arch. Nat. F14
2289 1).
13. L. Bucciarelli and N. Dworsky analyze this part of the paper in their book Sophie
Germain, an Essay on the History qf Elasticity, Dordrecht, Reidel Publishing
Company, 1980, Ch. 9, nO 3, pp. 139-140.
14. See C. Truesdell, 'The creation and unfolding ofthe concept of stress', in Essays in
the History of Mechanics, New-York-Berlin-Heidelberg, Springer-Verlag, 1968,
pp. 184-238.
262 Notes

15. See 'De la pression dans les fluides', Ex. math., 2, March 1827, pp. 23-24 (O.C., 7,
pp.37-39).
16. Bull. Phil., January 1823, p. 10 (O.C., 2, 2, p. 301).
17. See 'Supplement au memoire sur la double refraction' of January 13, 1822,
presented to the Academie on January 22 (A. Fresnel, Oeuvres, 2, no XLII, pp. 344-
347) and 'Second supplement au memoire sur la double refraction' of March 31,
1822, presented to the Academie on April 1, 1822 (A. Fresnel, Oeuvres, 2, no XLIII,
pp. 369-442). Cauchy set forth Fresnel's argument in the addition to his article 'De
la pression ou de la tension dans les corps solides', Ex. Math., 2, March 1827,
pp. 56-59 (O.c., 2, 7, pp. 79-81).
18. Bull. Phil., January 1823, pp. 11-12 (O.C., 2, 2, p. 303.
19. H. Freudenthal, 'Cauchy, Augustin-Louis', Dictionary of Scientific Biography, 3,
New York, 1971, pp. 131-148, especially p. 145.
20. Cauchy deduced the properties of the perfect fluids from the general laws of
continuum mechanics in the paper 'Sur les equations qui expriment les conditions
d'equilibre ou les lois du mouvement des fluides', Ex. Math., 3, March 1828, pp. 42-
57 (O.c., 2,8, pp. 128-146).
21. Abstract in Bull. Phil., January 1823, pp. 9-13 (O.c., 2, 2, pp. 300-304). See also the
Registre des seances de la Societe Philomatique, 1822-1823, Ms 2086, fo 60 r. and fo
61 v, at the Sorbonne Library.
22. See the following articles:
1. 'Surla pression ou tension dans les corps solides', Ex. Math., 2, March 1827, pp. 42-
57 (O.c., 2, 7, pp. 60-78). Cauchy deposited the manuscript with the bureau of the
Academie on March 12, 1827. It is kept in the meeting's packet. The subtitle of the
manuscript, 'Extrait de la premiere partie du memoire sur l'equilibre et Ie
mouvement interieur des corps solides ou fluides elastiques ou non elastiques',
proves that this article is an extract of the famous unpublished paper of September
30, 1822.
2. 'Sur la condensation et la dilatation des corps solides', Ex. Math., 2, April 1827,
pp. 60-69 (O.c., 2, 7, pp. 82-93).
3. 'Sur les relations qui existent, dans l'etat d'equilibre d'un corps solide ou fluicie, entre
les pressions ou tensions et les forces acceleratrices', Ex. Math., 2, April 1827,
pp. 108-111 (O.c., 2, 7, pp. 141-145).
4. 'Sur les equations qui expriment les conditions d'equilibre ou les lois du mouvement
interieur d'un corps solide, elastique ou non elastique', Ex. Math., 3, September
1828, pp. 160-187 (O.c., 2, 8, pp. 195-226). In this article, Cauchy presented both of
his continuum theories (for isotropic media), one with a single coefficient of elasticity
(from 1822) and one with two coefficients (from 1828). The draft, deposited in a
sealed envelope on August 18, 1828, is kept in the meeting's packet. The text is
identical with the printed article.
23. See Proces-verbaux, Ac. Sci., 7, p.420. The manuscript, which was initialed by
Fourier on January 27, 1823, was written in 1815. It is inserted in the Cahier sur la
Theorie des Ondes belonging to Mrs. de Pomyers, pp. 115-140.
24. The letter is kept in the meeting's packet of October 6, 1822.
25. Sorbo nne Library, Ms 2086, Registre des seances de la Societe Philomatique, 1822-
1823, fo 62 v.
26. Bull. Phil., March 1823, pp. 36-37.
27. All the authors who have written about the origins of the theory of elasticity and
compared the works of Navier and Cauchy repeat this mistake. Only Prony,
Poisson, and Fourier were actually members of the commission.
 Notes 263

28. Histoire de I'Academie, partie mathematique, 1822, pp. 30-31.

29. Bull. Phil., January 1823, p. 13 (D.C., 2, 2, p. 301), and Histoire de l'Academie, partie
mathematique, 1822, p. 31.
30. Mem. Ac., Sci., 7 (1824), 1827, pp. 375-394.
31. On November 17, 1823 (Proces-verbaux, Ac. Sci., 7, p. 585), while preparing the
publication of his award-winning paper of 1815, 'Sur la theorie des on des', he
presented to the Academie the paper 'Sur les effets de l'attraction moleculaire dans
Ie mouvement des ondes', in which the term molecule was used for the very first
time to mean not an infinitely small material element but a center of infinitely small
attractive action [the study, quoted in note XX of the paper 'Sur la theorie des
ondes', is unpublished; the manuscript abstract is kept in the meeting's packet of
November 17, 1823 (see the appendix I, p.298)]. In October 1827, Cauchy
specifically stated that he had been occupied 'for quite some time' with molecular
theory [Analyse des Travaux de l'Academie ... 1827, Mem Ac. Sci., 10 (1827),1831,
iii-iv].
32. Proces-verbaux, Ac. Sci., 8, p. 603.
33. Sealed envelope no 126, kept in the meeting's packet of October 1, 1827. The sealed
envelope was opened in October 5, 1977, and the paper published by C. Truesdell.
See C.R. Ac. Sci., 291, Vie academique, pp. 33-46.
34. See Proces-verbaux, Ac. Sci., 9, p. 52 and p. 55.
35. Poisson offered an offprint of his paper to the Academie on November 3, 1828
(Proces-verbaux, Ac. Sci., 9, p. 137). The paper came out in the Mem. Ac. Sci., 8,
1829, pp. 751-780.
36. Ex. Math., 3, September 1828, pp. 188-212, and October 1828, pp. 213-236 (D.C.,
2,8, pp. 227-252 and 252-277).
37. Sealed envelope no 128, registered on August 18, 1828, and opened on October 5,
1977. Cauchy authorized unsealing the envelope on September 8, 1828, after
having published the paper (Proces-verbaux, Ac. Sci., 9, p. 114).
38. Formally, Navier's equation for isotropic bodies can be deduced from Eq. (6.13) by
assuming G = 2R.
39. See the following articles:
1. 'Sur l'equilibre et Ie mouvement d'une lame solide' and 'Addition a l'article
precedent', Ex. Math., 3, September 1828, pp.245-326, and December 1828,
pp.326-327 (O.c., 2, 8, pp.288-380). Cauchy presented the paper 'Sur la
mouvement des lames elastiques ou non elastiques naturellement planes ou
naturellement courbes, d'une epaisseur con stante ou d'une epaisseur variable' to the
Academie on October 6,1828 (Proces-verbaux, Ac. Sci., 9, p. 130), and an addition to
this paper on October 13, 1828 (Proces-verbaux, Ac. Sci., 9, p. 131). On November
17, 1828, he presented a new paper, 'Sur les vibrations des lames courbes'.
2. 'Sur l'equilibre et Ie mouvement d'une plaque solide', Ex. Math., 3, December 1828,
pp.328-355 (O.c., 2, 8, pp.381-411). Cauchy presented the paper 'Sur Ie
mouvement des plaques et des verges elastiques ou non elastiques' on October 6,
1828 (Proces-verbaux, Ac. Sci., 9, p. 130). Two manuscripts are kept in the meeting's
packet:
'Premiere partie d'un memoire sur les equations d'equilibre ou de mouvement d'une
plaque solide naturellement droite et d'une epaisseur constante'.
'Sur J'equilibre et Ie mouvement d'une plaque solide'.
3. 'Sur J'equilibre et Ie mouvement d'une verge rectangulaire', Ex. Math., 3, December
1828, pp. 356-368 (O.c., 2, 8, pp.412-423). Cauchy presented the paper 'Sur Ie
264 Notes

mouvement des plaques et des verges elastiques ou non elastiques' on October 6,

1828 (Proces-verbaux, Ac. Sci., 9, p. 130). A manuscript is kept in the meeting's
packet:
'Sur l'equilibre et Ie mouvement d'une verge rectangulaire'.
On December 1, 1828, Cauchy presented a new paper, 'Sur l'equilibre et Ie
mouvement des verges elastiques rectangulaires, droites ou courbes, d'epaisseur
constante ou d'epaisseur variable', to the Academie.
40. See the following articles:
1. 'Sur l'equilibre et Ie mouvement d'une plaque elastique don! I'elasticite n'est pas la
memedans tous les sens', Ex. Math., 4, February 1829,pp. 1-14(O.C., 2,9, pp. 9-22).
Cauchy presented the paper to the Academie on February 26,1829 (Proces-verbaux,
Ac. Sci., 9, p. 187). A manuscript is kept in the meeting's packet:
'Sur l'equilibre et Ie mouvement des corps elastiques des plaques elastiques, etc... ,
dont l'elasticite n'est pas la meme dans tous les sens'.
2. 'Sur l'equilibre et Ie mouvement d'une verge rectanguiaire extraite d'un corps solide
dont l'elasticite n'est pas la meme en tous sens', Ex. Math., 4, February 1829, pp. 15-
29 (D.C., 2, 9, pp. 23-40). Cauchy presented the paper 'Sur Ie mouvement des lames
de surface et des verges elastiques lorsque l'elasticite n'est pas la meme dans tous les
sens' to the Academie on December 22, 1828 (Proces-verbaux, Ac. Sci., 9, p. 167).
Two manuscripts are kept in the meeting's packet of October 6, 1828:
'Sur Ie mouvement des lames de surface et des verges elastiques lorsque l'elasticite
n'est pas la meme dans tous les sens'.
'Sur l'equilibre et Ie mouvement d'une verge rectangulaire dont l'elasticite n'est pas
la meme en tous sens'.
One manuscript is kept in the meeting's packet of December 22, 1828:
'Lame elastique dans Ie cas ou l'elasticite varie dans toutes les directions'.
3. 'Sur les vibrations longitudinales d'une verge cylindrique ou prismatique a base
queIconque', and 'Sur la torsion et les vibrations tournantes d'une verge rectan-
gulaire', Ex. Math., 4, February 1829, pp. 43-46 and pp. 47-64 (O.C., 2, 9, pp. 56-60
and pp.61-86). Cauchy presented the paper 'Sur la torsion et les vibrations
tournantes d'une verge rectangulaire' on February 2 and 16, 1829 (Proces-verbaux,
Ac. Sci., 9, p. 190 and p. 196). One manuscript is kept in the meeting's packet of
October 6, 1828: 'Vibration tournante d'une verge rectangulaire'. Another manu-
script is kept in the meeting's packet of February 16, 1829: 'Sur la torsion et les
vibrations tournantes d'une verge rectangulaire'.
41. Cauchy explains his general molecular theory of elasticity in the article 'Sur les
equations differentielles d'equilibre ou de mouvement pour un systeme de points
materiels sollicites par des forces d'attraction ou de repulsion mutuelle', Ex. Math.,
4, August 1829, pp. 129-139 (O.c., 2, 9, pp. 162-173).
42. See especially the unpublished notes of Cauchy and Poisson in the meeting's
packets of April 6, 13, and 20, 1828.
43. On the theory oflight, see E. T. Whittaker, A History of the Theories of Aether and
Electricity, 1, London, New-York, T. Nelson, 1951. On Cauchy's research works
especially, see J. Z. Buchwald, 'Optics and the Theory of the Punctiform Ether',
Arch. Hist. Exact Sci., 21, 1980, pp. 245-278.
44. See R. H. Silliman, 'Fresnel Augustin', Dictionary of Scientific Biography, 5, New
York, 1972, pp. 166-171.
 Notes 265

45. 'Sur les mouvements d'un systeme de molecules qui agissent les unes sur les autres Ii
de tres petites distances et sur Ie mouvement de la lumiere', Bull. Ferussac, 11,
February 1829, pp. 111-112, and Mem. Ac. Sci., 9 (1826),1830, pp.114-116.
46. 'Sur l'integration d'une certaine c1asse des equations aux differences partielles et
sur les phenomenes dont cette integration sert Ii faire connaitre les lois'. This
unpublished paper is lost, but a note is kept in the meeting's packet of April 12,
1830. Cauchy published an abstract ofthe paper in Bull. Ferussac, 13, April 1830,
pp. 273-279, and the first paragraph in J.E.P., 13, February 1831, pp. 175-221. It
seems that Cauchy had already treated the subject in a paper he presented to the
Academie on March 22, 1830 (Proces-verbaux. Ac. Sci.,9, p. 423).
47. 'Application des formules qui representent Ie mouvement d'un systeme de
molecules sollicitees par des forces d'attraction ou de repulsion mutuelle Ii la
theorie de la lumiere', Ex. Math., 5, September 1830, pp. 19-72 (O.C., 2, 9, pp. 390-
450). See also Memoire sur la Theorie de la Lumiere. This detached paper is
reproduced in the Bull. Ferussac, 13, 1830, pp. 414-427 (O.C., 2, 2, pp. 119-133),
and in Mem Ac. Sci., 10(1827), 1830,pp. 293-316 (O.c., 1,2,pp. 91-110), with some
unimportant variants. Cauchy presented two parts ofthe paper to the Academie on
June 7 and 14, 1830.
48. From his lectures given at the College de France, Cauchy published two
lithographies:

Extrait des lefons domnees au College de France. Resume de la lefon de M. A. L.

Cauchy du samedi 8 mai 1830. An exemplar of this lithography is kept in the meeting's
packet of May 17, 1830. As for the manuscript, it is kept in the meeting's packet of May
10, 1830.

Extrait des lefons donnees au College de France par M. A. L. Cauchy sur la theorie de
la lumiere a dater du 8 mai 1830. Refraction et reflexion de la lumiere. Dispersion de la
lumiere. An examplar of this lithography, with autographic corrections, is kept in the
meeting's packet of June 21, 1830. A manuscript copy, not autographic and without the
paragraph on light dispersion, and two notes to the compositor of mathematics for the
M em Ac. Sci. are kept in the same packet. In fact, Cauchy prepared the publication of
his lectures in the M em Ac. Sci. He intended to entitle his paper 'Second memoire sur la
lumiere'.

49. Cauchy published the beginning of his paper 'Sur la dispersion de la lumiere' in
1830. In 1835, he published the whole paper at Prague, in Nouveaux Exercices de
Mathematiques, pp. 1-60 (O.c., 2, 10, pp. 196-260).
50. 'Sur l'equilibre et Ie mouvement interieur des corps consideres comme des masses
continues', Ex. Math., 4, May 1830, pp. 293-319 (O.c., 9, pp. 342-372).
51. Sept Le{:ons de Physique Generale ... , Paris, 1868, published by F. Moigno (O.C., 2,
15, pp. 412-447).

Notes to Chapter 7
1. This study, the manuscript for which could not be located, was published only in
1827 in the Memoires des Savants Etrangers, 2, 1, 1827, pp. 599-799 (O.c., 1, 1,
pp. 329-506) at the same time as the study 'Sur la theorie des ondes'. The long delay
between the presentation and the publication ofthe study can be explained by the
slowness that, since 1811, had characterized the Academie's publication of its
266 Notes

records and materials. The main results of Cauchy's study were known as far back
as the end of 1824 because of a rather lukewarm report that was published by
Poisson in the Bull. Phil., December 1814, pp. 185-188 (O.c., 2, 2, pp. 194-198).
The report of the Academie's evaluative commission was published in the Analyse
des Travaux de l'Academie des Sciences pour l'annee 1814. Finally, Cauchy added
two supplements, which were written in 1814, to the manuscript, following a
request from the commissioners. Moreover, a large number of important notes
were appended to this work, probably in 1825. On September 14, 1825, Cauchy
registered this paper with the secretariat of the Academie, so that it could be
published along with the additions. An analysis of the study can be found in the
general work& on Cauchy; a more detailed study is given in the article by H. J.
Ettlinger, 'Cauchy's 1814 paper on definite integrals', Annals of Mathematics, 23,
1921-1922, pp. 255-270.
2. Shortly before Cauchy, Laplace and Poisson published works with the same titles:
Laplace, 'Memoire sur les integrales definies et leur application aux probabilites',
Mem. Institut, 10, 1810-1811, pp. 279-347 (P. S. Laplace, Oeuvres, pp. 357-412);
Poisson, 'Memoire sur les integrales definies', first part, J.E.P., 9, 16th cahier, 1813,
pp. 215-246. As for Legendre, in 1811, he published the first volume of Exercices de
Calcul Integral.
3. Published in the Memoires Ac. Sci., 1782-1785, pp. 1-88 (Oeuvres, 10, pp. 209-
291). A sequel was published in Mem. Ac. Sci., 1783-1786, pp.423-467 (P. S.
Laplace, Oeuvres, 10, pp. 295-338).
4. Laplace, 'Memoire sur les integrales definies', Mem. Inst., 11, 1810-1811, p. 284
(P. S. Laplace, Oeuvres, 12, p. 361).
Cauchy cited this passage in his study, p. 612 (O.c., 1, 1, pp. 329-330). In his
1810 paper, Laplace used established, orthodox methods to recalculate certain
formulas that he had derived from the passage from the real to the imaginary
domain in 1782. He pursued this work in his additions to the second edition of
Theorie Analytique des Probabilites (November 1814).
5. This study was published only in 1844 in the J.E.P., 18, 28th Cahier, pp. 147-248,
(O.c., 2,1, pp. 467-567). In a note, Cauchy pointed out that the study of January 2,
1815, had received certain additions 'about this same time', Cauchy, in fact,
presented to the Academie an improved version of this study, which had the same
title as the original work, on April 1, 1816 (Proces-verbaux, Ac. Sci., 6, p. 44). It was,
no doubt, this later version that he published in 1844. The manuscript of the study
could not be located. Meanwhile, Cauchy deposited two abstracts of it with the
Academie on September 26, and November 7, 1825. The first reproduced the
introduction to the paper, while the second reproduced the section 'Sur la
conversion des differences finies des puissances en integrales definies'. The
manuscripts are kept in the meeting's packet.
6. See Memoires Sav. Err., 2, 1, 1827, note IV, pp. 140-145 (O.c., 1, 1, pp. 133-
139).
7. On Cauchy's teaching at the College de France in 1817, see Bull. Phil., Oct.
1822, p. 161 and p. 171 (O.c., 2, 2, p. 283 and p.295); J. E. P., 12, 9th cahier,
July 1823, p. 576 (O.c., 2, 1, p. 339); Memoires Sav. Etr., 2, 1, 1827, p. 715, foot-
notes (O.C., 1, 1, p.429, footnote); and Ex. Math., 2, 1827, p. 156 (O.c., 2, 7,
p. 194).
8. 'Sur la reduction des integrales finies et des sommes de series en integrales definies',
cited by J. Mandelbaum, La Societe Philomatique de Paris, 1, p. 200. It remained
 Notes 267

unpublished. As to its contents, see Bull. Phil., August 1817, pp. 123-124 (D.C., 2, 2,
pp.226-227).
9. Proces-verbaux, Ac. Sci., 6, p. 201. The manuscript could not be located. Cauchy
published this paper in August 1817: 'Sur une loi de reciprocite qui existe entre
certaines fonctions', Bull. Phil., August 1817, pp. 121-124 (D.C., 2,2, pp. 223-227).
10. See Appendix I, Registre d'instruction, 1817-1818, p. 313.
11. See Fourier's letter to the Permanent Secretary ofthe Academy in Fourier's papers,
BN Ms ffr 22529, p. 127.
12. 'Seconde note sur les fonctions reciproques', Bull. Phil., December 1818, pp. 178-
181 (D.C., 2, 2, pp. 228-237).
13. At first, Cauchy used an integral formula representing the cosine function, which
had been obtained by use of the theory of singular integrals, in order to integrate
the differential equations of the theory of waves. See 'Memoire sur la theorie des
ondes', note XVI, p. 187, and note XVIII, pp. 292-293 (D.C., 1,1, pp. 189-190 and
pp. 295-297; these notes were written for publication).
14. 'Sur la resolution analytique des equations de tous les degres par Ie moyen des
integrales definies', unpublished (see Proces-verbaux, Ac. Sci., 6, p. 507). The
manuscript could not be located. An announcement was published in Analyse des
Travaux de I'Academie des Sciences, partie mathematiques, 1819, pp.8-11,
republished in Mem. Ac. Sci., 4, (1819-1820),1824, pp. XXVI-XXIX (D.C., 1,2,
pp. 9-11). See also Bull. Phil., October 1822, p. 168, (D.C., 2, 2, pp. 293) and J.E.P.,
12, 9th cahier, July 1823, pp. 541-543 and pp. 580-581 (D.C., 2, 1, pp. 305-306 and
pp. 343-345). Among the studies in the initial project on the publication of
Cauchy's Deuvres Completes was a separate paper from 1819, entitled 'Observa-
tions sur les principes de la resolution des equations numeriques', which was
probably a copy of the paper of November 22,1819. This text could not be located
(see D.C., 2, 15; appendix, p. 584).
15. Proces-verbaux, Ac. Sci., 7, p. 380.
16. See Bull. Phil., October 1822, pp. 161-174 (D.C., 2, 2, pp. 283-299).
17. 'Sur l'integration des equations lineaires aux differences partielles et Ii coefficients
constants', J.E.P., 12, 9th cahier, July 1823, pp. 571-592 (D.C., 2, 1, pp. 333-357).
18. Resume des Lefons donnees a l"Ecole Royale Poly technique sur Ie Calcul InfiniteSimal,
Calcul Integral, 24th, 25th, and 34th lectures.
19. J.E.P., 12, 19th cahier, July 1823, p. 574, footnote (D.C., 2, 1, p. 337, footnote).
20. Mem. Sav. Etr., 2, 1, 1827, note IX, p. 158 (D.C., 1, 1, p. 146).
21. See the 'Memoire sur l'integration de quelques equations lineaires aux differences
partielles et particulierement de l'equation generale du mouvement des fluides
elastiques', which was presented by Poisson on July 19, 1819, to the Academie and
published in Mem. Ac. Sci.,3 (1818), 1820, pp. 121-176.
22. See Proces-verbaux Ac. Sci., 7, p. 231. The paper was published, in part, in Analyse
des Travaux de I'Academie des Sciences, partie mathematique, 1821, pp. 25-32, and
in Bull. Phil., October 1821, pp. 101-112, and November 1821, pp. 145-152 (D.C.,
2,2, pp. 253-275).
23. This paper was presented to the Academie on January 22, 1822 (Proces-verbaux,
Ac. Sci., 7, p. 271), and published in Analyse des Travaux de I'Academie des Sciences,
partie mathematique, 1821, pp. 6-13 [see also Memoires Ac. Sci., 5 (1821), 1826,
pp. 13-19], and in Bull. Phil., April 1822, pp. 49-54 (D.C., 2, 2, pp. 276-282).
24. See Bull. Phil., November 1821, p. 152 (D.C., 2, 2, p. 275). This issue ofthe Bulletin
appeared at the beginning of 1822.
268 Notes

25. Proces-verbaux, Ac. Sci., 7, p. 366. This study constitutes the substance of the
article of the J.E.P. cited above in note 17; see also the announcement in Appendix
I, p. 296.
26. 'Moyen d'integrer les equations lineaires aux differences totales ou partielles, finies
ou infiniment petites, avec un dernier terme variable et d'un ordre que1conque,
dans tous les cas possibles, lorsque les coefficients du premier membre sont
constants et dans certains cas, lorsque les coefficients varient, sans etre oblige de
resoudre aucune equation algebrique' (Proces-verbaux, Ac. Sci., 7, p. 503), for
which the first three sections were given in the article of the J.E.P. cited above in
note 17. The fourth section appeared in Memoires Ac. Sci., 9 (1826), 1830, pp. 97-
103 (D.C., 1,2, pp. 67-72), and in Bull. Fer., 4, August 1825, pp. 71-75 (D.C., 2, 2,
pp. 66-71). On the same day, May 26,1823, Cauchy presented another study, 'Sur
la determination des integrales definies et sur la resolution des equations
algebriques ou transcendantes par Ie moyen de ces memes integrales', as 'a
complement to the papers that the author presented in 1814, 1819 and 1822'
(Proces-verbaux, Ac. Sci., 7, p. 503). This study remained unpublished and the
manuscript could not be located. One month later, on July 21, 1823, Cauchy
presented to the Academie a paper entitled 'Divers theoremes servant it integrer les
equations propres it la theorie analytique de la chaleur', which remained
unpublished.
27. 'Memoire sur la theorie des ondes', Memoires Sav. Etr., 2, 1, 1827, note XVIII,
pp. 281-293 (D.C., 1,1, pp. 288-299).
28. S. D. Poisson, 'Sur les integrales des fonctions qui passent it l'infini entre les limites
de l'integration, et sur l'usage des imaginaires dans la determination des integrales
definies', J.E.P., 11, 18th cahier, January 1820, pp. 295-341.
29. See Bull. Phil., November 1821, pp. 171-174 (D.C., 2, 2, pp. 296-299), and J.E.P.,
12, 19th cahier, July 1823, P.S., pp. 590-591 (D.C., 2, 1, pp. 354-355).
30. See R. Taton, Brisson, Barnabe, Dictionary of Scientific Biography, 2, New York,
1970, pp. 473-475.
31. As to the dispute between Cauchy and Poisson, see Ch. 4, pp. 51-52. Cauchy's
report is published in the Proces-verbaux, Ac. Sci., 8, pp.223-226 (D.C., 2, 15,
pp.560-565).
32. 'Sur l'integration des equations lineaires et leur application it divers problemes de
physique', was published without the applications in Mem. Ac. Sci., 22, 1850,
pp. 39-130 (D.C., 1,2, pp. 195-281) under the title 'Memoire surle calcul integral'.
A part of this study was published in lithograph on May 2, 1825, under the title
'Memoire sur l'analogie des puissances et des differences et sur !'integration des
equations lineaires' (D.C., 2, 15, pp. 23-40). As to the applications, see Mem. Ac.
Sci., 7 (1824),1827, Histoire de l'Academie, partie mathematique, pp. XLV-XLVI
(prepared by Cauchy from the handwritten copy kept in meeting's packet of May
30, 1825), and the unpublished announcement of the study 'Sur l'integration des
equations lineaires et sur Ie mouvement des plaques elastiques rectangulaires' of
January 17, 1825 (See Appendix I, pp. 300-301).
33. Proces-verbaux, Ac. Sci., 8, p. 225 (D.C., 2, 15, p. 564).
34. Undoubtedly, the 'Recherches sur la determination des senes qui doivent
representer des fonctions donnees dans une partie seulement de leur etendue' was
presented to the Academie on August 27,1827, and never published. See Memoire
 Notes 269

Sur les integrales definies prises entre des limites imaginaires, p.2 (D.C., 2, 15,
p. 42), and Appendix I, p. 302.
35. Memoire sur les integrales definies prises entre des limites imaginaires, p. 2 (D.C., 2,
15, p. 42), and Appendix I, p. 302.
See A. Iushkevitch, Ostrogradski, Dictionary of Scientific Biography, 10, New
York, Chales Scribner's Sons, 1974, pp. 247-251, and Michel Dstrogradski et Ie
progres de la science au XIXeme siecle Conference du Palais de la Decouverte. Paris
1966.
36. Unpublished. The handwritten manuscript is kept in the meeting's packet for
February 13, 1826.
37. Unpublished. The handwritten manuscript is kept in the Ostrogradski file at the
Academie des Sciences.
38. See Appendix I, p. 302.
39. Ex. Math., t, March 1826, pp. 11-24 (D.C., 2, 6, pp. 23-37). As to Cauchy's teaching
at the College de France in 1824-1825, see Chapter 4, p. 49.
40. Unpublished. However, there is a handwritten announcement in the meeting's
packet of February 25, 1825 (see Appendix I, p. 301).
41. In the meeting's packet of February 28, there is a handwritten manuscript of the
paper, with the corrections. It is identical to the printed edition, but does not
contain the addition. Undoubtedly, this manuscript was deposited with the bureau
of the Academie at the meeting of February 28. The minutes of the meeting only
indicate that Cauchy read a paper on analysis that day. Moreover, the manuscript
was neither dated nor initialled by the secretary of the Academie, as was
customary. This manuscript is quite likely the study registered with the Academie
on August 8, 1825, the printed edition having been sent the following week (see
Proces-verbaux, Ac. Sci., 8, p. 189 and p. 192). Cauchy published a resume of the
paper in Bull. Fer., 3, April 1825, pp. 214-221 (D.C., 2, 2, pp. 57-65 and D.C., 2, 15,
pp.41-89).
42. Memoire sur les integrales definies prises entre des limites imaginaires, p. 26 (D.C., 2,
15, p. 59).
43. In these notes, Cauchy discusses the progress that had been made since 1814 on the
theory of singular integrals, and he used his theory of imaginary integrals from
1822-1823 to rewrite the double equations connecting the real integrals.
44. 'Memoire sur les integrales definies ou l'on donne une formule generale de laquelle
se deduisent les valeurs de la plupart des integrales definies deja connues et celles
d'un grand nombre d'autres', Annales de mathematiques, 16, no 4, October 1825,
pp. 97-108 (D.C., 2, 2, pp. 343-352). .
45. 'Sur diverses relations qui existent entre les residus des fonctions et les integrales
definies', Ex. Math., 1, June 1826, pp. 95-124 (D.C., 2, 6, pp. 124-145).
46. The method of decomposing rational fractions by means of the calculus of residues
is given in the article 'Sur un nouveau genre de calcul analogue au calcul
infinitesimal', Ex. Math., 6, March 1826, pp. 11-24 (D.C., 2, 6, pp. 23-37). Cauchy
resumed the theory of extraordinary integrals by using the formalism of the
calculus of residues in the article 'Sur un nouveau genre d'integrales', Ex. Math., 1,
May 1826, pp.57-65 (D.C., 2, 6, pp.78-88). He extended the method of
decomposing to cover the case offunctions that have an infinite number of poles in
the article 'Sur diverses relations qui existent entre les residus des fonctions et les
270 Notes

integrales definies', Ex. Math., 1, June 1826, pp. 95-124 (D.C., 2, 6, pp. 124-145).
See also the two following articles in Ex. Math.:
1. 'Usage du calcul des residus pour la sommation de plusieurs suites composees d'un
nombre fini de termes', which was presented to the Academie on December 26,1825.
The manuscript is contained in the meeting's packet for that day. A corrected and
completed version of this study was published in Ex. Math., 1, May 1826, pp. 44-53
(O.c., 2, 6, pp. 62-73).
2. 'Sur Ie developpement des fonctions d'une seule variable en fractions rationnelles',
was presented to the Academie on December 10, 1827. The handwritten manuscript
is contained in the meeting's packet for that day. The study was published in Ex.
Math., 2, November 1827, pp. 277-296 (O.c., 2, 7, pp. 324-344). In 1843, Cauchy
designated the function w(x) as a complementary function (C.R., 19, p. 138; a.c., 1,8,
p.361).
47. 'Sur les limites placees it droite et it gauche du signe E dans Ie calcul des residus', Ex.
Math., 1, September 1826, pp. 205-232 (D.C., 2, 6, pp. 256-286).
48. 'Sur quelques propositions fondamentales du calcul des residus', Ex. Math, 2,
November 1827, pp.245-276 (D.C., 2, 7, pp.291-323), was presented to the
Academie on November 5, 1827. The handwritten announcement is in the
meeting's packet for that day. See also the articles, 'Usage du calcul des residus
pour la sommation ou la transformation des series dont Ie terme general est une
fonction paire du nombre qui represente Ie rang de ce terme', Ex. Math., 2,
December 1827, pp. 298-314 (D.C., 2, 7, pp. 345-362), which was presented to the
Academie on December 17, 1827 (the manuscript is contained in the meeting's
packet for that day), and 'Methode pour developper des fonctions d'une ou de
plusieurs variables en series composees de fonctions de meme espece', Ex. Math., 2,
December 1827, pp. 317-340, (D.C., 2, 7, pp. 366-392).
49. 'Usage du calcul des residus pour determiner la somme des fonctions semblables
des racines d'une equation algebrique ou transcendante', Ex. Math., 1, January
1827, pp. 339-357 (D.C., 2, 6, pp. 401-420).
50. See the following articles in Ex. Math.:
1. 'Application du calcul des residus a I'integration des equations differentielles
lineaires a coefficients constants', Ex. Math., 1, July 1826, pp. 202-204 (D.C., 2, 6,
pp.252-255).
2. 'Application du calcul des residus a I'integration de quelques equations differen-
tielles lineaires a coefficients variables', Ex. Math., 1, October 1826, pp.262-264
(O.c., 2, 6, pp. 316-319).
3. 'Sur la determination des constantes arbitraires renfermees dans les integrales des
equations differentielles lineaires', Ex. Math., 2, March 1827, pp. 25-27 (O.C., 2, 7,
pp.4O-54).
4. 'Sur la transformation des fonctions qui representent les integrales generales des
equations differentielles lineaires', Ex. Math., 2, October 1827, pp. 211-220 (O.c., 2,
7, pp. 255-266).
51. See the following two papers:
1. The paper 'Usage du calcul des residus pour la solution des problemes de physique
matbematique' was presented to the Acadernie on December 26, 1826 (a hand-
written abstract dated December 26, 1826, is contained in the meeting's packet for
that day), and the full version of this study was presented on February 5,1827. The
incomplete manuscript is in the meeting's packet for the February 5, 1827. The first
part of this study was published separately in January 1827, the second part
appeared in February 1827 (O.c., 2, 15, pp. 90-137).
2. The paper 'Deuxieme memoire sur I'application des residus aux questions de
 Notes 271

physique mathematique' was presented to the Academie on September 17, 1827. The
manuscript is contained in the meeting's packet for the day. The study was published
in Memoires Ac. Sci.,7 (1824), 1827, pp. 463-472 (D.C., 1,2, pp. 20-28).
52. See Chapter 6, p. 105.
53. 'Memoire sur les developpements des fonctions en series periodiques', presented to
the Academie on February 27, 1826 (the manuscript is in the meeting's packet for
that day). This study was published in Memoires Ac. Sci., 6 (1823),1827, pp. 603-
612 (D.C., 1,2, pp. 12-19).
54. G. P. Dirichlet 'Sur la convergence des series trigonometriques qui servent a
representer une fonction arbitraire entre des limites donnees', Journal de Crelle, 4,
1829, pp. 157-169.

Notes to Chapter 8
1. See Ch. Magnin, 'Note biographique sur M.1. de Bure', Journal general de
l'Imprimerie et de la Librairie, Feuilleton, July 17, 1847 p. 240, and A. Delavenne,
Recueil Genealogique de la Bourgeoisie Ancienne, Paris, 1954.
2. Marriage contract between A. L. Cauchy and A. de Bure, Arch. Nat., Minutier
central des notaires, Etude LXXIII, 1260, April 4, 1818.
3. Vers a['occasion du mariage de M. A. L. Cauchy avec Melle Aloise de Bure, Paris,
undated.
4. C. A. Valson, La Vie et les Travaux du Baran Cauchy, 1, p. 69.
5. F. R. de Lamennais, Correspondance Generale, 3 (1825-June 1828), 1971, docu-
ment 28, letter from Abbe Gerbet to M. de SenlTt, dated August 19, 1826, Paris.
6. See the letter from Cauchy to Libri, March 28, 1828, in Appendix III, p. 324.
7. See Arch. Nat. F 1dIVC 4, request for the Legion d'Honneur, and Arch. Nat.
F1411872. Letters from L. F. Cauchy requesting the Legion d'Honneur and
promotion to the rank of engineer ordinary first class on behalf of A. L. Cauchy are
found here. Since 1816, Cauchy was no longer in practice as engineer.
8. See 1. B. Duroselle, 'Les "filiales" de la Congregation', Revue d'Histoire Ecclesiasti-
que, 1950, p. 867, and Arch. Nat. F7 6699, file 1, Societe des Bonnes Etudes.
9. F. R. de Lamennais, Correspondance Generale, 1, (1805-1819), Paris, 1971, p. 423,
letter 276 to Brute, dated July 22,1818, where he wrote 'Teysseyre, Binet, Cauchy,
etc., are busy from dawn to dusk with charitable works'. According to C. A. Valson,
La Vie et les Travaux du Baron Cauchy, 1, pp. 195-196, Cauchy was especially
involved in the religious instruction of young chimney sweeps.
10. J. B. Duroselle, 'Les "filiales" de la Congregation', pp. 878-880.
11. For more on this school of counterrevolutionary thought, see D. Bagge, Les Idees
Politiques en France so us la Restauration, Paris, 1952.
12. The following handwritten note is kept at the Academie in the folder of the
meeting's packet for July 19, 1824:

I make no pretensions to having a knowledge of anatomy. But, I do swear that I was

singularly shocked to have here heard praise given to the system propounded by
Doctor Gall as concerns the brain. We all know how often his blunders have made him
truly ridiculous in the eyes of the whole world. I do not think they will be accepted at the
Academie; and what convinced me of this is precisely that the Academie has never put
him down on its list of candidates for the sections (departments) of medicine and
anatomy. Now, precisely what is the principle that M. Gall claims to have discovered?
It has long been known that the material brain is made of a great number of parts, and
272 Notes

this fact was never in question-even before Doctor Gall. But, what Doctor Gall did
not prove-and what he will never prove-is that the diversity of the parts of the
material brain destroys the unity of the 'self, that thought can be dissected and various
geometrical forms assigned to it. Such a principle is as repulsive to true philosophy as it
is to the very foundations on which the peace and happiness of society rests.
13. See Proces-verbaux, Ac. Sci., 7, p. 5.
14. See Proces-verbaux, Ac. Sci., 8, pp. 136-138 (~.C .. , 2, 15, pp. 557-560).
15. Memorial Catholique, 1st year, 2, 1824, p. 192.
16. New Monthly Magazine, June 1825 (see also Stendhal, Courrier anglais, 3, Paris,
1935, pp. 228-229).
17. New Monthly Magazine, November 1825 (see also Stendhal, Courrier anglais, 3,
Paris, 1935, pp. 228-229).
18. The following note is in Cauchy's handwriting: it is contained in the meeting's
packet for January 22, 1827, of the Academie:
M. Cauchy made some observations that have to do with this study. It should be noted
that the results obtained with respect to the departments that constitute Old Brittany
confirm the statements he made at one of the preceding meetings by showing the
beneficial influence that religious instruction has exerted on the habits and manners of
people. In fact, as to the province in question, the study indicates that the department of
Ille-et-Vilaine presents the most natural and unaffected manners and habits. But, it is
precisely in this department that schools directed by a religious order known as the
Petits Freres were were founded and operated most successfully; the Petits Freres go
forth into the countryside to be received by the poorest communities. The study that
was just read, M. Cauchy added, goes to show to what degree the religious
establishment of the Petits Freres is really worthy of the government's protection.
Cauchy showed less mistrust with regard to statistics than he did in the
introduction of Analyse Algebrique, stating that the statistical results seem to agree
with his own opinions.
19. See Geoffroy de Grandmaison, La Congregation (1801-1830), Paris, 1889.
20. For more on the association, J.B. Duroselle, art. cit., pp. 880-882, and G. Bertier de
Savigny, 'Le role des lalcs dans l'Eglise de France so us la Restauration', L'Anneau
d'Or, 32, March-April 1950, pp. 98-104.
21. See the prospectus of the College de Juilly, from the summer of 1828, in F. R. de
Lammenais, Correspondance Genera/e, 6, 1834-1835, Paris, 1977, document 33,
p.32:
Instruction in mathematics would come next. There would be established special
classes for those students who had the intent of taking the examinations for the naval
school, the military school, or the Ecole Polytechnique. Two scholars to whom we owe
our recognition, M. Ampere and M. Cauchy, both of whom are professors at the Ecole
Polytechnique, are willing to take charge of this important part of the instructional
program: they will enlighten the professors with their advice and they will keep abreast
of the students' progress by regular examinations.
22. F. R. de Lammenais Correspondance Generale, 6, (1834-1835), Paris, 1977, annexe,
p. 488 and p. 520.

Notes to Chapter 9
1. Liste des presences, Proces-verbaux, Ac. Sci., 9, p.748. Nevertheless, this list
indicates that Cauchy did not attend the meeting of August 23, 1830, even though,
 Notes 273

according to the minutes of the meeting, he read a paper 'Sur la dispersion de la

lumiere', on that day. This obvious contradiction leaves us uncertain about the
length of his absence.
2. According to the Proces-verbaux, Ac. Sci., 9, pp. 748-753, from September 6 on,
Cauchy regularly missed the meetings of the Academie. Moreover, we know from
the Registry of Permits of Fribourg that on August 11, 1830, he obtained a passport
that was delivered in Paris and was valid until July 15, 1831. (See G. Castella,
'Documents inedits sur un projet de fonder une "Academie Helvetique" a Fribourg
en 1830', Revue dHistoire Ecclesiastigue Suisse, 21, 1927, p. 309, note 2.)
3. See, for example, C. A. Val son, La Vie et les Trauaux du Baron Cauchy, 1,
pp.73-77.
4. The circular from the Ponts et Chaussees requiring the loyalty oath is dated
September 4, 1830. At the Faculte des Scicnces, the professors took the oath on
September 18, 1830.
5. The expression is from R. Taton, 'Sur les relations scientifiques d' Augustin Cauchy
et d'Evariste Galois', Rev. Rist. Sci., 24,1971, pp. 123-148, esp, p. 125. This article
gives a portrayal of A. L. Cauchy in 1830.
6. Undated letter written shortly after the death of A. L. Cauchy's youngest brother,
Amedee, who was born in 1802 and died in 1831. Cited by C. A. Valson, La Vie et
les Travaux du Baron Cauchy Valson, 1, p. 76.
7. Arch. Nat. F14 2187 2 , Cauchy file.
8. On September 18, in a letter that Alexandre Cauchy wrote to the Dean of the
Facuite des Sciences, the 'short trip' became an 'extended trip' (Arch. Nat. AJ 16
5120, Proces-verbaux des deliberations de la Faculte des Sciences, September 18,
1830). On November 28, 1830, Alolse wrote that the 'extended trip' was now 'an
extended stay in Italy' (Arch. Nat. F14 2187 2 , Cauchy file).
9. For example, we know that Fran(,:ois Moigno left Paris for Switzerland on
September 6, 1830.
10. From Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 75.
11. From his letters to the Emperor of Austria and to the Czar, published by A.
Terracini, 'Cauchy a Torino', note 53 bis, p. 195, and by G. Castella, 'Documents
inedits pour un Projet de fonder une Academie Helvetique .. .', pp. 312-313.
12. Nouveau Dictionnaire des Girouettes ..., by an immovable weathervane, 1832,
pp. 200-201; cited by Dr. Delaunay, 'Le Parnasse du temps de Napoleon ... ',
p.139.
13. See also Cauchy's letter of October 7, 1830, to the Emperor of Austria: 'Some
members of the Institut and some professors, because of their decision to remain
true to their oaths, are obliged to give up their professorial chairs.'
14. A letter from Lamennais to the Abbe Gerbet, dated October 8, 1830, gives us an
understanding and appreciation of the state of mind of the promoters of the
project:
Herewith are two letters for you from Fribourg. There is nothing in them to say in reply.
They have completely lost their minds; they are every persuaded that within two years
events will lead them to success. They are now busy trying to set up some kind of
academy or university and expect the most beautiful things to result from it.

F. R. de Lamennais Correspondance Generale, 4, (July 1828-June 1831), Paris,

1973, p. 362, letter 1693).
15. See A. Terracini, 'Cauchy a Torino', note 53 bis, pp. 193-199.
274 Notes

16. On March 8,1831, Cauchy's name was written in the Registre des permis de sejour
du canton de Fribourg (see G. Castella, 'Documents inedits pour un projet de fonder
une Academie Helvetique', p. 309).
17. Confirmed by the Proces-verbaux, Ac. Sci., 9, p. 750, Cauchy was present on March
14, 1831.
18. Cauchy's letter to the president of the Academie is kept in the meeting's packet of
July 4, 1831. See Appendix III, p. 325.
19. 'Sui metodi analitici', Biblioteca italiana, 60, pp. 202-219; 61, pp. 321-324; and 62,
pp. 373-386 (D.C., 2, 15, pp. 149-181).
The French manuscript ofthe last two articles sent by Cauchy in Geneva to the
Academie des Sciences in Paris has not been published. This manuscript, entitled
'Sur Ie calcul differentiel et Ie calcul des variations presentant Ie resume de ces deux
calculs', has been preserved in the meeting's packet of July 4, 1831.
20. See A. Terracini, 'Cauchy a Torino', p. 182.
21. Bibliographical studies of the first Turin memoir have been published by A.
Terraccini, 'Cauchy a Torino', pp. 183-185; R. Taton, in D.C. 2, 15, pp. 262-263; I.
Grattan-Guinness, 'On the publication ofthe last volume ofthe works of Augustin
Cauchy', Janus, 62, 1975. pp. 179-191, and J. Peiffer, Les Premiers Exposes
Globaux de la Theorie des Fonctions de Cauchy, 1840-1860, 3rd cycle Thesis,
typewritten, Paris, 1978.
Here is a bibliographical analysis of the first Turin memoir:

1. An introductory resume was read before the Academy of Turin on October 11,
1831 (the handwritten manuscript is in the archives of that Academy, and it was
lithographed several days later and presented to the Academy of Turin on
November 27,1831). This resume was published in France in the Bull. Fer., 15,
May 1831, pp. 260-269 (D.C., 2, 2, pp. 158-168).
2. Extrait du memoire presente aI'Academie de Turin Ie 11 Dctobre 1839,153 pages,
lithographed in Turin in 1832, was presented to the Academy of Turin on
January 27,1833. An Addition of 51 pages, dated March 6,1833, was presented
by Cauchy on April 14, 1833, as the third and final part of the study. An Italian
translation of the Extrait was printed in Milan in the Dpuscoli Matematici e
Fisici,2, 1834. The third paragraph of the first part, the second part, and the
addition of the Extrait are printed in D.C., 2, 15, pp. 262-411. For the first and
second paragraphs of the first part, see below, 3 and 7.
3. In 1837, section 1 ofthe first part and the first lines of section 1 of the second part
ofthe 1832 Extrait appeared in J.L, 2,1837, pp. 406-412 (D.C., 2, 2, pp. 1823).
4. In a letter to Coriolis, dated January 29, 1837, an extract of which appeared in
c.R. Ac. Sci., 4, February 13, 1837, pp. 216-18 (D.C., 1,4, pp. 38-41), Cauchy
gave the statement of the Turin theorem and indicated possible applications. He
also referred to 'a new study in analysis in which I will give a greater study of the
methods developed in the preceding.' It appears that this study was never
published, however.
5. In C.R. Ac. Sci.,9,August 5, 1839,pp. 184-190 (D.C., 1,4, pp. 483-490), Cauchy
gave a proof for the Turin theorem. For the first time, he required that the
derivative of the function should be continuous. This proof was published in Ex.
An. Phy. Math., 1, 1840, pp. 27-32 (D.C., 2, 11, pp. 43-50).
6. In C.R. Acad. Sci., 11, April 20, 1840, pp. 640-650 (D.C., 1,5, pp. 180-191),
Cauchy presented a new proof of the Turin theorem using the mean value of a
function of a complex variable on a circle with center O. This proof was
 Notes 275

published in the Ex. An. Phy. Math., 1,1840, pp. 269-287 (D.C., 2,11, pp. 331-
353).
7. In 1841, the introductory resume of 1831, along with section 2 of the first part of
the Extrait of 1832, with some changes, which were given in a note, appeared in
the Ex. An. Phys. Math., 2, 1841, pp. 41-98 (D.C., 2, 12, pp. 48-112).
22. H. Burckhardt (see 'Entwicklungen nach oscillierenden Funktionen und In-
tegration der Differentialgleichungen der mathematischen Physik', lahresbericht
der Deutschen Mathematiker Vereinigung, 10 (1904-1908, p. 24) and H. Freuden-
thal (see Dictionary of Scientific Biography, 3, p. 140) found in Cauchy's papers 'Sur
la determination du reste de la serie de Lagrange par une integrale definie' and
'Regles de convergence de la serie de Lagrange et d'autres series du meme genre',
Memoires Ac. Sci. 8 (1825),1829, pp. 97-129 (O.c., 1,2, pp. 29-66), a first outline of
the method presented at Turin in 1831. Nevertheless, the calculus of residues was
not used in these studies and the methods of proof were very different.
23. Here is a bibliographical analysis of the second Turin memoir:
1. An introductory resume, read before the Academy of Turin on November 27,
1831, was lithographed in Turin and dated December 17, 1831. The text of the
lithograph was reproduced in France in the Bull. Fer., 16, Sept. 1831, pp. 116-
130 (O.c., 2, 2, pp. 169-183).
2. The complete memoir was lithographed at Turin and dated August 8,1832. It
was presented to the Academie des Sciences in Paris on October 8, 1832 (O.c., 2,
15, pp. 182-261). An Italian translation of this memoir was published in the
Memorie della Societa Italiana delle Scienze in Modena, 1,22, 1838, pp. 91-183.
24. This study, 'Sur un certain type d'equations', which was presented to the Academie
des Sciences in Paris on October 8, 1832, has not been published. A resume of the
work in Italian (no doubt the translation of the presentation made by Cauchy on
September 10, 1832) was published in the Gazzetta Piemontese, no. 113, for
September 22, 1832. This text has been reproduced by A. Terracini, 'Cauchy a
Torino', pp. 178-179. In 1837, in several letters and notes that appeared in the c.R.
Ac. Sci. between May and September (~.C.. 1,4, pp. 42-99), Cauchy came back to
the method of solving equations that he had discussed in the study of September 10,
1832. See c.R. Ac. Sci., 12, June 21,1841, pp. 1133-1145 (O.c., 1,6, pp. 175-186) and
Ex. An. Phys. Math., 2, November 1841, pp. 109-136 (O.c., 2,12, pp. 125-156).
25. The text of this lithograph constitutes section 1 ofthe article 'Calcul des indices des
fonctions', J.E.P., 15, 25th cahier, 1837, pp. 176-226 (O.c., 2, 1, pp. 416-466).
26. A. Terracini, 'Cauchy a Torino', p. 160, and note 3, pp. 170-171.
27. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 75.
28. Cited by A. Terracini, 'Cauchy a Torino', p. 160.
29. B. Boncompagni, 'La vie et les travaux du Baron Cauchy par C. A. Valson ... ',
Bulletino di bibliografia e di storia delle scienze matematiche efisiche, 2,1869, p. 22,
note 10.
30. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 77. Valson probably
paraphrased certain letters that he had examined but are now no longer available.
31. The chronology of events that we propose is, in fact, a hypothesis that we infer from
the following passage of C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1,
p. 77:
His family, in fact, had not been able to even think of his decision to go into exile; but, all
their efforts to dissuade him from doing so bore no fruit. Two short trips that Cauchy
made to Paris and some 'reciprocal' visits by various members of his family were not
276 Notes

enough to compensate for the emptiness caused by his absence. His parents and his
friends were always waiting for his next visit. .. [Valson here tells of the trip made by
Eugene Cauchy]. He seems to have been deeply impressed by all of this and appears to
have been on the point of agreeing to come back when new trouble broke out in Paris
and this caused him to put away any thoughts of returning.
What the troubles were that Valson refers to is not clear, unless the reference is to
the dramatic events of June 1832. As for the two trips that Cauchy made to Paris,
which were undertaken earlier, they undoubtedly were those of March 1831 and of
March 1832, as is confirmed by the Proces-verbaux, Ac. Sci.
32. See Archives Ac. Sci., meeting packet of October 19, 1832.
33. Published by F. Moigno in 1868 (O.C .. , 2, 15, pp. 412-447).
34. A. Terracini, 'Cauchy a Torino', note 33, pp. 180-181.
35. Ibid., p. 165, and note 43, p. 190.
36. L. Menabrea, M emorie, published by the Centro per la storia della tecnica in Italia
del Consiglio nazionale delle ricerche, 1971, pp. 14-15.
37. Proces-verbaux de la classe des sciences physiques et mathematiques de I'Academie
des Sciences, 3, October 11, 1831, cited by A. Terracini, 'Cauchy a Torino', note 35,
p.186.
38. From L. Menabrea, Memorie, p. 15:
In the ultramontain sense, Cauchy was a saint; this, however, does not mean that he did
not have certain 'little faults' or that he was not lacking in charitable spirit. Thus, Plana
enjoyed a position in Turin that, though modest, sufficed for him to devote himself to
science. Aside from the chair in analysis at the university, he was responsible for the
direction of the observatory as well as for the direction of studies at the military
academy where he also lectured. While the total payment that he received was not
much, it is nevertheless true that he was content with it. Cauchy thought this was
excessive and asked that Plana be stripped of one of his positions so that it could be
awarded to him [i.e., to Cauchy]. This behavior raised quite a stir, because Plana was
very popular, and Cauchy, despite the support of the Jesuits, was never able to get one
of Plana's positions awarded to him.

39. From L. Menabrea, Memorie, p. 11:

When any geometer came up with some new transcendental theory, strewn through
with some more or less elegant formulas, he [Plana] would say, by way of judging, 'I
expect numbers to be used'. Frequently, there would be no numbers in the
formulas .... His opposition to what is called the new transcendental analysis, which
Cauchy had promoted, grew in intensity as he himself grew older.

40. 'I will always recall with true delight the scientific conversations that we had while
you were in Turin' wrote Bidone in a letter of thanks to Cauchy on November 4,
1835. Cauchy had sent him his study, 'Sur l'interpolation', along with a letter cited
by A. Terracini, art. cit., note 69, p. 202.
41. See the letter of September 24, 1833, published in Appendix III, pp. 327-328. In
1833, after Cauchy left for Prague, Father La Cheze took charge of the printing of
the editions of the Resumes Analytiques.
42. This letter from Sophie Germain, dated April 18, 1831, was published by C. Henry.
See C. Henry, 'Les manuscrits de Sophie Germain et leur recent editeur.
Documents nouveaux', Revue Philosophique de fa France et de I'Etranger, 8, 1879
pp. 619-641, esp. p. 632.
 Notes 277

Notes to Chapter 10
1. F~om a copy written by Cauchy, undated, kept in the Musee National de
I'Education in Rouen, Ms A 10822-1.
2. See the letter by Cauchy in Appendix III, p. 326.
3. Barande remained in Bohemia after his dismissal, where he studied geology,
gaining a reputation and name for himself as a geologist.
4. For more on this affair, see Comte de Damas d'Anl{:zy, 'L'Education du duc de
Bordeaux', Revue des Deux Mondes, October 1902, pp.602-640, in particular,
pp. 609-611. In this article, the author who had access to private papers, stated that
the Marquis de Foresta, 'always tireless in serving his prince', called on Cauchy in
Turin to ask him to serve as Barande's replacement (pp. 619-620). This assertion
does not agree with the letters from Baron de Damas that Cauchy received in
Turin.
5. See Appendix III, pp. 327-328.
6. Brochure published in Prague in September 1833. This declaration appeared in
Fribourg, in LInvariable, 4, pp. 64-78, with the following foreword:
We are all the more happy to publicize such noble thoughts because the so-called
royalist newspapers have raged with insolent violence and perfidious hypocracy
against the new teachers of the Duke of Bordeaux and have carefully omitted M.
Cauchy. That European name was an obstacle to their declamation; and, in order to
lament over the departure of M. Barande and present it as irreparable, it was necessary
not to tell how it has, in fact, been repaired.
7. A. d'Hautpoul, Quatre Mois a la Cour de Prague, published by Count de Fleury,
Paris, 1902, p. 129, and p. 147.
8. Ibid., p. 160.
9. Ibid., p. 242:
Of all the people around me, the one who most obstructed my principle of education
was Cauchy.
10. Ibid., p. 327.
11. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 90.
12. Ibid., 1, p. 90. It is difficult to identify the paper to which AloYse referred. Perhaps it
is 'Sur l'interpolation', which was lithographed in September 1835, or 'Memoire sur
l'integration des equations differentielles', which was lithographed at the end ofthe
same year.
a
13. A. d'Hautpoul, Quatre Mois la Cour de Prague, p. 160.
14. F. Moigno, Le{:ons de Calcul Differentiel et de Calcul Integral, 1,1840, Introduction,
p.XIV:
M. Cauchy has developed, on a totally new basis, an elementary treatment of arithmetic
and geometry. The reason for his having done so is, of course, known; and, it is a
pleasure to see a great thinker, inspired with self-sacrifice, suspend the pursuit of his
own brilliant discoveries and research in order to make the most important secrets of
science available to a young royal person who is in exile.

Unfortunately, the archives of the Duke of Bordeaux have been entirely destroyed.
Two of Cauchy's notebooks relating to this elementary level teaching are kept in
the Sorbonne Library: a notebook of arithmetic exercices (Ms 1760) and a
notebook containing Section 1 of Chapter 7, 'Exponentielle et logarithme', from an
278 Notes

elementary treatise referred to by Moigno. That section is devoted to powers using

arbitrary exponents or exponentials (11 pages, unfinished, Ms. 1761).
15. Method published in c.R. Ac. Sci., 11, November, 16 and 23,1840, pp. 789~ 798 and
(O.c., 1,5, pp. 431~454). Cauchy described its origins in the following terms:
Because of the honorable mission that was conferred upon me at the time that I was
publishing these tables [that is, in 1836], having presented me with an opportunity to
investigate whether it is possible to make the various rules of computing more easy and
more certain, I came to realize that some very simple procedures might be of advantage
in this regard, even in arithmetic.
16. Those assertions by General d'Hautpoul that we have been able to verify all appear
to be correct; and this compels us, regardless of d'Hautpoul's resentment of
Cauchy, to take account of his testimony.
17. A. d'Hautpoul, Quatre Mois a la Cour de Prague, pp. 246~292.
18. Ibid., p. 291.
19. Ibid., p. 292.
20. Ibid., pp. 245~246.
21. Ibid., p. 244.
22. Ibid., p. 243.
23. P. L. F. de Villeneuve, Charles X et Louis XIX en Exit. Memoires Inedites, Paris,
1889, pp. 242~243.
We also note the imprecise testimony dated February 24,1826, in the Invariable
(Fribourg), 8, 1836, p. 196:
We then moved on to the examination in the physical sciences. The prince responded
correctly to the questions that were addressed to him, on applications to astronomy
and navigation. His very learned teacher was M. Cauchy, to whom the sciences were
indebted for his analytical treatment of the theory of light, a work that had the
unexpected result of leading this famous geometer to create some new interpolation
formulas, which have at once made an immense step in analysis as a whole and in the
theory of light.
24. A. Terraccini, 'Cauchy a Torino', note 33, pp. 180~181.
25. W. R. Hamilton, 'On a general method in dynamics', Mathematical Papers, 2,
Cambridge (England), 1940, p. 103.
26. Cauchy maintained that he had been thinking about this reduction for a long time.
In 1835, he wrote:

The fine study by M. Hamilton, in which he made the integration of the differential
equations of dynamics depend on the determination of a single function, represented by
a definite integral that satisfies two second-order partial differential equations, has
brought my ideas to a point that occurred to me a long time ago and is worthy of being
examined with particular care. I thought that there perhaps would be some advantage
in reducing the integration of a system of differential equations to the problem of
integrating a single partial difference equation of the first order.

(Ex. An. Phys. Math., 1, December 1840, p. 331, O.c., 2, 11, p. 404). Five years latter
he wrote:
The method of reduction that I have just applied to a system of differential equations
does not differ from the one that I gave in the 1835 study; and it is one that I have been
considering for a long time. In fact, I have just found it again in a note dated August 31,
1824, placed behind various studies that were presented to the Academie during 1823.
 Notes 279

(C.R. Ac. Sci., 10, June 29, 1840, p. 957, O.C., 1,5, p. 236). The note could not be
found. Cauchy probably gave this method of reduction in his analysis teaching at
the Ecole Poly technique.

27. The first installment of Nouveaux Exercices, pp. 1-24 (O.c., 2,10, pp. 195-220),
goes back to the 'Memoire sur la dispersion de la lumiere' of 1830, As to financing
the publication, see the advice to the reader, dated June 10, 1836.
28. Lithograph dated September 26, 1835, published in 1837 in J. L., 2, pp. 193-205
(O.c., 2, 2, pp. 5-17). On this study, see C. C. Heyde and E. Seneta, I. J. Bienayme.
Statistical theory anticipated, Springer-Verlag, New York, Heidelberg, Berlin,
1977, pp. 71-76.
29. This study, which is not widely known, is not contained in the Oeuvres Completes
(O.c.). A copy of the lithograph is preserved in the Houghton Library at Harvard
(FCB. 02103. B843d).
30. From the 'observation' published at the end of his lithographed study Sur la
Theorie de la Lumiere (August 1836). Cauchy brought several notebooks back to
France with him; these notebooks contained manuscripts devoted to his research
on light; some of these are dated 1836. An entire notebook, entitled 'Researches
nouvelles sur la lumiere' and dated January 1838, is kept in the Sorbonne Library
(Ms. 1762).
According to c.R. Ac. Sci., 15, September 26, 1842, p. 606 (O.C., 1, 7, p. 157),
another notebook, which is now lost, contained two unpublished investigations:

1. a first study, 'Sur la theorie de la lumiere', in 4 sections: the first three dealt with the
theory of spherical and cylindrical waves (see c.R. Ac. Sci., 9, November 18, 1839,
pp. 637-649 (O.c., 1,5, pp. 5-20); the 4th section presented a discussion of the two
phenomena of shadows and diffraction (C.R. Ac. Sci., 7, September 26, 1842,
pp.157-170).
2. the manuscript of the lithographed study Sur la theorie de la lumiere.

31. These letters have been published almost completely in the Compte Rendus
Hebdomadaires des Seances de I'Academie des Sciences. Handwritten copies
containing some modifications compared to the printed versions are contained in
the manuscript book Melanges, which is kept in Ivoy-Le-Pre at the residence of
Madame de Pomyers. The dates on which these letters were written are given in
that manuscript:

1. February 12, 1836: letter to Ampere, published in the c.R. Ac. Sci., 2, February 22,
1836, pp. 182-185 (O.c., 1,4, pp. 5-8);
2. February 19, 1836; letter to Ampere, published in the c.R. Ac. Sci., 2, February 29,
1836, pp. 207-209 (O.c., 1,4, pp. 9-11);
3. March 19, 1836; letter to Libri, published in the c.R. Ac. Sci., 2, April 4, 1836,
pp. 341-343 (O.c., 1,4, pp. 11-13);
4. March 28, 1836; letter to Libri, published in the c.R. Ac. Sci., 2, April 4, 1836,
pp. 343-349 (O.C., 1,4, pp. 13-21);
5. April 1, 1836: letter to Ampere, published in the C.R. Ac. Sci., 2, April 11, 1836,
pp. 364-371 (O.c., 1,4, pp. 21-27);
6. April 16, 1836: letter to Ampere, published as addressed to Libri, in the C.R. Ac. Sci.,
2, May 2. 1836, pp. 424-428 (O.c., 1,4, pp. 30-32);
7. April 22, 1836: letter to Libri, published in he c.R. Ac. Sci., 2, May 9,1836, pp. 455-
456 (O.C., 1,4, pp. 32-34):
280 Notes

8. April 26, 1836: letter to Libri, published in the C.R. Ac. Sci., 2, May 9, 1836, pp. 456-
461 (O.c., 1,4, pp. 34-36).

32. The handwritten manuscript book Melanges, which is kept in Ivoy-Ie-pre at the
residence of Madame de Pomyers, contains 'Recherches sur la theorie de la
lumiere' (unpublished), with 63 pages, signed on November 4, 1837 by Augustin
Cauchy. These 'Recherches' are divided into 5 sections:

1. polarisation rectiligne (sent to Moigno on December 15, 1837);

2. axes optiques dans la polarisation rectiligne;
3. surface des ondes;
4. ombre et diffraction; and
5. suite du 4 erne §.

Aside from this, one can also find section 9 of'Theorie de la lumiere', entitled 'Lois
de propagation de la lumiere dans Ie vide et dans les milieux qui ne dispersent pas
les couleurs' (4 p.), and 'Notes sur la theorie de la lumiere envoyees a M. L'abbe
Moigno' (27 p.):

Note 1 'Equations du mouvement de l'ether dans un milieu Oil la propagation de la

lumiere s'effectue de la meme maniere en tous sens autour de l'axe' (sent to Moigno
on October 6, 1837).
Note 2 'Sur les vibrations de l'ether dans un milieu transparent' (sent to Moigno on
October 6, 1837).
Note 3 'Ellipsoi'de'.

33. See p. 285, note 20.

34. The manuscript of the 1835 study, 'Sur !'integration des equations differentielles's,
is kept at the Academy of Sciences of Czechoslavakia. See K. Rychlik, 'Un
manuscrit de Cauchy aux Archives de I'Academie tchecoslovaque des Sciences',
Revue d'Histoire des Sciences, 10, 1958, pp. 259-261.
35. On relations between Bolzano and Cauchy in Prague, see R. and D. J. Struik,
'Cauchy and Bolzano in Prague', Isis, 11,1928, pp. 364-366, and K. Rychlik, 'Sur
les contacts personnels de Cauchy et Bolzano', Revue d'Histoire des Sciences, 15,
1962, pp. 163-164 and H. Sinaceur, 'Cauchy et Bolzano', Revue d'Histoire de
Sciences, 20, 1973, pp. 97-112.
36. See C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 91, who gives as the
places where Cauchy resided Hradshin, Toeplitz in 1835 (probably during the
summer), Budweis, Kirchberg, Goeritz in 1836, and Kirchberg in 1838. See also L.
Bader, Les Bourbons de France en Exil a Gorizia, Paris, Perrin 1977, p. 100, and
appendix 10, Etat du personnel des maisons etde la suite de lafamille royale aGoeritz
en Novembre 1837 pp. 372-374.
37. Two notes, one from September 18 and the other from October 23, 1837, have not
been published in the C.R. Ac. Sci. The unpublished manuscript is kept in the
meeting's packet of September 18, 1837. In it, Cauchy suggested that he was
working on a study of Ampere's interpolation function. As to these investigations,
see also the three manuscripts by Cauchy that are kept at the Sorbonne Library,
Mss. 1759, 1760, and 1786.
38. From C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 94.
 Notes 281

Notes to Chapter 11
1. See G. Bigourdan, 'Le Bureau des Longitudes, son histoire et ses travaux de
l'origine Ii ce jour', 3rd part, Annuaire du Bureau des Longitudes, 1930, A 18-26.
2. Published in the c.R. Ac. Sci., 9, August 5, 1839, pp. 184-190 (D.C., 1,4, pp. 483-
490). Repeated in the Ex. An. Phys. Math., 1, September 1839, p. 27 (D.C., 2, 2,
p.43).
3. c.R. Ac. Sci., 9, August 5, 1839, p. 190 (D.C., 1, 4, p. 490).
4. Proces-verbaux des seances du Bureau des Longitudes, November 6,1839, cited by
G. Bigourdan, 'Le Bureau des Longitudes, .. .', A 29.
5. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 97-98. The problem at
the Bureau des Longitudes is treated extensively in this work (pp. 97-104).
6. See C.R. Ac. Sci., 8, April 15, 1839, pp. 553-661 [Cauchy] (D.C., 1,4, pp. 312-321);
April 22, 1839, pp. 581-582 [Poisson]; and pp.582-589 [Cauchy], (D.C., 1, 4,
pp.322-330).
7. Proces-verbaux des seances du Bureau des Longitudes, November 13, 1839, cited by
G. Bigourdan, 'Le Bureau des Longitudes, .. .', A 29-30.
8. Proces-verbaux des seances du Bureau des Longitudes, November 27,1839, cited by
G. Bigourdan, 'Le Bureau des Longitudes, .. .', A 30.
9. From the Proces-verbaux des seances du Bureau des Longitudes. The discussion of
this problem is handled very discretely. Arago seems to have played a very impor-
tant role in these developments. G. Bigourdan,'Le Bureau des Longitudes, .. .',
A 30.
10. J. B. Biot, 'Lettre a M. de Falloux', abstract from Le Correspondant, 1857, p. 9. See
also 1. Bertrand, Eloges Academiques, 2, Paris, 1902, pp. 116-117:
It is in jest told that, being pressed to accept an unimportant formality, he replied: 'May
they chop off my head!' That was his way of saying 'No!' most emphatically.
11. A remark by J. B. Biot reported in the Proces-verbaux des seances du Bureau des
Longitudes; cited by G. Bigourdan, 'Le Bureau des Longitudes, .. .', A 30.
12. In a study on celestial mechanics that was presented to the Academie on August 3,
1840 (C.R. Ac. Sci., 11, August 3, 1840, pp. 179-185 (D.C., 1,5, pp. 260-276), he
expressed regret at not being able to present his works 'to the gathering of scholars
who had done so much to advance celestial mechanics' and 'to offer to have them
[his works] placed in the Connaissance des Temps'. He concluded his study with
these words:
The only thing I can do is exert every possible effort to respond to the kindness that the
friends of science have shown toward my works. Moreover, if it is possible, I hope to be
able to prove that the title of geometer is not altogether at odds with the ususal
concerns of the old professor, who, in times past, was kindly consulted by the masters of
science (celestial mechanics).
(See also the foreword of the first volume of the Exercices d'Analyse et de Physique
Mathematique, probably written in the spring of 1841.)

13. F. Hoefer, Nouvelle Biographie Universelle, Paris, 1852-1866,9, article on Cauchy.

14. 'Discussion des lignes et des surfaces du second degre', Ex. math., 3, June 1828,
pp. 65-127 (D.C., 2, 8, pp. 83-149):
This manner of arriving at the equation ofthe tangent plane [Cauchy wrote in a note]
has been pointed out to us by a young ecclesiastic, who is as learned in the divine
282 Notes

sciences as he is in the human sciences. He is a member of that illustrious society that

has rendered great services to civilization in both hemispheres.
15. In the preface to his Le(:ons de Mecanique Analytique, which was published in
1867, Moigno pointed out that he had used the lectures that Cauchy had given
prior to 1830 in his courses at the Faculte des Sciences as the basic text for his own
teaching at the Ecole Normale Ecclesiastique from 1838 to 1843.
16. F. N. M. Moigno, Le(:ons de Calcul DifJerentiel et de Calcul Integral, 1, 1840,
introduction, p. XIII.
17. Ibid., 1, introduction, pp. XVIII-XIX and Lessons 42 and 45.
18. c.R. Ac. Sci., 11, November 2, 1840, pp. 687-693 (O.c., 1, 5, pp. 424-430).
19. See 1. B. Duroselle, Les Debuts du Catholicisme Social en France, 1822-1870, Paris,
PUF, 1951, pp. 248-249, and Archives ofthe Compagnie de Jesus, Bibliotheque du
Centre des Fontaines, Chantilly, Moigno file.
20. F. N. M. Moigno, Le(:ons de Calcul Differentiel et de Calcul Integral, 2, 1844,
introduction.
21. C. R. Ac. Sci., 23, November 16, 1846, pp. 911-914 (O.C., 1,10, pp. 202-205).
22. See C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 201-213.
23. Many authors have wrongly confused the two conference centers (L. Baunard,
Frederic Ozanam d'apres sa Correspondance, Paris, 1912, p. 345, J. B. Duroselle, op.
cit., pp. 247-248). Valson also makes this error. Jeanne Caron, however, correctly
distinguishes between the Institut Catholique and the Cercle Catholique in the
critical edition of the Lettres de Frederic Ozanam, 2 (1841-1844), Paris, CELSE,
1971, p. 295, note (2). The Cercle Catholique met in the rue de Grenelle and
published its own bulletin.
24. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 205.
25. Notice sur l'Institut Catholique, February 1842, published with the Bulletin de
I'Institut Catholique containing the 'Seance d'ouverture de l'annee 1842'.
26. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 205 and p. 206. See also
Bulletin de I'Institut Catholique, p. 42, meeting of March 3, 1842.
27. In this Bulletin, the lectures given at the opening meeting and the general meetings
were published. Collection of the Bulletin, that is kept at the Bibliotheque
Nationale, concludes with the meeting of March 13, 1844.
28. See Notice sur l'Institut catholique, February 1842.
29. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 205-206.
30. Ibid., 1, p. 199.
31. See G. Bertier de Sauvigny, Le Comte Ferdinand de Bertier, 1782-1864, et I'Enigme
de la Congregation, Paris, 1949, pp. 536-599, and Lettres de F. Ozanam, 2, p. 74,
note 6.
32. Letter from Dominique Meynis, March 8,1841, in Lettres de F. Ozanam, 2, p. 96.
33. Letter to Dominique Meynis, March 17, 1841, ibid., p. 96.
34. Letter to Domique Meynis, March 23, 1841, ibid., p. 119.
35. Letter from Cauchy to Father Roothan, A. R. S. J., Franc. 1007-XXXIV,
8... Rome. Reactions that it provoked in Rome, A. R. S. 1., Reg. Provo Franciae, II,
239.
36. G. Monod, 'Les troubles du College de France en 1843', Seances et Travaux de
I'Academie des Sciences Morales et Politiques, 1909,2, pp. 407-423.
37. Cauchy had known Libri since the 1820s. In his letter of June 19 (See Appendix IV,
p. 343), he noted the importance that was attached to his candidacy by the other
candidates-Liouville and Libri-who had attended his lectures in the past.
 Notes 283

38. Letter from Cauchy to Letronne, dated June 11, 1843, Archives of the College de
France, B-II Mathematiques, b-1. See Appendix IV, p. 339.
39. Michelet wrote in his diary for May 6,1843; 'A visit from Libri: rubbiano. Shocking
disclosures!'
40. G. Libri, 'Lettres sur Ie clerge': 'I. De la liberte de conscience', Revue des Deux
Mondes, 1843, 3, pp. 329-356; II. 'Y-a-t-il encore des jesuitesT, Revue des Deux
Mondes, 1843,3, pp. 968-981.
41. Archives of the College de France, Libri file.
42. Archives ofthe College de France, B-II, Mathematiques, b-1, letter from Cauchy to
Letronne, undated. See Appendix IV, p. 340.
43. Archives of the College de France, CXII, Liouville, 1B, letter from Liouville to
Letronne, dated June 19, 1843. See Appendix IV, p. 341.
44. Letter from Cauchy to the members of the Academie des Sciences, June 19, 1843,
Archives Ac. Sci., Libri file, rough draft in the Archives of the Compagnie des Jesus,
Bibliotheque du Centre des Fontaines, Chantilly, Moigno file. See Appendix IV,
p.343.
45. Archives Ac. Sci., Libri file. See Appendix IV, p. 345. Relations between Libri on
the one hand and Sturm and Liouville on the other hand had been very bad since
about 1838-1839.
46. This declaration was published in the c.R. Ac. Sci., 16, p. 1365.
47. In the end, Liouville only published the paper by Galois in his journal in 1846. He
declined to publish the commentaries that were proposed.
48. Proces-verbaux des seances du Bureau des Longitudes, November 15, 1843, cited by
G. Bigourdan, 'Le Bureau des Longitudes', A 31.
49. Proces-verbaux des seances du Bureau des Longitudes, November 29, 1843, cited
by G. Bigourdan, 'Le Bureau des Longitudes', A 31.
50. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 101-104.
51. For more on the relationship between Father Fran~ois de Ravignan and Cauchy,
see A. de Ponlevoy, Vie du R. P. Xavier de Ravignan 2, Paris, 1876, pp. 386-387,
and C. A. Valson, op. cit., 2, pp. 111-112.
52. Ibid. 2, pp. 189-190.
53. E. Gossin, Vie de M. Jules Gossin. 1789-1855, Paris, 1907, pp. 325-328, and c.R.
Ac. Sci., 12, May 11, 1846, p. 751.
54. E. Gossin, Vie de M. Jules Gossin, p. 326, and C. A. Valson, La Vie et les Travaux
du Baron Cauchy, 2, pp. 188-189.
55. H. de Riancey, article in the Union, January 16, 1857, cited by C. A. Valson, La Vie
et les Travaux du Baron Cauchy, 2, pp. 190-191.
56. Ibid., 2, pp. 191-193.

Notes to Chapter 12
1. Le National, October 19, 1842.
2. See the letter from Cauchy to Moigno dated October 26, 1842, Appendix III,
p.331.
3. J. B. Biot, 'Comptes rendus hebdomadaires des seances de I'Academie des sciences,
publies par MM. les secretaires perpetuels, commen~ant au 3 aout 1836', Journal
des Savants, November 1842, pp. 641-661. See especially pp. 659-660.
4. C.R. Ac. Sci., 15, November 14, 1842, pp. 910-916 (O.C., 1, 7, pp. 200-207).
284 Notes

5. c.R. Ac. Sci., 15, December 19,1842, p. 1075 (D.C., 1,7, p. 212). The scientific writer
for the National, always ridiculing Cauchy, had the following to say in the
December 21, 1842 issue:
On reading in the Journal des Savants, his criticism on publications abuses, which was
addressed to an illustrious member ofthe Academie, M. Cauchy exclaimed: 'Well, truly,
if the article was not signed, then one would swear that the author intended to speak of
M. Biot'. Is the remark authentic? That could not be ascertained. However, it might
well be.

6. From December 1842 to December 1843, Cauchy published 43 articles in the

Comptes Rendus Hebdomadaires des Seances de I'Academie des Sciences (which in
the D.C. amounts to 370 pages), that is, about the same number of articles as from
December 1841 to December 1842 (or about 310 pages in the D.C.).
7. J. Bertrand, 'La Vie et les travaux du Baron Cauchy, par C. A. Valson', Journal des
Savants, 1869, p. 210.
8. The following sarcastic remark about Cauchy is due to Poinsot and is cited by
Barre de Saint-Venant in a letter to Boussinesq on July 9, 1876 (Bibliotheque de
l'Institut, Ms. 4227):
Poinsot said that the majority of Cauchy's works were like a notebook with blank
pages.
9. From June 1839 to February 1848, Cauchy and Liouville worked together on 51
evaluation committees; similarly, from October 1838 to February 1848, he worked
on 31 committees with Sturm; from October 1838 to February 1848 he served on 22
committees with Poncelet; from July 1843 until February 1848, he worked with
Binet on 30 committees; and from March 1843 to February 1848, he worked on 21
committees with Lame.
10. See R. Taton, 'Liouville, Joseph', Dictionary of Scientific Biography, 8, New York.
1973, pp. 381-387, and P. Speziali, 'Sturm, Charles Fran<;:ois', ibid., 13, 1976,
pp. 126-132.
11. In a letter written in 1840, Liouville wrote to Dirichlet concerning two notes by
Cauchy (C.R. Ac. Sci., 10, March 16, 1840,pp. 437-452, and May 11, 1840,pp. 719-
731 (D.C., 1,5, pp. 135-152 and pp. 199-212)) on a question that had been studied
by Dirichlet in his papcr 'Sur l'usage des series infinies dans la tMorie des nombres'
and that Liouville presented in a course at the College de France.
You can imagine my astonishment at seeing that I had been cited by M. Cauchy in the
two notes that he ventured to publish with reference to yours. This especially concerns
the one of May 11. I see scarcely anything in these notes that is not like an amplification
of the teacher's paper by a good student. In the second note, the plagiarism is
particularly evident. I clearly explained to him that he had forced me to playa dirty role
by citing me thusly without my knowledge and [I also spoke of] the injustice that he
had committed with regard to you. I must tell you that during the conversation he
appeared to be much more fair-minded that he had been on paper. He promised me
that he would not mention me and would be more mindful of your rights in the second
edition, which I think will appear in the memoirs of the Academie. In any event, we will
see what he will do. If he should fail to acknowledge you, then I strongly advise you to
take up the matter yourself when you publish a new work.
12. In January 1836, Liouville founded the Journal de Mathematiques Pures et
Appliquees. Practically all of the leading mathematicians of the day-especially
young authors-published in it. Liouville was closely connected with many
 Notes 285

mathematicians and corresponded regularly with them, thus providing a link

between the French and German scientific communities.
13. The handwritten report by Cauchy accompanying a letter to the president of the
Academie is kept in the Bertrand collection at the Academie des Sciences.
14. Bulletin de l'Institut Catholique, 1st issue, 'Seance d'ouverture', January 13, 1842, p. 13.
15. Letter from Cauchy to J. B. Dumas, June 20, 1843, Archives Ac. Sci., Libri file.
16. Following are some figures: from 1 December 1843 to 1 December 1844, Cauchy
presented to the Academie 19 communications and 4 reports (or 186 pages in the
O.c.) from December 1, 1842 to December 1, 1843, and 44 communications and 4
reports (or 536 pages in the O.C.), as opposed to 39 communications and 5 reports
(or 362 pages in the O.c.) from December 1, 1844 to December 1, 1845. On the
other hand, from December 1, 1843 to December 1, 1844, he served on 11 academic
commissions, 5 of which took place during November 1844, as opposed to 20
academic commissions from December 1, 1842 to December 1, 1843, and on 19
academic commissions from December 1, 1844, to December 1, 1845.
17. A. L. Cauchy, Recueil de Memoires sur la Physique Mathematique, Paris, Bachelier,
1839.
18. In the first volume ofthe new series of Exercices, 11 of 15 papers were devoted to
questions relative to mathematical physics and, more particularly, to the theory of
light.
19. See c.R. Ac. Sci., 7, October 29, November 19 and 26, and December 10, 1838,
pp. 751-759, pp. 865-867, pp. 907-912, and pp. 985-992; C.R. Ac. Sci., 8, January
7,14, and 28, and February 4,11,18, and 25,1839, pp. 7-13, pp. 39-46, pp. 114-
119, pp. 146-155, pp. 189-196, pp. 229-231 and pp. 985-1000 (O.c., 1,4, pp. 99-
186 and pp. 427-443); c.R. Ac. Sci. April 8, 22, and 29, May 20, and June 24,1839,
pp. 505-522, pp. 589-597, pp. 659-673, and pp. 767-778 (O.C., 1,4, pp. 237-312);
C. R. Ac. Sci., 9, July 1,8, and 15, 1839, pp. 1-9, pp. 59-68, and pp. 91-103 (O.c., 1,
4, pp. 427-483); and the three papers, 'Memoire sur les mouvements infiniment
petits d'un systeme de molecules sollicitees par les forces d'attraction et de
repulsion mutuelle', Ex. An. Phys. Math., 1, September 1839, pp. 1-15 (O.C., 2,11,
pp. 11-27); 'Memoire sur les mouvements infiniment petits dont les equations
presentent une forme independante de la direction des trois axes coordonnees,
supposes rectangulaires, ou seulement de deux de ces axes', ibid., September 1839,
pp. 101-132 (O.C., 2, 11, pp. 134-174); and 'Memoire surla polarisation rectiligne
et la double refraction', May 20, 1839, Mem. Ac. Sci., 18 (1840), 1842, pp. 153-216
(O.C., 1,2, pp. 111-166).
20. See C.R. Ac. Sci., 8, April 22, and May 20, and 27,1839, pp. 597-598, pp. 719-731,
and pp. 779-783 (O.C., 1,4, pp. 343-369); C.R. Ac. Sci., 9, November 11, 1839,
pp. 589-590 (O.c., 1,4, pp. 520-523); and 'Memoire sur les mouvement infiniment
petits de deux systemes de molecules qui se penetrent mutuellement', Ex. An. Phys.
Math., 1, September 1839, pp. 33-52 (O.c., 2, 11, pp. 51-74).
21. See, for example, C.R. Ac. Sci., 9, August 26,1839, pp. 283-288 (O.C., 1,4, pp. 491-
496).
22. c.R. Ac. Sci., 9, November 25, and December 2,1839, pp. 676-691 and pp. 726-730
(O.c., 1,5, pp. 20-42), and 'Memoire sur la reflexion et la refraction d'un mouvement
simple transmis d'un systeme de molecules a un autre, chacun des deux systemes
etant suppose homogene et tellement constitue que la propagation des mouvements
infiniment petits s'y effectue en tous sens suivant les memes lois', Ex. An. Phys. Math.,
1, December 1839, pp. 133-177 (O.c., 2, 11, pp. 173-226).
286 Notes

23. An exception, the 'Memoire sur la theorie de la polarisation chromatique', c.R. Ac.
Sci., 18, May 27,1844, pp. 961-972 (D.C., 1,8, pp. 213-224).
24. 'Memoire sur l'integration des equations lineaires', C.R. Ac. Sci., 8, May 27, and
June 3,10, and 17, 1839, pp. 827-830, pp. 845-865, pp. 889-907, and pp. 931-937
(D.C., 1,4, pp. 369-427), and Ex. An. Phys. Math., 1, September 1839, pp. 53-100
(D.C., 2, 11, pp. 75-133).
25. 'Memoire sur la reduction des integrales generales d'un systeme d'equations
lineaires aux differences partielles', c.R. Ac. Sci., 9, August 26, and November 18,
1839, p. 288 and pp. 637-649 (D.C., 1,4, p. 497, and 5, pp. 5-19), and Ex. An. Phys.
Math., 1, December 1839 and April 1840, pp. 178-211 (D.C., 2, 11, pp. 227-264).
26. P. H. Blanchet, 'Memoires sur la propagation et la polarisation du mouvement
dans un milieu homogene indHini, cristallise d'une maniere quelconque', J.L., 5,
1840, pp. 1-30.
27. Communications each week from July 5 until September 13, 1841 (D.C., 1, 6,
pp.202-340); numerous other communications until March 1842 (D.C., 1, 6,
pp. 367-421).
28. c.R. Ac. Sci., 23, August 9, 1841, pp. 339-340.
29. c.R. Ac. Sci., 23, November 15, 1841, pp. 958-960.
30. c.R. Ac. Sci., 23, December 20, 1841, p. 1152.
31. c.R. Ac. Sci., 10, April 20, 1840, pp. 640-650 (D.C., 1,5, pp. 180-191), and Ex. An.
Phys. Math., 1, August 1840, pp. 269-287 (D.C., 2, 11, pp. 331-353).
32. C.R. Ac. Sci., 12, December 16, 1844, p. 339 (D.C., 1, 8, pp. 336).
33. C.R. Ac. Sci., 9, August 5, and November 11, 1839, pp. 184-190 and 587-588 (D.C.,
1,4, pp. 483-491 and pp. 518-519); c.R. Ac. Sci., 10, April 20, 1840, pp. 650-656
(D.C., 1,5, pp. 191-198); and Ex. An. Phys. Math., 1, August 17, 1840, pp. 279-187
(D.C., 1, 11, pp. 343-353).
34. C.R. Ac. Sci., 15, July 11 and 25,1842, pp. 44-59 and pp. 85-102 (D.C., 1,7, pp. 17-
49 and pp. 62-67).
35. c.R. Ac. Sci., 17, October 30, 1843, pp. 938-942 (D.C., 1,8, pp. 115-119).
36. J.L., 1, 1836, pp. 197-210, and c.R. Ac. Sci., 11, September 14 and 21, 1840,
pp. 453-475 and pp. 501-511 (D.C., 1,5, pp. 288-311 and pp. 311-331).
37. c.R. Ac. Sci., 13, August 9,1841, p. 317 (D.C., 1,6, p. 282). See also c.R. Ac. Sci., 13,
October 4 and 24,1841, pp. 682-687 and pp. 850-854 (D.C., 1,6, pp. 341-346 and
pp. 354-358). A letter from Leverrier to Cauchy about Lexell's comet (C.R. Ac. Sci.,
18, April 29, 1844, pp. 826-827) shows that relations between the two men
remained close during the following years.
38. C.R. Ac. Sci., 15, August 8, August 16, August 22, August 29, and September 5,
1842,pp. 255-269,pp. 303-305,pp. 357-366,pp. 411-418,andpp. 487-483 (D.C.,
1, 7, pp. 86-126).
39. c.R. Ac. Sci., 20, March 17, March 24, March 31, April 7, and April 21, 1845,
pp.767-796, pp.825-847, pp.907-927, pp.996-999, pp.1166-1180 and
pp. 1612-1626 (D.C., 1, 9, pp. 121-129). See J. B. Biot, Melanges scientifiques et
litteraires, 3, Paris, 1858, p. 153, and C. A. Valson, La Vie et les Travaux du Baron
Cauchy, 2, pp. 167-169. Cauchy was criticized for having published studies on
planetary perturbations, because this was the subject of a competition of the
Academie. See C.R. Ac. Sci., 20, April 7, 1845,pp. 996-999 (D.C., 1,9, pp. 186-190).
40. C.R. Ac. Sci., 21, September 15, 22, and 29, October 6, 13,20, and 27, November 3,
10,17, and 24, and December 1, 8, 15, 22, and 29,1845, pp. 593-607, pp. 668-679,
pp. 727-742,pp. 779-797,pp. 835-852,pp. 895-902,pp.931-933,pp. 1025-1041,
 Notes 287

pp.1042-1044, pp.1093-1101, pp.1123-1134, pp.1188-1201, pp.1238-1255,

pp. 1287-1300, pp. 1356-1369, and pp. 1401-1409; and C.R. Ac. Sci., 22, January
5,12,19, and 26, February 2 and 9, and April 11, 1846, pp. 2-31, pp. 53-63, pp. 99-
107, pp. 159-160, pp. 193-196, pp. 235-238, and pp. 630-632 (D.C., 1,9, pp. 277-
505, and 10, pp. 5-68).
41. Bertrand's postulates states that for any integer n> 1, there exists at least one
prime between nand 2n.
42. Ex. An. Phys. Math., 3, p. 250 (D.C., 2, 13, p. 280).
43. These papers were finally published by Liouville in the November 1846 edition of
the J.L., 11, pp. 381-444.
44. C.R. Ac. Sci., 21, December 8,1845, and January 5,1846,21, pp. 1247-1248, and 22,
pp. 30-31 (D.C., 1, 9, p. 459, and 10, p. 34).
45. See A. Dahan, 'Les travaux de Cauchy sur les substitutions. Etude de son approche
du concept de groupe', Arch. Hist. Ex. Sci., 23, 1980, pp. 296-310.
46. A. Lamarle, 'Note sur Ie theoreme de M. Cauchy relatif au developpement des
fonctions en series', J.L., 11, April 1846, pp. 129-141.
47. J.L., 11, August 11, 1846, pp. 313-330 (D.C., 2, 2, pp. 35-54). Lamarle responded to
Cauchy in 1847 in J.L., 12, August 1847, pp. 305-342.
48. J.L., 11, August 1846, p. 323 (D.C., 2, 2, p. 46).
49. c.R. Ac Sci., 23, August 8,1846, pp. 251-255 (D.C., 1,10, pp. 70-75). Following
Saint-Venant's paper, 'Sur les sommes et les differences geometriques et sur leur
usage pour simplifier la mecanique', for which he had been appointed commis-
sioner, Cauchy had defined the senses, direct or backward, of a rotation. See c.R.
Ac. Sci., 23, August 3, 1846, p. 251 (D.C., 1,8, pp. 69-71), and 'Memoire sur les
resultantes que I'on peut former, soit avec les cosinus des angles, soit avec les
coordonnees de deux ou trois points', Ex. An. Phys. Math., 4, pp. 5-86 (D.C., 1,14,
pp.I-92).
50. C.R. Ac. Sci., 23, September 21, 1846, pp. 557-563 (D.C., 1,10, pp. 135-143).
51. C.R. Ac. Sci., 17, October 2, 9, 16, 23, and 30, 1843, pp. 572-581, pp.640-651,
pp. 693-702, pp. 921-925, and pp. 1159-1164 (D.C., 1,8, pp. 65-115).
52. C.R. Ac. Sci., 19, December 16, 1844, pp. 1262-1263 (D.C., 1, 8, pp.366-375).
Anotherproofisin the C.R. Ac.Sci., 19,December 23, 1844,pp. 1377-1384 (D.C., 1,
8, pp. 143-153). It is known that this theorem can easily be deduced from the
Cauchy inequalities.
53. C.R. Ac. Sci., 23, October 12, 1846, pp. 689-702 (D.C., 1,10, pp. 152-168).
54. Without explanation, Cauchy ceased publishing on his theory on October 26,
1846. Puiseux, nevertheless, indicates the existence of a study in which Cauchy
recovered the known double periodicity of elliptic functions.
55. c.R. Ac. Sci., 23, September 21, 1846, p. 566 (D.C., 1, 10, p. 146).
56. 'Memoire sur les fonctions de variables imaginaires', c.R. Ac. Sci., 23, August 10,
1846, pp. 271-275 (D.C., 1,10, pp. 75-80), and Ex. An. Phys. Math., 3, October
1846, pp. 361-387 (D.C., 2, 13, pp. 405-436).
57. 'Memoire sur la theorie des equivalences algebriques, substituee ala theorie des
imaginaires', C.R. Ac. Sci., 24, June 28,1847, pp. 1120-1130 (D.C., 1, 10, pp. 312-
324), and Ex. An. Phys. Math., 4, pp. 87-110 (D.C., 2,14, pp. 93-120).
58. The two polynomials X(x) and p(x) are equivalent moduli of the polynomial w(x)
[notation: X(x) == p(x) mod. w(x)] if upon division by the polynomial w(x), the
remainders are equal.
59. See Ex. An. Phys. Math., 4, pp. 97-100 (D.C., 2, 14, pp. 104-108).
288 Notes

60. C.R. Ac. Sci., 24, March 1, 1847, p. 310. Lame had long been interested in Fermat's
last theorem and had competed in 1818 in the Grand Prix de Mathematiques of the
Academie des Sciences. In 1839, he showed the impossibility of solving, in integers,
theequationx 7 + y7 = Z7. Cauchy wrote the report, which was highly laudatory. C.
R. Ac. Sci., 9, September 16, 1839, pp. 359-363 (D.C., 1, 4, pp. 499-504).
61. C.R. Ac. Sci., 24, March 15, 1847, pp. 430-434.
62. Sealed messages no. 726 and no. 727.
63. C.R. Ac. Sci., 24, March 22 and 29, and April 5, 12, and 19, 1847, pp. 469-481,
pp. 516-528, pp. 578-584, pp. 633-636, and pp. 661-666 (D.C., 1, 10, pp.240-
285).
64. C.R. Ac. Sci., 24, May 17, 1847, pp. 899-900.
65. C.R. Ac. Sci., 24, May 24 and 31, and June 7,14, and 28,1847, pp. 885-887, p. 943,
pp. 996-999, pp. 1022-1030, and pp. 1117-1130, and C.R. Ac. Sci., 25, July 5, 12,
19, and 26, and August 2, 9, and 23,1847, p. 6, pp. 37-54, pp. 93-99, pp. 129-136,
pp. 177-182, pp. 242-243, and pp. 285-288 (D.C., 1,10, pp. 292-371).
66. c.R. Ac. Sci., 24, May 31, 1847, p. 88 (D.C., 1, 10, p. 295).
67. For example, he showed that a polynomial formed from a 23rd root of unity could
not be factored into prime factors. C.R. Ac. Sci., 24, June 14, 1847, pp. 1022-1030
(D.C., 1,10, pp. 299-308).
68. C.R. Ac. Sci., 24, June 28, 1847, p. 1121 (D.C., 1, 10, p. 312), and 'Memoire sur la
theorie des equivalences algebriques, substituee ala theorie des imaginaires', Ex.
An. Phys. Math., 4, pp. 87-110 (D.C., 1,14, pp. 93-120).
69. J. B. Biot, 'M. Ie Baron Cauchy', Melanges Scientifiques et Litteraires, 3, Paris,
1858, p. 152.

Notes to Chapter 13
1. F. N. M. Moigno, preface to A. L. Cauchy, Sept Lerons de Physique Generale ... ,
Paris, 1868.
2. A. L. Cauchy, 'La Chandeleur', Bulletin de l'Institut Catholique, February 1, 1843.
3. A. Fourcy, Histoire de l'Ecole Poly technique, Paris, 1828, Summary Table, pp. 255-
257.
4. See Chapter 5, pp. 61-63.
5. Cited by G. Bertier de Sauvigny, La Restauration, Paris, Flammarion, 1955, p. 346.
6. A. L. Cauchy, 'Sur la recherche de la verite', Bulletin de l'Institut Catholique, April
14, 1842, p. 21.
7. A. L. Cauchy, Sept Lerons de Physique Genf!rale ... , D.C., 2, 15, p. 413.
8. Ibid., p. 413.
9. A. L. Cauchy, 'Sur les !imites des connaissances humaines', D.C., 2, 15, p. 7.
10. M. Jullien, 'Quelques souvenirs d'un etudiant jesuite a la Sorbonne et au
College de France, 1852-1856', Les Etudes, 127, 1911, pp. 329-348, especially,
p.336.
11. A. L. Cauchy, Quelques Reflexions sur la Liberte de I'Enseignement, Paris, 1844,
especially Chapter 3, 'De l'enseignement scientifique et de l'enseignement re-
ligieux', pp. 14-16.
12. A. L. Cauchy, Considerations sur les Moyens de Prevenir les Crimes et de Reformer
les Criminels, Paris, 1844.
13. A. L. Cauchy, 'Sur la recherche de la verite', Bulletin de l'Institut Catholique, April
14, 1842, p. 21.
 Notes 289

14. A. L. Cauchy, Considerations sur les Moyens de Prevenir les Crimes et de Reformer
les Criminels, Paris, 1844.
15. Sept Ler;ons de Physique Generale ... ,D.C., 2,15, p. 413.
16. Practically in the same terms in Sur les limites ... of 1811, D.C., 2, 15, pp. 5-6; Sept
Ler;ons de Physique Generale ... of 1833, ibid., pp. 412-413; and 'Sur la recherche de
la verite', Bulletin de l'Institut Catholique, Apri114, 1842, p. 20.
17. Sept Ler;ons de Physique Generale ... , D.C., 2, 15, p. 412.
18. A. L. Cauchy, 'Sur quelques prejuges contre les physiciens et les geometres',
Bulletin de l'Institut Catholique, March 3, 1842, p. 43.
19. Analyse Algebrique, introduction, pp. V-VI (D.C., 2, 3, pp. V-VI).
20. 'Sur quelques prejuges ... " Bulletin de l'Institut Catholique, March 3, 1842, p. 43.
21. Sept Ler;ons de Physique Generale ... , D.C., 2, 15, p. 418.
22. Ibid., p. 415.
23. Ibid., p. 419.
24. 'Sur quelques prejuges ... ', Bulletin de l'Institut Catholique, March 3, 1842, p. 45.
25. A. L. Cauchy, 'Epitre d'un mathematicien a un poete, ou la lecon d'astronomie',
Bulletin de l'Institut Catholique, January 13, 1842, p. 20.
26. 'Sur la recherche de la verite', Bulletin de l'Institut Catholique, April 14, 1842, p. 21.
27. Analyse Algebrique, introduction, p. VII (D.C., 2, 3, p. VII).
28. The manuscript for this unpublished note from 22, January 22,1837, is kept in the
meeting's packet for that day.
29. C.R. Ac. Sci., 21, July 14, 1845, pp. 134-143 (D.C., 1,9, pp. 240-253).
30. Ibid., p. 134 (D.C., 1,9, pp. 240).
31. 'Sur les limites .. .', D.C., 2, 15, p. 6.
32. Analyse Algebrique, introduction, pp. III-V (D.C., 2, 3, pp. III-V).

Notes to Chapter 14
1. Proces-verbaux du Bureau des Longitudes, cited by Bigourdan, art. cit., 1930, A 33.
2. C. R. Ac. Sci., 26, April 3 and 17, and May 1, 1848,pp. 404-408,p. 429,pp. 441-443,
pp. 448-449, and pp. 469-472 (D.C., 2, 11, pp. 30-49).
3. See C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 105.
4. Eugene Briffault wrote in Dictionnaire de la Conversation et de la Lecture, Paris, 2nd
ed., 1852-1876, article on Cauchy:
During the reign of Louis-Philippe, the Cauchys were not just a family, but a triple-
almost a dynasty. This race of bureaucratic climbers was attached to whichever party
happened to be on top at the Palais du Luxembourg; there, they formed a colony; they
were identified with the place, part and parcel, to such an extent that they themselves
constituted a party. It was said that in order to get rid of the Cauchys you would have to
tear the place (i.e., the Palais du Luxembourg) down. Thus, for a long time,
revolutionary events would come and go without apparently touching the Cauchy's
pleasant position.
The heredity of a peerage failed to materialize, but aside from this, the Cauchy
heredity remained untouched. This succession seemed destined to continue indefinitely
into the future, with no end in sight. Vanitas vanitatum! Then! All at once, the fatal hour
sounded on February 24, 1848. The Republic was proclaimed, and the first blows of the
clock's hammer scattered the Cauchys into flight. Nothing about this revolution was so
unbelievable as that they had been forced to clear out of the Palais du Luxembourg.
5. See Arch. Nat. F17 20356.
290 Notes

6. These letters were written by Cauchy after certain accusations had been made
against the Jesuits at the meetings of February 23 and 25, 1850. In the first letter,
Cauchy took up the arguments that had been used in the opuscules of 1844; the
second letter contains a historical account of the suppression of the order during
the 18th century.
7. D.C., 2, 15, pp. 511-513.
8. See Chapter 11, pp. 185-186.
9. See A. L. Cauchy, Note read before the Academie on January 6,1851, Appendix IV,
p.357.
10. See Appendix IV, p. 346.
11. Archives of College de France, G II 5, Registre des deliberations de l'assemblee des
professeurs du college, December 1, 1850. See E. Neuschwander, 'Joseph Liouville
(1809-1882): correspondance inedite et documents biographiques proven ant de
differentes archives parisiennes', Bolletino di Storia delle Scienze M atematiche, 4,
1984, pp. 55-132, especially p. 125.
12. See Appendix IV, p. 347.
13. Archives of College de France, G II 5, Registre des deliberations de l' assemblee des
professeurs du college, December 8, 1850. See E. Neuschwander, 'Joseph
Liouville .. .', p. 125.
14. See Appendix IV, p. 354.
15. Letter from Cauchy to the administration of the College de France, Appendix IV,
p.355.
16. c.R. Ac. Sci., 32, January 6, 1851, p. 3.
17. Report by Cauchy, c.R. Ac. Sci., 32, March 31, 1851, pp. 442-450 (D.C., 1, 11,
pp. 363-373); remarks by Liouville, ibid., pp.450-452; note by Cauchy, ibid.,
pp. 452-454 (D.C., 1, 11, pp. 373-376). The manuscript of Liouville's course on
doubly periodic functions is kept in the meeting's packet for that day.
18. See B. Belhoste et J. Liitzen, 'Joseph Liouville et Ie College de France', Rev. Hist.
Sci., 1984, 37, pp. 255-304, especially pp. 274-278.
19. Drafts of letters from Liouville to Cauchy, Bibliotheque de I'Institut, Ms 3623,
book note 1, dated July 9, 1856. See E. Neuenschwander.. 'Joseph Liouville ... "
p.112.
20. See Appendix III, p. 336.
21. The course of mathematical astronomy was held during the second semester ofthe
school year (the first semester was devoted to physical astronomy). Faye was
appointed on March 13, 1853, as a substitute for Cauchy and Leverrier on
December 24, 1853 (Bulletin administratif du Ministere de l'Instruction Publique, 1,
4, 1853, p. 89 and p. 649). Moigno in his preface to Sept Le(:ons de Physique
Generale ... of 1885 wrongly stated that Cauchy gave up his chair in 1851 and
resumed it in 1853.
22. See Arch. Nat. F17 20356 and Moigno, preface to Sept Le(:ons de Physique
Generale . .. , p. VII.
23. M. Jullien, 'Quelques souvenirs d'un etudiant jesuite.. .', p. 127, p. 191, and
p.336.
24. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 2, introduction, p. VIII.
25. Reports presented in July 1848 by Laurent; in January 1849 and July 1850 by La
Provostaye and Desains. During this time, Cauchy also examined, but did not
make evaluation reports on, several important studies by Saint-Venant on the
theory of elasticity.
 Notes 291

26. c.R. Ac. Sci., 27, July 31, August 14,21, and 28, October 9, and 16, November 13,
20, and 27,December4, 11, and 18, 1848,p. 105,p. 133,p. 162,p. 198,p. 225,p. 356,
p. 373, p. 433, p. 499, p. 525, p. 537, p. 572, and p. 596 (O.c., 1,11, p. 73 and pp. 76-
91).
27. C.R. Ac. Sci., 27, December 18, 1848, pp. 621-622; c.R. Ac. Sci., 28, January 2 and 8,
1849, pp. 2-6 and pp. 25-28 (O.C., 1, 11, pp. 92-104); and M em. Ac. Sci., 22, pp. 29-
37 (O.C., 1,2, pp. 187-194).
28. C.R.Ac. Sci., 28, January 15, 1849,pp. 57-65 (O.C., 1, 11,pp. 104-113), and C.R. Ac.
Sci., 31, July 29, August 5, 19, and 26, September 2, 9, and 16, October 7, and 14,
November 11, and December 2 and 231850, pp. 112-114, pp.160-166, pp. 225-
232, pp. 257-262, pp. 297-306,pp. 331-342, pp. 532-533, pp. 666-667, and p. 766
(O.c., 1, 11, pp. 245-289). Cauchy based the theory of reflection and refraction of
light on two principles: that of corresponding motions and the continuity of motion
in the ether.
29. C.R. Ac. Sci., 27, January 2, 1849, p. 2 (O.c., 1, 11, p. 95).
30. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 252.
31. C.R. Ac. Sci., 25, September 20, October 4, 18,25, November 8, 15, and 29, and
December 13 and 27,1847, pp. 401-413, pp. 475-478, pp. 531-538, pp. 572-579,
pp.650-656, pp.700-705, pp. 775-781, pp. 879-883, and pp.953-959, and 26,
January 10, 17,24, and 31, and February 21, 1848, pp. 29-33, pp. 57-61, pp. 133-
136, pp. 157-162, and pp. 236-240 (O.c., 1, 10, pp. 403-499).
32. C.R. Ac. Sci., 29, July 23 and 30, 1849, pp. 65-67 and pp. 103-106 (O.c., 1, 11,
pp.141-147).
33. C.R. Ac. Sci., 33, December 29, 1851, pp. 649-709; c.R. Ac. Sci., 34, January 5, 19,
and 26, and February 2,1852, pp. 8-9, pp. 70-77, pp. 121-124, and pp. 156-159
(O.c., 1, 11, pp. 385-403).
34. C.R. Ac. Sci., 38, May 22 and 29, and June 5, 12, and 26, 1854, pp. 910-913, p. 945,
p. 952, pp. 990-993, and p. 1033 (O.C .• 1, 12, pp. 148-167).
35. On August 11, 1856, Cauchy deposited a sealed envelope (no 1591) containing the
study 'Note sur l'integration des equations differentielles qui renferment les
mouvements des astres dont se compose Ie systeme solaire'. This unpublished
paper is in the meeting's packet for that day.
36. C.R. Ac. Sci., 44, April 27, and May 4,1857, p. 528, p. 595, and pp. 805-807 (O.c., 1,
12, pp. 445-446). These were the last papers that Cauchy wrote; a few days before
his death, he was working on a paper in astronomy (C. A. Valson, La Vie et les
Travaux du Baron Cauchy, 1, p. 259).
37. J. Bertrand, 'La Vie et les Travaux du Baron Cauchy par C. A. Valson', p. 214.
38. 'Astronomie mathematique, cours de M. Cauchy', contained in the handwritten
notebook Mathematiques. Cours Inedits by G. Lespiault, Bibliotheque de la
faculte des sciences de Bordeaux, Ms 52 (old mark). These notes, filling eight leaves
recto verso, concern the theory of geometric quantities and the theory of the
functions of geometric quantities.
39. This fragment, which Professor Roos of the University of Stockholm has
communicated to me, is kept in the papers of C. A. B. Bjerknes in the library of the
University of Oslo.
40. A copy of Report to the ministry for the church and education about ajourney abroad
to study pure mathematics, in Norwegian, is kept in the papers of Bjerknes, at the
University of Oslo, Professor Roos has translated for me in English the passage of
the report concerning Cauchy:
292 Notes

I arrived in Paris on October 15 [1856]. The winter semester had not yet started
there ....
On November 15, the courses started at the Sorbonne.... Cauchy was supposed to
lecture about the movements of the celestial bodies, when they are attracted by the sun.
In this course, which in fact was the last one that the eminent mathematician taught, he
developed a whole series of mathematical theories. He developed in a particularly
instructive and original way the foundations for almost all of those beautiful and
important theories, that current science owes to him, e.g., his theory of indices, his
calculus of residues, the theory about the criteria for convergence of the Taylor series,
and the theory of the so-called geometric quantities. In view of the completeness of his
treatment ofthe preliminaries, he only had a few opportunities to treat the real subject
of his lectures, that by the way for me only would have had a minor interest, as being
something that was outside my real subject of interest....
Incidentally, the lectures that I attended had a very small audience. While the
elementary lectures about, e.g., integral calculus and analytic geometry, were attended
by rather many auditors, it turned out that the higher, purely mathematical subjects,
even in Paris, only attracted a small audience. Thus, Liouville's and Cauchy's lectures
were only attended by 3, 4, or 5 persons, and the lectures of Bertrand had an audience of
6-8.
41. See J. Peiffer, Les Premiers Exposes Globaux de la Theorie des F onctions de Cauchy,
3rd cycle Thesis, typewritten, 1978, pp. 94-123.
42. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 253.
43. See J. Peiffer, Les Premiers Exposes Globaux ... , pp. 82-93.
44. Ex. An. Phys. Math., 4,1849, pp. 157-180 (O.C., 2, 14, pp. 175-202), and Mem. Ac.
Sci., 22, 1850, pp. 131-180 (O.C., 1,2, pp. 282-328).
45. Two letters from Saint-Venant to Cauchy from the end of 1845, which are kept in
Archives E. P., respond to certain objections that Cauchy had raised and further
reveal that he (Cauchy) was not yet convinced of the importance of this new
geometrical calculus. In a letter dated December 27,1845, Saint-Venant defended
his calculus in the following terms:
I realize that all that is proved by way of the kind of geometrical calculus that I can
develop here, perhaps, is established without recourse to it. But, by use of it, proofs can
be made more simple when, once and for all, the terms have been defined, and it has
been shown that geometrical equations are to be treated in exactly the same way as
ordinary equations. It is rare that a new algorithm solves questions that cannot be
solved in some other way; in general, it will only provide a simplification. If it is to be
granted that I may compare my work to something that is clearly superior, then I
should compare it to M. Poinsot's theory of couples, or to other analytical theories that
focus on the same matters.
46. Ex. An. Phys. Math., 4, 1849, p. 157 (O.c., 2, 14, p. 175).
47. C.R. Ac. Sci., 29, November 19, 1849, p. 594, Charles Hermite, Oeuvres, 1, Paris,
1905, p. 74.
48. See C. Houzel, 'Histoire de la theorie des fonctions elliptiques' in Abrege d'Histoire
des Mathematiques, 2, Ch. VII, 1-113, especially p. 22. The study by Hermite was
not published. It is not in his Oeuvres, 'having disappeared from the archives of the
Academie' (Emile Picard). However, today it is kept in the meeting's packet of
November 19, 1849, except for the first few pages, which are missing.
49. See G. Darboux, Notice Historique sur Charles Hermite, Paris, 1905, particularly
p.29.
50. C. A. Valson La Vie et les Travaux du Baron Cauchy, 1, p. 253.
 Notes 293

51. M. Jullien, 'Quelques Souvenirs d'un etudiant jesuite ... " p. 340.
52. Memoir published in the J.L., 15, 1850, pp. 365-480.
53. c.R. Ac. Sci., 32, January 20,1851, pp. 68-75 (D.C., 1, 11, pp. 292-300).
54. c.R. Ac. Sci., 32, February 10, 1851, pp. 160-162 (D.C., 1, 11,pp. 301-304). In 1829,
Cauchy had already given the definition ofthe derivative of a function of a complex
variable, but he had not given the conditions for the differentiability of such a
function.
55. c.R. Ac. Sci., 32, April 7, 1851, pp. 484-487 (D.C., 1, 11, pp. 376-380).
56. c.R. Ac. Sci., 32, May 12, 1851, pp. 704-705 (D.C., 1, 11, pp. 384-385).
57. C.R. Ac. Sci., 34, February 23, 1852, pp. 265-273 (D.C., 1,11, pp. 406-415).
58. Ex. An. Phys. Math., 4, November 1853, pp. 308-313 (D.C., 2, 14, pp. 359-366).
Cauchy first defined a real function of a real variable:
Two real variables are said to be functions of each other when they vary simultaneously
in such a way that one of them determines the value of the other.
For functions of geometrical quantities, he adopted the following definition:
... Z is to be deemed a function of z when the value of z determines the value of Z ....
Then, the position ofthe moving point A always determines the position of the moving
point B.
59. Ex. An. Phys. Math., 4, November 1853, pp. 314-335 (D.C., 2, 14, pp. 367-392).
60. Ex. An. Phys. Math., 4, November 1853, pp. 336-347 (D.C., 2, 14, pp. 393-406).
61. See the papers of January 15, February 12, and December 31,1855; the reports by
Cauchy of March 12, 1855, C.R. Ac. Sci., 40, March 12, 1855, pp. 557-567 (D.C., 1,
12, pp. 243-246) and of July 7,1856, c.R. Ac. Sci., 43, July 7,1856, pp. 26-29 (D.C.,
1,12, pp. 330-333); and c.R. Ac. Sci., 43, pp. 13-20 and pp. 69-75 (D.C., 1,12,
pp. 323-330 and pp. 333-342).
62. C.R. Ac. Sci., 40, April 2, 1855, pp. 787-789.
63. From a letter that Meray wrote to the Minister of Public Instruction in 1863. It was
brought to our attention by Professor Pierre Dugac:

Among my scientific works, I can mention to your Excellency, aside from my thesis,
... the publication of the Cours d'Ana/yse, by M. Cauchy, which was undertaken at the
request of his family and ofM. Saint-Venant; this work is presently being published.

The book obviously never appeared.

64. Ex. An. Phys. Math., 4, November 1853, p. 356 (D.C., 1, 14, p.417). The first
mention ofthe calculus of algebraic keys appears on January 10, 1853, C.R. Ac. Sci.,
36, January 10, 1853, pp. 70-75 (D.C., 1, 11, pp.439 fl). Cauchy also used his
calculus of keys in the course he taught at the College de France.
65. Not knowing Saint-Venant's address, Grassmann wrote to Cauchy in April 1847,
asking him to forward his letter to Saint-Venant, along with one ofthe two copies
of the Ausdehnungslehre of 1844. On June 17, 1847, Saint-Venant replied to
Grassmann who sent him another letter, along with several papers, in January
1848 (H. G. Grassmann, Gesammelte mathematische und physikalische Werke, 3,
Leipzig, 1911, pp. 120-122).
66. See the meeting's packet for that day. Grassmann claimed priority for the theory of
algebraic keys that Cauchy presented in 1853, as well as on the geometrical
calculus that Saint-Venant presented in 1845 (H. G. Grassmann, Gesammelte
mathematic he und physikalische Werke, 3, pp. 174-198).
294 Notes

67. The draft of the note from Saint-Venant to Cauchy is kept in the Fonds Saint-
Venant, Archives E. P., and the letter from Saint-Venant to Grassmann is
published along with his reply in Grassmann, Gesammelte Werke, 3, pp. 199-200.
68. See Archives Ac. Sci., Cauchy file. The list gives the investigations that the
Academie obtained from the Cauchy family.
69. See the critical analysis of this dispute in H. Freudenthal, 'Cauchy, Augustin-
Louis', Dictionary of scientific biography, 3, pp. 141-142.
70. See M. lullien, 'Quelques souvenirs d'un etudiant jesuite.. .', p. 338.
71. See the account by Guyon, Mayor of Sceaux, at the funeral services for Cauchy:
Almost every day he comes to visit me; often sometimes several times a day. They are
short visits and are free of all foolish chatter. Time was too valuable for someone who
was making such a valuable use of it.

Also see that of M. lullien 'Quelques souvenirs d'un etudiant jesuite... " pp. 334-
335:
From time to time, he would come to the little chateau (a tiny building from wood at the
bottom of the entrance to the house across from the door, the pressed and busy famous
Parisien) and find us in our little study. He came like a puff of wind, without knocking,
without uttering any compliments, and started to talk about his business... On another
day, as I was quietly meditating at five o'clock in the morning, I suddenly heard
someone coming hurriedly up the wooden stairs. I thought something untoward was
happening. M. Cauchy came in like someone who was about to miss his train: 'Father,
read this quickly; it is a hymn that I was asked to give for the rete de Sainte-Genevieve
at Saint-Etienne du Mont. It is to be sung at the seven o'clock mass. I forgot to show
it to any theologian. Maybe it contains some heresay or some incorrect, improper
expression. Read it over, and I will correct it right away'.
Then, see that of Armand de Melun, cited by Alexis Chevalier in Le Vicomte
Armand de Melun, Fondateur de ['Oeuvre des Jeunes Apprentis et des Jeunes
Ouvriers, Versailles, 1893, p. 25:

How considerable were the efforts-the canvassing and the discussions-that he went
through to assure this 'conquest' [the works of the Ecoles d'OrientJ. Cauchy often
woke me up at daybreak to advise me about ways of getting around this inconvenience
or of overcoming that obstacle.

72. See C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 223-242.
73. See the letter from Armand de Melun cited by A. Chevalier, Le Vicomte Armand de
Melun ... , p. 25:
In the Middle Ages, Baron Cauchy would have been the first person to take up the cross
to march to the conquest of the holy places. But, in the present age, he has taken up a
new crusade: to establish the Ecoles d'Orient.
See also C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 226.
74. Ibid., 1, p. 236.
a
75. H. de Lacombe, Note sur l'Oeuvre d'Orient ['Occasion du Cinquantenaire de sa
Fondation (1856-1906), Paris, 1906.
76. C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, p. 237, and the letter from
Cardinal Lavigerie to E. Beluze, which serves as a preface to E. Beluze, La Vie de
Mgr Dauphin, Prelat de la Maison de leurs Saintetes Pie IX et Leon XIII, 1806-
1882, Paris, 1886.
 Notes 295

77. From the speech by Guyon, Mayor of Sceaux, at the funeral services of Cauchy,
cited by C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 272-274.
78. For more about this controversy, see c.R. Ac. Sci., 43, December 8,1856, p. 1065
(remarks by Bertrand), p. 1067 (remarks by Cauchy); December 22, 1856,
pp. 1137-1139 (D.C., 1, 12, pp. 395-398, notes by Cauchy); December 29, 1856,
pp. 1165-1166 (remarks by Duhamel), and pp. 1166-1167 (D.C., 1, 12, pp. 398-
401, reply by Cauchy). c.R. Ac. Sci., 44, January 5, 1854, pp. 3-5 (new remarks by
Duhamel), January 19, 1857, pp. 80-81 (D.C., 1, 12, p.405, reply by Cauchy),
pp.81-82 (reply by Duhamel), pp.82-89 (remarks by Poncelet), pp.89-91
(remarks by Morin), pp. 101-104, 'Sur quelques propositions de mecanique
rationelle', by Cauchy (not published in the D.C.), p. 104 (reply by Duhamel), and
pp. 104-107 (new intervention by Poncelet).
79. C. A. Val son, La Vie et les Travaux du Baron Cauchy, 1, p. 255.
80. Ibid. 1, p. 259. Cauchy's mother died in 1839, and his father died in 1848. His two
sisters were married, the eldest to Felix de I'Escalopier, an official at the Court of
Accounts, and the younger to Alfred de Saint-Pol. His younger brother Eugene
had been widowed since 1839. Alolse Cauchy died on June 13, 1863 at Sceaux.
81. Augustin Cauchy, Discours aux Fum!railles de Jacques Binet, Paris, 1856.
82. Letter written by Alicia to Pere Coue, S. 1., on May 23, 1857, cited by C. A. Valson,
that reports, in vivid terms, the pain and exemplary death of the great scientist. See
C. A. Valson, La Vie et les Travaux du Baron Cauchy, 1, pp. 259-267.
 Appendix I

Presentation Notes of Various Works

Presented by Cauchy to the Academie des
Sciences (1816-1830)

Study on the Integration of Linear Partial Differential Equations

with Constant Coefficients (Memoire sur l'integration
des equations lineaires aux differences partielles et a
coefficients constants)l

This study, in which I discuss the main results of my lectures at the Faculte des
Sciences, is an expansion of the material that I presented, under the same title,
to the Academie des Sciences on October 8, 1821,2 There is, however, a
difference in the way that the main question is examined, and it includes some
important additions. The work is divided into two parts. In the first part, I
establish certain formulas which will be used later on, either to solve the partial
differential equations or to reduce the integrals obtained. Among these
formulas should be noted those that, in the third section, are used to transform

ffr··
the values of the multiple integral

f(a. 2 + 1J2 + y2 + ... )cos aa. cosbP cos cy··· da. dP dy* (1)

taken between the limits - 00 and + 00 over all the variables, for a simple or
double integral, and those that are used to determine one or several real or

1 Presentation of the study to the Academie; the reading was on September 16, 1822. The
manuscript is in the meeting's packet for that day. The study is published in J.E.P., 12, 19th Cahier,
July 1823, pp. 510-570 (D.C., 2, 1, pp. 275-333).
a
2The paper mentioned here is 'Sur !'integration generale des equations lineaires coefficients
constants'; published in Bull. Phil., October 1821, pp. 101-112 (D.C., 2, 2, pp.253-266) and
November 1821, pp. 145-152 (D.C., 2, 2, pp. 267-275).
*Ifthe number ofvariables ... is odd, by using n to denote this number and writing a2 + b2 + c 2

f
+ ... = s, I find that the integral in Eq. (1) is equivalent to the product:
0(n-1)/2
(_I)n-1/22n-171n-1/2___
os(n-1)/2
cos(sta)J(a 2 )da (a =- 00, a = + 00).
[Note by Cauchy.]

296
 Appendix I 297

imaginary roots of an arbitrary algebraic or transcendental equation-or

more generally, the sum offunctions similar to several of the systems of values
that may have various variables subjected to certain given equations.
In the second part of the study, I apply certain formulas that are found in
the first part to the solution of linear equations with constant coefficients
whose last term is a variable. I determine the general value of the principal
variable expressed by means of the functions that represent the initial values of
this variable and its derivatives; and I show that in this determination one can
dispense entirely with the solution of the algebraic equations. Finally, I point
out the reductions that hold for a special class of partial differential equations
that are related to a large number of problems in physics and mechanics.

Research on Definite Integrals That Contain Imaginary

Exponentials (Recherches'sur les integrales definies
qui renferment des exponentielles imaginaires)3
These investigations were undertaken in 1819 following the study on the
solution of algebraic equations by means of definite integrals. 4 These
questions lead me to some important formulas, which are related to the
transformation of the functions and serve to replace an arbitrary function of
the variable x by a definite integral in which there enters no more than one
rational fraction of this variable whose denominator is of the first degree.

Research on the Motion of Two Superimposed Fluids, One of

Which Is Compressible While the Other Is Incompressible
(Recherche sur Ie mouvement de deux fluides superposes,
l'un compressible, l'autre incompressible)5
After having finished the study on the theory of waves, which won a prize from
the Academie in 1815, I undertook some investigations on the motion of two
superimposed fluids when one of them is compressible and the other is
incompressible; and I was able to establish the formulas that serve to
determine the waves produced by the contact between the two fluids. These
formulas are found in the notebook that includes the notes appended to my
first study. I thought about removing them and to develop a separate; but, time
would not allow me to do so, and this subject, not being suitable for extensive,
detailed treatment in a lecture, I am bound to offer them to the Academie in

3Presentation of the study to the Academie. The reading was on September 16, 1822; the
manuscript is in the meeting's packet for that day. The study is lost.
4The study is 'Sur la resolution analytique des equations de tous les degres par Ie moyen des
integrales definies' of November 22, 1819 (Memoires Ac. Sci., 4 (1819-1820), 1824, pp. XXVI-
XXIX; O.c. 1,2, pp. 9-11).
5 Presentation of the study was on January 27,1823. The manuscript is in the meeting's packet for
that day. The study, which is unpublished, is inserted in the Cahier sur la Theorie de Ondes,
belonging to Mrs. de Pomyers, pp. 115-140.
298 Appendix I

their original form. I do not know if the analysis that I make use of and that, it
seems to me, should merit the consideration of geometers is related in some
way to the one by M. Poisson in which he established the laws of reflection and
refraction of light in systems of undulations, as he stated in the secret
committee at the last meeting. 6

On the Effects of Molecular Attraction in the Motion

of Waves (Sur les effets de l'attraction moleculaire dans
Ie mouvement des ondes)7
The more or less considerable viscosity of a liquid gives rise to the attractions
that its molecules exercise on each other at very small distances and necessarily
exerts an influence on the motion of waves at the surface of the liquid. I
propose to determine the nature of this influence. In order to accomplish this,
it is necessary to integrate the equations of motions for fluids, after having
introduced new terms that depend on the molecular attraction and to then
suitably determine the arbitrary functions, based on the initial data of the
problem. These are, in effect, the questions that I have resolved by means of an
analysis that offers the following noteworthy fact: it provides a way of
integrating the equations of motion of a fluid even in the situation when the
accelerations along axes multiplied, respectively, by the differentials of the
coordinates do not yield exact differentials. This analysis, when applied to the
case in which the viscosity is zero, yields the formulas that M. Poisson and I
obtained by other methods in earlier studies. Moreover, when one takes the
viscosity of the liquid into consideration, the variable that represents the
mutual attractions of the two molecules and is regarded as a function of the
distance (which decreases very rapidly) vanishes as soon as the integrations
have been performed and the arbitrary functions determined-and, in their
places, in the definitive formulas, only two constants are left. One of these
constants is proportional to the attractions exerted in a point on the surface of
the liquid by the neighboring molecules, while the other one is proportional
the sum of the actions exerted on this point and on those that are below it
along the same vertical. In particular, in order to obtain the equation that shall
determine the vertical ordinate ofthe surface of the liquid, at the end ofa given

6 The study is Mernoire sur La propagation du mouvement dans Lesfluides elastiques, read by Poisson
to the Academie on March 24, 1823. Some extracts were published in Annales de Chimie, 2,22,
1823, and Mbnoires. Ac. Sci. (1827) 1, 1831.
'Presentation of the study to the Academie. The reading was on November 17, 1823; the
manuscript is in the meeting's packet for that day. At the end of note XX of the study 'Sur la theorie
de la propagation des ondes', which was added again for publication in 1824, Cauchy declared:
In another study, we will derive by the preceding methods the formulas that we presented to the
Academie Royale des Sciences on November 17 last, and by means of which we have determined
the motion of waves on the surface of a heavy liquid, keeping in mind the adhesion that exists
between its molecules.
This study was never published.
 Appendix I 299

time t, when the initial ordinate is known, it suffices to take the equation that
furnishes this ordinate in the case in which the viscosity is zero, and then (1) the
weight of a quantity equal to the first constant and (2) the initial ordinate and
the ordinate at the end of time t by a quantity equal to the second.

A Study on Various Points in Analysis (Memoire sur

divers points d'analyse)B

I have the honor of presenting to the Academie a study on various points in

analysis. In this study, I begin by deriving certain formulas that I have
previously given for definite integrals, not only the Taylor and Lagrange series,
but also a considerable number of other series of the same type and on the
remainder terms needed to complete these various series. The same principles
furnish the means for expressing the sum of similar functions of the roots of an
algebraic equation by means of the coefficients of the (given) equation. I then
show how it is possible to directly establish (1) the formulas that serve to
determine the sum of the functions in question and (2) those formulas that are
used to determine their product. In making use of these formulas, it is possible
to directly compose, without recourse to the theory of symmetric functions, the
final equation that should result from the elimination of several unknowns
between the algebraic equations.

A. L. Cauchy

A Study on the Integration of Linear Equations and Their

Application to Various Problems in Physics (Memoire sur
l'integration des equations lineaires et leur application it divers
problemes de physique)9

I am honored to present to the Academie two new papers on the integration of

linear equations and the determination of the arbitrary functions that they
involve. In these studies, I establish several notable properties of M. Fourier's
formula, and I show how, by applying to this formula a system of notation
proposed by M. Brisson, the results that he obtained can be rigorously proved.
I then propose a direct method by means of which one can determine, in a large
number of cases, the arbitrary functions by assuming only that the initial
functions are known between certain limits, and I apply this method to the
solution of several problems in mathematical physics. Among these problems,

8 Presentation of the study to the Academie. The reading was on August 9, 1824. The manuscript is
in the meeting's packet of August 9, 1824.
9Note of presentation read to the Academie on December 27, 1824. The manuscript is in the
Staatsbibliothek Preussischer KuIturbesitz, Dokumentensammlung Darmstaedter F lc 1836 (1).
This note was used almost completely by Fourier in the mathematics part of the Histoire de
I'Academie for 1824 (Memoire Ac. Sci., 7, (1824),1827, pp. XLV-XLVI).
300 Appendix I

I mention the propagation of waves in a canal of finite length or in a

rectangular basin, regardless of the depth of the liquid. One result following
from my formulas is that the waves produced in a rectangular basin are
absolutely the same as if the basin were extended indefinitely in all directions;
the initial surface of the liquid in the basin is continually reflected off its sides in
such a way that two continuous rectangles with sides equal to those of the
basin are always recovered by two symmetrical and symmetrically placed
surfaces on all sides of the vertical plane formed by the joint side of these two
rectangles. If, for the sake of greater convenience, one should assume that the
surface of the basin is contained by four vertical plane mirrors, the surfaces we
will construct outside the basin will not be different from the images of the
original surface of the liquid reflected an infinity of times by the four mirrors.
This conclusion holds even when one or several of the mirrors recede to
infinity, that is to say, when the basin is extended indefinitely in one or several
directions. If three of the mirrors recede to infinity, there will be no more than
one image; and, at the same time, one would obtain the motion ofthe waves on
the surface of a liquid bounded by a vertical plane. If, in this last case, the
ordinate of the initial surface of the liquid is always zero except in the
neighborhood of a certain point, then this surface, when extended, has an
ordinate that is zero except for the neighborhood of a second point, which will
coincide precisely with the image of the first, and the surface of the liquid will
be as if it were extended indefinitely beyond the plane and if the two points in
question would be two centers of the motion of the waves. From this, it can be
immediately concluded that the circumferences of the circles formed by
different incident and reflected waves are cut along the given plane in such a
way that the radius of the incident wave and the radius of the reflected wave
always intersect at the same angle with the normal to the plane. This is what
takes also place, as is known, in the theory of light where it is regarded as a
product of the motions of the waves of an etheral fluid.
December 27, 1823 (sic)
A. L. Cauchy

A New Study on the Integration of Linear Equations

and on the Motion of Elastic Plates (Nouveau
memoire sur l'integration des equations lineaires et sur
Ie mouvement des plaques elastiques)10
I am honored to offer to the Academie a new study on the integration oflinear
equations in which I present the general solution for several equations of this
type; particular attention has been given to those that are related to the motion

lONote of presentation read before the Academie on January 17, 1825. The manuscript is in the
meeting's packet for that day. This paper has never been published.
 Appendix I 301

of an elastic plate that is held at its borders or to the motion of a rectangular

elastic plate held fixed along its four sides.* Those equations that I have
obtained have been presented in a form such that it will be easy to recognize
those cases in which the motion becomes periodic in the nature of the sounds
that correspond to motions ofthis type. The method by which I obtained this
integral can be applied to a large number of questions ofthis type, and for this
reason merits some attention.

A. L. Cauchy

A New Study on the Calculus of Residues and Definite Integrals

(Nouveau memoire sur Ie calcul des residus
et les integrales definies)l1
I am honored to present to the Academie certain new applications of the
calculus of residues along with some new formulas having to do with the
determination and the transformation of certain definite integrals. These
formulas complete the method that I proposed in a study in 1814 and in which
I presented the evaluation of certain definite integrals as depending on the
theory of singular integrals, that is, integrals taken between limits but in which
the function under the Jsign becomes either indeterminate or infinite. In the
1814 study, I was restricted to the consideration of the case in which the terms
divided by the function under the Jsign gives, as a quotient, an expression that
becomes infinitely small in the first order when an infinitely small increase of
order one is attributed to the variable. In the other studies that I presented
during this period, I considered the case in which the quotient in question is an
infinitely small quantity of order greater than unity, but is represented by an
integer. In my new investigation, I examined the case in which this order is
replaced by a fraction, and I exhibited the formulas that are to replace those
that I obtained earlier. The new formulas, of course, include as special cases
certain results that are already known, as well as a multitude of other formulas
that would appear to be noteworthy. Aside from this, in order to recapture the
previously known formulas, it suffices to substitute integers diminished by
very small quantities (which are soon to be reduced to zero) for the fractions I
referred to earlier.
[Paris, February 7, 1825
A. L. Cauchy]12

*These two motions have already been considered in the last study by M. Brisson. But his
solutions are incomplete, even with regard to elastic plates, considering that the integrals that he
gave or indicated also contain arbitrary functions [note by Cauchy].
11 Note of presentation read before the Academie on February 14, 1825. The manuscript is in the
meeting's packet of that day. The study has never been published.
12The words in brackets are cancelled by Cauchy.
302 Appendix I

The last paragraph of the new study is devoted to the investigation of the
number and the nature ofthe values that can be assumed by a definite integral
taken between imaginary limits. It is known that M. Laplace has drawn some
rather curious results from the consideration of such integrals. M. Ost-
rogradsky, a young Russian, who is very talented and wise, as well as very
learned in infinitesimal analysis, had deduced from there one of the general
formulas that we published in Journal de I'Ecole Poly technique. Finally, M.
Brisson has declared to us that in a work that he is presently engaged in he has
used these integrals for the summation of periodic series and for those whose
different terms include exponentials obeying a known law.
Nevertheless, I do not know that if at the present anyone has determined in
a sufficiently precise way the sense that ought to be attached to integrals taken
between imaginary limits and the different values that they may assume. Such,
then, is the question that I now propose to resolve in the fourth section of my
new study. The solution is derived quite easily from the calculus of variations
combined with the theory of singular integrals. Aside from this, these questions
led me to certain new formulas, which include those that I gave in my earlier
investigations and can be quite fruitfully applied, either to the determination
or the transformation of definite integrals taken between real limits.
Paris, February 15, 1825
A. L. Cauchy

A Study on the Equilibrium and the Motion of Fluids (Memoire

sur l'equilibre et Ie mouvement des fluides)13
On finishing the note that I read at the last meeting of the Academie,14 I
observed that the theory of fluids can be deduced from the principles
established in Exercices de M athematiques. I am now going to give a few
expansions on this subject.
When one considers a system of molecules that attract or repel each other
and if one supposes that the interior pressures reduce to zero in a natural state,
the equations of equilibrium or of motion of this system include fifteen
coefficients that depend on the way in which the molecules are distributed
about each point. The same equations include six coefficients more, if in a
natural state the interior pressures are assumed to be nonzero. These various
coefficients will be represented by certain triple sums with finite differences,
and if the total action exerted on a given molecule by those around it can be
regarded as essentially independent of the more neighboring molecules, then
the sums in question can be transformed, with no trouble, into simple integrals
of infinitesimally small differences.

13Studypresented to the Academie on May 4, 1829, and read on May 25,1829. The manuscript is
in the meeting's packet of May 4, 1829.
14 Addition to the study 'Sur la dilatation et la condensation lineaire des corps solides ou fluides, et
sur i'equilibre ou Ie mouvement des fluides', read on April 27, 1829.
 Appendix II
Documents on Cauchy's Analysis Course
at the Ecole Poly technique (1816-1819)

First-Year Program of the Analysis Course by Augustin

Cauchy (November 1816)1
Algebraic Analysis

General topics. Review of the theory of exponential, logarithmic, and

circular quantities and the most useful of the trigonometric formulas.
Imaginary expressions. Proof of the formula

(coscp ± ~ sin CPt =cosncp ± ~ sinncp).

Some notions on functions in general; the distinction between continuous

and discontinuous functions; the distinction between entire functions
and fractional functions, rational or irrational, simple or composite,
explicit or implicit.
Decomposition of the first member of an equation into real factors of the
second degree. Solutions of equations of degree three and four by
algebraic methods or by means of tables of sines; solutions of binomial
equations of the form xm - 1 = O.
Theorem of Cotes
Decomposition of rational fractions.
Degree of the final equation that results from an elimination performed on
several algebraic equations.
[To determine the functions that verify the equations f(x + y) = fxfy,
f(x + y) = fx + fy, etc.] (cancelled by Cauchy.)

1 Archives Ac. Sci., F6nds Ampere, box 4, chapter 4, file 75, The manuscript, which was written in
Cauchy's hand, is undated, but examination of the records of the deliberations of the Conseils and
of an abridged version of this project, which is kept in the Archives suggest a date of November
1816 with great certitude.

303
304 Appendix II

Expression of integral powers of sines and cosines by series of sines and cosines
of multiple arcs.
Rules on the convergence of series.
Expansion of certain functions III a series by means of the method of
undetermined coefficients.
Theory of recurrent series.
Proof based on a series for the formula
e"'P = cos ¢ + J=1 sin ¢.
Interpolation formula obtained by the method of undetermined coefficients.
Calculus of (Finite) Differences
General topics. Finite differences and integrals of the first order. Finite
differences and integrals of different orders.
Analogies of powers and differences. The most simple notions on solving
certain finite difference equations.
On the values that correspond to certain integral indices and on the number of
arbitrary constants that appear in these same values.
Interpolation formula.
Differential and Integral Calculus
Basic principles.
Differentials of the first order of functions and integrals.
Differentials of the first order of functions of functions and of composite
functions. Formulas for the integration of these same functions and
particularly of rational fractionals; differentials affected by a second-order
radical; differentials of binominals and of radicals that contain exponential,
logarithmic, or circular functions of a single variable. Changes of the
independent variable.
Differentials of the first and higher orders for functions of several variables.
The conditions for integrability and the integration of the same differentials
when they satisfy these conditions.
Taylor's theorem. Theory of minima and maxima for functions of a single
variable.
Values of fractions that appear in an indeterminate form.
Taylor's theorem extended to functions of several variables. Maxima and
minima of these classes of functions. Theorem on homogeneous functions.
Differentials of the first and higher orders for implicit functions.
Elimination of constants between an equation and its differentials of different
orders. The number of arbitrary constants that appear in the integral
[solution] of a differential equation.
Determining the factor that renders a first-order differential equation
integrable.
Particular solution of first-order differential equations.
Complete integration of the equation y - PX = !(p), where P = dy.
dx
 Appendix II 305

Integration of simultaneous differential equations, in particular, of those

equations that are linear with constant coefficients.
The most simple analytical notions on the integral calculus of partial differ-
ences. Integration of the first-order equation or of linear equations with
constant coefficients of arbitrary order.
Integration by series of differential formulas, certain differential equations of
different orders, and partial differential equations.

Applications of Integral Calculus to Geometry

Analytical formulas for the determination of tangents, normals, asymptotes,
etc.
Various expressions for the radius of curvature. Properties of values.
Applications to conic sections, cycloids, etc.
Geometrical considerations on the rectification of curves, the quadrature of
curves and of curved surfaces; the curvature of solids.
Analytical formulas for the solution of the preceding questions, including
formulas related to solids of revolution and those that have as their
boundaries arbitrary surfaces by application of interpolation formulas to
these same questions.

Official First-Year Program of the Analysis Course (1816)

In order to provide a deeper appreciation of the implications of Cauchy's
project for the first-year course, we will now present the official first-year
program, which was finally adopted in 1816.

Algebraic Analysis
Solution of algebraic equations of the 3rd and 4th degree. Demonstration of
how any equation of the 3rd or 4th degree can be solved by use of tables of
sines.
Series expansions of certain functions by means of the method of undeter-
mined coefficients.
General law of recurrent sequences as observed in the development of rational
fractions.
Special examination of recursive sequences depending on two terms, their
decomposition into two geometric progressions, and their general terms.
Review of the most useful trigonometric formulas and of the exponential
equation that arises between a number and its logarithm. To derive the
general properties of logarithms from this equation. To compare the
different systems and how to 'pass' from one system to another. To establish
the formulas
(cosq> + F1 sinq»m = cosmq> + F1 sinmq>
and
e'PP = cos q> + j=1 sin q>
306 Appendix II

Use of the first formula in obtaining the roots of the equations xm - 1 = 0,

xm + 1 = 0, which leads to the theorem of Cotes.
Use of the same formulas to express the powers of sines and cosines by series of
sines and cosines of multiple arc.
Some notions on functions in general and on their classification as entire,
rational, etc.

Differential Calculus
To exhibit the principles underlying differential calculus by consideration of
infinitesimals. To show, in the most simple cases, the agreement between this
method and methods based on limits and series expansions.
To determine the differentials of xm,xy, and ~ after which it is easy to
y
determine the derivative of any algebraic function, whether of one or several
variables, implicit or explicit.
Differentials of circular, logarithmic, and exponential functions, both simple
and compound.
Second and third derivatives.
Proof of Taylor's theorem.
Proof of the binominal formula for the case of negative or fractional
exponents.
Theory of equal roots using differential calculus.
Application of Taylor's theorem to series expansions, which determine
logarithms, exponentials, sines, and cosines, as arc functions, and
conversely.
The same theorem extended to two variables.
Notions on partial derivatives.
Theory of the maxima and minima of functions of one or two variables.
Ways of distinguishing maxima and minima.
Applications to some selected examples.
Formulas for sub tangents, subnormals, tangents, etc. The determination of
asymptotes.
Expression for the radius of curvature.
General properties of the involute and the method of finding its equation.
Application to conic sections, the cycloid, etc., to determine the involute of the
parabola and its rectification.
To change a function of a differential equation of the second order, in which a
first derivative has been assumed to be constant, into one in which no
differential is assumed constant.
To show succinctly that, if in a fraction ~, the two terms vanish when x = 0(,

then it is equal to ~~. This is almost always sufficient to determine the

value [of the fraction].
 Appendix II 307

Integral Calculus
General notions on integration.
Integration of monomial differentials and entire functions.
Methods of making rational differentials that are of the form
Ja +hX+CX2.
Integrability of binominals.
To exhibit the basic formulas to which other formulas are related and to show
how these integrals may be expressed.
Formulas on the quadrature of curves, rectification, quadrature of surfaces,
and the curvature of solids of revolution. Emphasis on applications and the
determination of constants.
Cauchy's analysis course, first-year, 1816-1817. 2

Lecture
Date number Subject of the lessons

Thursday, January 16 Point de leyon. L'installation de l'Ecole n'ayant

eu lieu que Ie 17
Saturday, January 18 Considerations generales sur les nombres et les
quantites, les variables et les fonctions.
Monday, January 20 2 Suite des considerations generales sur les
fonctions, somme et difference. La somme de
plusieurs variables a pour limite la somme de
leurs limites.
Tuesday, January 21 Point de leyon a cause de l'anniversaire de la
mort de Louis XVI (cette leyon a eu lieu
Lundi de 1h a 2h t).
Thursday, January 23 3 Suite des considerations generales. Produits et
quotiens. Lorsque B, B' ... sont des quantites
A+A' ...
de meme signe, la fraction est
B+B' ...
comprise entre la plus petite et la plus
A B
grande des fractions - , -... Le produit de
A' B'
plusieurs variables a pour limite Ie produit
des limites de ces memes variables.
Saturday, January 25 4 Revue de la theorie des quantites
exponentielles.
Tuesday, January 28 5 Suite des matieres traitees dans la leyon
precedente. a, a', a" ... b, b', b" ... designant
des nombres quelconques,
b+b'+b"+ "'J a + a' + a" + ... est comprise entre
la plus petite et la plus grande des racines
(Continued)
2Archives E. P., X II C7, Registre d'instruction 1816-1817.
308 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Ja,
b b'jd, b".jd' ... Lorsque les variables
x et y ont respectivement pour limites X et
Y, x Y a pour limite X Y • Distinction entre les
puissances arithmetiques des nombres et les
puissances algebriques des quantites.
Thursday, January 30 6 Application des principes etablis dans les
lecons precedentes. Limite du deveioppement
du binome.
Saturday, February 1 7 Developpment du produit des differences
a-b, a-c, a-d ... , b-c, b-d ... , c-d,
etc. Comment on peut reconnaitre les signes
des differents termes. Transformation des
exposants en indices.
Tuesday, February 4 8 Application des formules trouvees dans la lecon
precedente aux probh~mes de l'interpolation.
Thursday, February 6 9 Revue de la theorie des logarithmes.
Saturday, February 8 10 Revue des fonctions circulaires.
Tuesday, February 11 11 Continuation de meme sujet. Dans un triangle
Ie rapport d'un cote au sinus de l'angle
oppose est Ie diametre du cerc1e circonscrit.
Thursday, February 13 12 Relations diverses entre les !ignes
trigonometriques des trois angles d'un
triangle. Lorsque les cotes sont representes
par les sinus, les distances des sommets au
point d'intersection des trois perpendiculaires
sont representees par 1e4l cosinus.
Saturday, February 15 13 Exprimer dans un triangle les angles, la surface
et Ie rayon du cerc1e circonscrit au moyen
des trois cotes. Resolution des triangles
rectilignes en general.
Tuesday, February 18 14 On a etabli la formule
(cos a + j=-1 sin a) (cos b + j=-1 sin b)
= cos (a + b) + j=-1 sin (a + b).
Application de cette formule a la
determination de sin na et cos na exprimes
en fonction de sin a et cos a.
Thursday, February 20 15 Exprimer les puissances du sinus et du cosinus
au moyen des sinus et cosinus des arcs
multiples.
Saturday, February 22 16 Des diverses valeurs de l'expression
(f + g ~)!!J (m etant premier a n) Si ron
fait f = r cos Q, g = r sin Q, r etant positif,
on aura=(f+g~)~=
 Appendix II 309

(Continued)

Date Lecture Subject of the lessons

number

f ( cos~
mQ + F1 mQ) (l)n
-1 sin~ I

I 2Kn
(l)n = cos--±
J=1 2Kn
-1 sin--
n n
n
K etant = ou <-.
2
Tuesday, February 25 17 Quotient de deux expressions imaginaires.
Puissances negatives de ces memes
expressions.
Thursday, february 27 18 Puissances irrationnelles des expressions
imaginaires. On les deduit des puissances
irrationnelles de I'unite. Ces dernieres sont en
nombre indefini. Si b etant irrationnel, on
suppose (l)b = cosx + ~ sinx,x pourra
recevoir a peu pres dans cette equation
to utes les valeurs possibles.
Saturday, March 1 19 Classement des diverses fonctions d'une seule
variable. Distinction des fonctions continues
et discontinues. Des valeurs des fonctions
d'une seule variable dans quelques cas
particuliers.
Tuesday, March 4 20 Resolution generale de l'equation f(u) - b = 0,
f(x) etant une fonction continue de x,
lorsque l'on connait deux valeurs de x qui,
substituees dans Ie premier membre, donnent
des resultats de signes contraires.
Classement des fonctions de plusieurs variables.
La limite d'une fonction continue de
plusieurs variables est la meme fonction de
leurs limites. Consequence de ce theoreme
relativement a la continuite des fonctions
composees qui ne dependent que d'une seule
variable.
Thursday, March 6 21 Toute equation qui n'a pas de racines reelles a
necessairement des racines imaginaires de la
forme r(cos Q + ~ sin Q). Demonstration
des differents lemmes qui conduisent a cette
proposition.
Saturday, March 8 22 Continuation du meme sujet Decomposition
des polyn6mes x" + 1, Xl" - 2a" x" cos Q + al "
en facteurs reels.
(Continued)
310 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, March 11 23 Theofli:me de Moivre et de Cotes. Resolution

generale de l'equation du 3e degre.
Thursday, March 13 24 Resolution generale de I'equation du 4e degre.
Discussion des racines.
Saturday March 15 25 Dans une equation quelconque, Ie produit des
quarres des differences entre les racines est
positif ou negatif suivant que les couples des
racines imaginaires sont en nombre pair ou
en nombre impair. Determiner les fonctions
qui satisfont a des conditions donnees.
Tuesday, March 18 26 Suite de la le<;on precedente.
Thursday, March 20 27 Regles sur la convergence des series.
Saturday, March 22 28 Suite de la le<;on precedente.
Tuesday, March 25 29 Developpemens en series des exponentielles, des
sinus et des cosinus. Les series obtenues sont
convergentes pour toutes les valeurs
possibles de la variable. Des series
imaginaires convergentes.
Thursday, March 27 30 On a etabli au moyen des series la formule
e'R = cos r + vi=l sin r. Definition des
expressions cos (xvi=l), sin (xvi=l),
(f + gj=ly+ yJ=\, log(u + v./=l).
Saturday, March 29 31 Developpemens en series des fonctions (1 + x)~,
(1 +xcos<p+xsin<p)~, log(l +x),
log(l + 2x cos <p + x 2 ), arc(tang'x),

arc (tang x sin cp )

1 + x cos cp
Decomposition des fractions rationnelles.
Tuesday, April 1 32 Theorie des series recurrentes. Definition du
coefficient differentiel.
Thursday, April 3 33 Principes fondamentaux du cal cui differentiel et
integral.
Saturday, April 5 Point de le<;on (Samedi Saint).
Tuesday, April 8 34 Des integrales definies, considerees comme des
sommes d'elements.
Thursday, April 10 35 Differentielles des fonctions simples, des
fonctions de fonctions et des fonctions
composees.
Saturday, April 12 36 Des integrales definies. Diverses methodes
d'integration.
 Appendix II 311

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, April 15 37 DitTerentielles de divers ordres. Changement de

variable independante.
Thursday, April 17 38 DitTerentiation et integration sous Ie signe J
Integrales definies. Des integrales prises entre
des limites infinies, et de celles OU la fonction
J
so us Ie signe devient infinie entre les limites
donnees. Decider si une integrale prise entre
des limites donnees est finie ou infinie.
Saturday, April 19 39 Suite des matieres precedentes.
Tuesday, April 22 40 Idem.
Thursday, April 24 41 Integration par series. Theoreme de Taylor.
Maxima et minima pour les fonctions d'une
seule variable. Valeurs des fractions qui se
presentent sous une forme indeterminee.
Saturday, April 26 42 DitTerentielles des fonctions de plusieurs
variables. Conditions d'integrabilite et
integration de ces memes ditTerentielles.
Tuesday, April 29 43 Theoreme de Taylor etendu aux fonctions de
plusieurs variables.
Thursday, May 1 44 Maxima et minima des fonctions de plusieurs
variables. Theoreme des fonctions
homogenes.
Saturday, May 3 45 Revue generale du calcul infinitesimal
Tuesday, May 6 46 Idem.
Thursday, May 8 47 Idem.
Saturday, May 10 48 DitTerentielles des fonctions implicites. Theorie
des tangentes.
Tuesday, May 13 49 Theorie des contacts de divers ordres.
Application au cerc1e.
Thursday, May 15 Point de lec;:on. Fete de I'Ascension.
Saturday, May 17 50 Sur Ie rayon de courbure generale. Application
aux courbes du second degre. Theorie des
developpees. Application a l'ellipse.
Tuesday, May 20 51 Developpee de la para bole. Discussion des
paraboles des divers ordres. Points de
rebroussement de ler et de 2e especes.
Cyc101de.
Thursday, May 22 52 Deveioppee de la cyc101de. Logarithmique.
Asymptote. Points multiples et points isoles.
Coordonnees polaires des spirales en general
et en particullier de la spirale logarithmique.
(Continued)
312 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Saturday, May 24 53 Contact des divers ordres entre les surfaces

courbes. Equation du plan tangent.
Tuesday, May 27 54 Sur les rayons de plus ou moins grande
courbure d'une surface courbe.
Considerations geometriques sur la
quadrature des surfaces courbes et la
cubature des solides.
Thursday, May 29 55 Formules analytiques pour les quadratures et
les cubatures.
Saturday, May 31 56 Rectification des courbes et quadratures des
surfaces courbes.

Cauchy's analysis course, second-year 1817-1818. 3

Date Lecture Subject of the lessons

number
Tuesday, December 2 1 Considerations generales sur les integrales

{-oo} Q
definies.
Thursday, December 4 2 Valeur de l'integrale S-P dx , -P etant
Q +00

une fraction rationnelle. Application au cas

P x2m
ou l'on suppose - = - - 2 - .
Q 1 +x m

Saturday, December 6 3 Valeur de l'integrale

l+z
dz_
Sz a - I _ {O} des
00
,

dz
SZ·-I_-b • Valeurs de r(n) et de
(+ z)
r(n + !), n etant un nombre entier
que1conque.

3Archives E. P., XII C7, Registre d'instruction, 1817-1818.

 Appendix II 313

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, December 9 4 Passage du reel a l'imaginaire. Valeurs des

integrales Jx·- l e- ax cos bx dx,
J x·- l e- ax sin bx dx, entre les limites
x=O, X= 00.
Thursday, December 11 5 Suite de la le<;on precedente. Valeur de

J
l'integrale e - x 2 cos 2 bx dx { : : }.

Saturday, December 13 6 Usage de l'equation dp/dy = dq/dx pour

determiner les relations qui existent entre
certaines integrales definies. Pour trouver
des fonctions p et q qui satisfassent a
l'equation precedente, il suffit de poser
J(x + y J=1)
= p +q J=1.
Sommation
t
des series 1 + t + + /6' etc. Etablir les

J
formules sin bx dx { 0 } =~,
x 00 2

e-X 4::b
2
dx {~} = tnt e- b.
Tuesday, December 16 7 Usage des integrales definies pour la
sommation des series convergentes.
Thursday, December 18 8 Sur les fonctions reciproques de la premiere et
de seconde especes.
Saturday, December 20 9 Sommation des series convergentes par Ie
moyen des fonctions reciproques. Limite du
produit {1 + (Xl - xo)F(xo)}
{1 + (X2 - xl)F(xd} ...
{1 + (x. - x.-l)F(x.- l )}, etc.,
tandis que les elemens de la difference
X - Xo decroissent indefiniment.
Tuesday, December 23 10 Considerations generales sur les equations
differentielles. Integration des equations
differentielles du premier ordre
dy = F(x) dx, dy = yF(x) dx.
Thursday, December 25 Point de le<;on. Fete de Noel.
Saturday, December 27 11 L'integrale d'une equation differentielle du
premier ordre renferme une con stante.
Methode pour obtenir par approximation
une integrale particuliere correspond ante a
une valeur determinee de la constante.
( Continued)
314 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, December 30 12 Suite de la le90n precedente.

Thurday, January 1 Point de le90n.
Saturday, January 3 13 Sur la distinction des integrales generales, des
integrales particulieres et des solutions
particulieres.
Tuesday, January 6 14 Du facteur propre a rendre une equation
integrable. On peut trouver autant de
facteurs que l'on veut propres a remplir
cette condition. Relation entre ces facteurs.
Thursday, January 8 15 Separation des variables. Integration de
l'equation lineaire du premier ordre et de
l'equation homogene.
Saturday, January 10 16 Suite de l'integration des equations
homogenes. Solution particuliere de ces
equations. Application a l'equation du
premier degre.
Tuesday, January 13 17 Integrale generale et solution particuliere de
l'equation y = f(y') + xf(y') dans laquelle
dy Integratton
y, =~. · . de I'·equatIOn
.
dx
dx dy
+ Q
J(l + IXX 2 + f3x 4) J(l + lXy2 + f3 y4)
Thursday, January 15 18 Toutes les solutions particulieres de I'equation
differentielle dy - pdx = 0, verifient la
1
formule ~~- = O. Mais la reciproque
(dp/dy)
n'est pas vraie.
Saturday, January 17 19 Methode generale pour distinguer les
solutions particulieres des integrales
particulieres.
Tuesday, January 20 20 Considerations generales sur I'integration des
equations differentielles simultanees du
premier ordre.
Thursday, January 22 21 Suite de la le(fon prec6dente.
Saturday, January 24 22 Les integrales generales de plusieurs
equations differentielles simultanees du
premier ordre renferment autant de
constantes arbitraires qu'il y a d'equations.
Methode pour obtenir par approximation
les integrales particulieres correspondantes
 Appendix II 315

(Continued)

Date Lecture Subject of the lessons

number

a des valeurs determinees de ces memes

constantes.
Tuesday, January 27 23 Sur la distinction des integrales generales des
integrales particulieres et des solutions
particulieres dans les equations
differentielles simultanees du premier ordre.
Methode pour la determination de to utes
les solutions particulieres de ces memes
equations.
Thursday, January 29 24 Determination des integrales generales et des
solutions particulieres de quelques
equations differentielles simultanees du
premier ordre. Integration des equations
lineaires simultanees.
Saturday, January 31 25 Suite de la le<;on precedente. Cas des racines
egales et des racines imaginaires.
Tuesday, February 3 26 Considerations generales sur l'integration des
equations differentielles d'un ordre
quelconque entre deux variables.
Thursday, February 5 27 Integration de l'equation lineaire a coefficients
constants avec un dernier terme variable.
Saturday, February 7 28 Theoremes relatifs a l'integration des
equations lineaires.
Tuesday, February 10 29 Suite des theoremes relatifs a I'integration des
equations lineaires.
d2 z
Thursday, Februrary 12 30 Integrale generale de l'equation lineaire - 2 =
dx
axmz. On en deduit l'integrale generale
de l'equation de Riccati par une integrale
definie.
Saturday, February 14 31 Ordre de l'equation finale obtenue par
I'elimination entre plusieurs equations
simultanees. Integration par series des
equations differentielles.
Tuesday, February 17 32 Integration des equations aux differences
partielles du premier ordre.
Thursday, February 19 33 Suite.
Saturday, February 21 34 Application des theoremes exposes dans la
seance precedente. Observation sur
l'integration des equations aux differences
partielles d'un ordre quelconque.

(Continued)
316 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, February 24 35 Suite.

Thursday, February 26 36 Integration des equations lineaires aux
differences partielles a coefficiens cons tans
sans dernier terme variable.
Saturday, February 28 37 Calcul direct aux differences finies.
Tuesday, March 3 38 Calcul inverse aux differences finies.
Thursday, March 5 39 Application a divers exemples et a la
sommation des series.

Cauchy's analysis course, first-year, 1818-1819. 4

Date Lecture Subject of the lessons

number

Tuesday, November 3 Revue des diverses especes de quantites. Sur les

moyennes entre plusieurs quantites.
Thursday, November 5 2 Suite des theoremes sur les moyennes.
Considerations generales sur les fonctions,
etc.
Saturday, November 6 3 Des fonctions continues et discontinues.
Tuesday, November 10 4 Valeurs singulieres des fonctions dans quelques
cas particuliers. Representation geometrique
des fonctions continues par des courbes.
Equation de la logarithmique.
Thursday, November 12 5 Discussion des courbes. Determiner leur
inclinaison en un point quelconque.
Saturday, November 14 6 Suite de la le90n precedente. Des courbes
representees par les equations y = x+ a,
y = sin x et y = cos x.
Tuesday, November 17 7 Suite de la discussion des courbes qui ont pour
1
equations y = x a, y = sinx,y = lx, y = sin-.
x
Thursday, November 19 8 Lorque i'on considere une fonction de plusieurs
variables x,y,z, continue par rapport a
chacune d'elles, si ces variables convergent
vers les valeurs X, Y,Z, ... ,f(x,y,z, . .. )
convergera vers la limite f(X, Y, Z, . ..). De

4 Archives E. P., XII C7, Registre d'instruction, 1818-1819.

 Appendix II 317

(Continued)

Date Lecture Subject of the lessons

number

la continuite des fonctions composees d'une

seule variable. Fonctions symetriques. Usage
de ces fonctions pour la resolution des
equations lineaires.
Saturday, November 21 9 Des fonctions alternees et de leur usage pour la
resolution des equations lineaires.
Tuesday, November 24 10 Interpolation.
Thursday, November 26 11 Application des principes etablis dans la leyon
precedente.
Saturday, November 28 12 Determiner les fonctions Q(x) qui satisfont a
l'une des equations Q(x + y) = Q(x) + Q(y),
Q(x + y) = Q(x)Q(y), etc.
Tuesday, December 1 13 Determiner la fonction Q(x) qui satisfait a
l'equation Q(x + y) + Q(x - y) = 2Q(x)Q(y).
Thursday, December 3 14 Sur les series convergentes. Valeur de
1 1
e= 1 +-+-+ ...
1 1.2
Saturday, December 5 15 RegIe sur la convergence des series.
Tuesday, December 8 16 Developpement du binome dans Ie cas d'un
exposant queleonque.
Thursday, December 10 17 Developpement des exponentielles et des
logarithmes.
Saturday, December 12 Point de leyon (M. Cauchy indispose).
Tuesday, December 15 18 Notions sur les maxima et minima des
fonctions entieres.
Thursday, December 17 19 Suite.
Saturday, December 19 20 Fin de la theorie des maxima et minima des
fonctions entieres
Tuesday, December 22 21 Sur les expressions imaginaires. Exprimer les
sinus et les cosinus d'arcs multiples en
fonction des puissances des sinus et cosinus
d'arcs simples, et reciproquement.
Thursday, December 24 22 Tout polynome est decomposable en facteurs
reels du second degre.
Saturday, December 26 23 Suite de la leyon precedente.
Tuesday, December 29 24 Resolution des equations binomes.
Thursday, December 31 25 Resolution des equations trinomes.
Saturday, January 2 26 Resolution des equations du troisieme degre.
Tuesday, January 5 27 Resolution des equations du quatrieme degre.
Decomposition des fractions rationnelles en
fractions simples dans Ie cas des racines
reelles.
(Continued)
318 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Thursday, January 7 28 Decomposition des fractions rationnelles, dans

Ie cas des racines imaginaires et des racines
egales.
Saturday, January 9 29 Series imaginaires.
Tuesday, January 12 30 Des fonctions imaginaires, exponentielles,
logarithmiques et circulaires.
Thursday, January t4 31 Elevation des expressions imaginaires Ii des
puissances quelconques.
Saturday, January 16 32 Developpemens en series entieres des fonctions
imaginaires {1 + x(cos 0 + j=t sin O)}I',
log {1- x(cosO + j=t sinO)} et des
fonctions reelles 1 + 2x cos 0 + x 2 ,
xsinO
arctang .
1 + 2x cos 0 + x 2

Tuesday, January 19 33 Suites recurrentes. Principes du ca1cul

infinitesimal.
Thursday, January 21 Point de lecon (anniversaire de la mort de
Louis XVI).
Saturday, January 23 Point de lecon (M. Cauchy (:tant indispose).
Tuesday, January 26 34 Differentielles des fonctions simples et des
fonctions de fonctions.
Thursday, January 28 35 Suite de la lecon precedente.
Saturday, January 30 36 Differentielles des fonctions de plusieurs
variables independantes et des fonctions
composees d'une seule variable.
Tuesday, February 2 37 Differentielles des series convergentes et des
fonctions imaginaires. Notions sur les
integrales definies.
Thursday, February 4 38 Proprietes des integrales definies. Integration
de l'equation dy = f(x) dx.
Saturday, February 6 39 Valeurs d'integrales definies et indefinies.
Tuesday, February 9 40 Examen des diverses methodes d'integration.
Thursday, February 11 41 Emploi des expressions imaginaires dans
l'integration indefinie.
Saturday, February 13 42 Revue des formules differentielles auxquelles les
methodes d'integration exposees dans les
seances precedentes sont applicables.
Tuesday, February 16 43 Idem.
Thursday, February 18 44 Formules de reduction pour les differentielles
binomes.
Saturday, February 20 45 Suite.
 Appendix II 319

(Continued)

Date Lecture Subject of the lessons

number

Tuesday, February 23 46 Differentielles de divers ordres des fonctions

d'une seule variable independante, et
integrales successives.
Thursday, February 25 47 Changement de la variable independante.
Differentielles des fonctions implicites.
Saturday, February 27 48 Differentielles totales de divers ordres pour les
fonctions de plusieurs variables
independantes et les fonctions composees
d'une seule variable. Theon':me des fonctions
homogenes.
Tuesday, March 2 49 Differentiation et integration sous Ie signe J.
Conditions d'integrabilite pour la
differentiation des formules differentielles qui
renferment plusieurs variables independantes.
Integration de ces memes formules
lorsqu'elles satisfont a ces conditions.
Thursday, March 4 50 Theoreme de Taylor.
Saturday, March 6 51 Applications du theoreme de Taylor aux
maxima et minima des fonctions d'une ou de
plusieurs variables.
Tuesday, March 9 52 Valeur des fractions qui se presentent sous la
forme 0/0. Integration par series.
Thursday, March 11 53 Considerations generales sur les applications
du calcul differentiel a la geometrie.
Tangentes aux courbes, sous-tangentes,
normales et sous-normales.
Saturday, March 13 54 Contact des divers ordres. Les courbes se
traversent quand Ie contact est d'ordre pair.
Points d'inflexion. Cerc1es osculateurs et
rayons de courbure.
Tuesday, March 16 55 Maxima et minima par la geometrie. De la
courbe qui est produite par les intersections
successives d'un systeme de droites. Le centre
de courbure est Ie point de rencontre de
deux normales infiniment rapprochees.
Thursday, March 18 56 Differentielle de l'arc d'une courbe. Equations
des developpees. Description de la
developpante par Ie moyen d'un fil applique
sur la developpee.
Saturday, March 20 57 Proprietes de la cyc101de.
Tuesday, March 23 58 Asymptotes, points isoles, etc.
(Continued)
320 Appendix II

(Continued)

Date Lecture Subject of the lessons

number

Thursday, March 25 59 Coordonnees polaires. Determination des

tangentes, des rayons de courbure,
developpees, lorsque les equations des
courbes sont exprimees a l'aide des
coordonnees polaires. Des spirales, et en
particulier des spirales hyperbolique,
logarithmique, etc.
Saturday, March 27 60 Quadratures et cubatures.
Tuesday, March 30 61 Exemples de quadrature. Application de
l'interpolation aux quadratures.
Thursday, April 1 62 Sur les limites des integrales simples et doubles
qui representent les aires et les volumes. Du
plan tangent a une surface courbe.
Saturday, April 3 63 Rectification des courbes, et quadratures des
surfaces courbes.
 Appendix III

Selection of Letters from Cauchy to

Various Persons (1821-1857)

Letter from Cauchy to Libri, February 3, 1821, Biblioteca

Moreniana of Florence, Fonds Palagi, File 431, Insert 76:
Paris, February 3, 1821
Sir,
Several days ago, I received and read with great interest the study Sur la
Theorie des N ombres that you were kind enough to send me. 1 I particularly
noted the ingenius method by which you solve the equation
zn = axn + bxn - 1 + ... + px + q
over the integers. However, at the same time, it would seem to me that one
should be able to substitute for this method a general principle that supplies
the solution for a rather broad class of indeterminate equations. The
confidence that you have been kind enough to place in me is evidence that I
may here expand on this principle.

Suppose, then, that two variables x and z are related to a certain equation
f(x,z) = O. (1)
One attempts to find a function u of these variables that, as regards the given
relation, can never assume any numerical value less than a certailllimit U, as
long as the variables are positive. If the function u is rational and if, at
the same time, its coefficients are integers, then this function should, for
positive values of the variable, be equivalent (up to the sign) to one of
the integers between 0 and U. By successively letting it assume the values of
these numbers, one obtains new equations that determine one or several

1 The Academie received Libri's paper on January 22,1821, and Cauchy presented a verbal report
on the work at the following meeting (O.C., 2, 15, Annexes documentairs, p. 519).

321
322 Appendix III

systems of values for the variables x or z. Among these systems, those that
constitute the positive values yield all the possible solutions, over the
integers, of the given equation.

Let us consider, for example, that from the given equation one obtains
for the values z or Z2, or Z3, •.. , a function of x with rational coefficients
and a series ordered in terms of descending and negative powers of x. Then,
one can take for u a multiple ofthe difference between z or Z2, or Z3, ... , etc. and
the integral function that was just mentioned.

For the sake of greater clarity, I will present here the application of the
general method given above to the solution of some particular equations.

1st Problem: To find the positive values of x or of z that satisfy the equation

(2)

Solution: From Eq. (2), one obtains

1 1
z=x 2 ( 1 +-+-+-+-
X
1 1
x 2 x 3 X4
)t
and
2x3a b b .
z=x +-+-+-+2""
2 8 x x
a, , ... are mtegers.

For u, we can take a multiple of the expression

-3 + -a + -,
b
etc. = z - x --
2 X
8 X 2 x 2

and express u by
u = 2z - 2X2 - X.

°
One sees that u ought to remain less that the limit 2. Thus, u = or u = + 1.
Foru = 0, we have x = O,z = 0, and for u = 1, we have x = 3, z = 11. Such, then,
are the positive solutions of equation (2).

2nd Problem: To find the positive solutions of the equation

Z4 + 8z = X4 + 21x. (3)
Solution: Let us assume u = Z - x. The values of u are determined by the
equation
(u + X)4 + 3 = X4 + 21x,
 Appendix III 323

and its maximum for the formula ~: = 0, which reduces to

4(x + U)3 + 3 = 4x 3 + 21,

or, what is the same thing,

Then, always,
1
U < (9/2)' < 2.
This means that either u = °or u = 1; u = °gives x = 0, Z = 0, and u = 1 gives
x = 1, Z = 2.

3rd Problem: To find the positive solutions of

(Z2 - x)(z + x) = 5 + 4x. (4)
Solution: In this case, we can set u = Z2 - x. Then, u < 4. Since 5 + 4x is odd, u
cannot be even. Thus, U = 1 or U = 3. Moreover, the value u = 1 does not yield
any solutions. But, U = 3, Z = 1, Z = 2. The investigation of negative
solutions is the same as the one of positive solutions, only the signs of
the variables are to be changed. Consider also that in a large number of cases it
is possible to solve indeterminate equations in more than two variables by
principles similar to the ones that we have just discussed. I close by asking that
you be assured of my personal regards.

Augustin Cauchy

Letter from Cauchy to Libri, May 18, 1821, Biblioteca Moreniana

of Florence; Fonds Palagi, File 431, Insert 76
Paris, May 18 1821

I have received the letter that you were so kind as to write to me and with
which you sent me a new work on prime numbers. It will be a true pleasure to
make myself familiar with this study and to transmit it, as you desire, to the
Academie Royale des Sciences. Please be sure, I pray you, of my highest
regards.

Augustin Cauchy
324 Appendix III

Letter from Cauchy to Libri, March 28, 1828, Biblioteca Moreniana

of Florence, Fonds Palagi, File 431, Insert 76
Paris, March 28, 1828

My dear Count,

It has always been a special pleasure to answer your kind letters. But, my poor
health or the multitude of things that I must do always come up and do not
permit me to respond. Today, I have a few moments ofleisure, and I hasten to
take advantage of it in order to tell you that I have fulfilled the commission
with which you charged me. I was delighted to nominate you as correspondent
of the Academie des Sciences. 2 I also thank you for the works that you have
been so kind as to send me at various times, and I pray that in exchange you
will accept the first fourteen fascicles of my Exercices de Mathematiques along
with a study on the applications of the calculus of residues to questions in
physics. I will be very flattered if this letter should be of some interest to you.
M. De Bure, my father-in-law, will be responsible for sending it, as well as the
various booklets of Exercices, to you.

I pray you be assured, my dear Count, of my highest regards,

A. L. Cauchy

Letter from Cauchy to Libri, July 21, 1829, Biblioteca Moreniana

of Florence, Fonds Palagi, File 431, Insert 76
Paris, July 21, 1829

Sir,

I have received the interesting studies that you were so kind as to send me. I
will be delighted to read them, and I beg you to accept all my thanks in this
matter. I regret that the person who was responsible for sending you Exercices
for me has only sent you the first seventeen fascicles; and, in order to make up
for this, I have just addressed to M. Freddoni the installments up to no. 39, as
well as several studies that were printed separately. I hope that they will be
worthy of your interest.

The formulas that are contained in the 39th fascicule of Exercices and that had
to do with the torsions of rectangular elastic rods have quite recently been

2Libri was presented for a position as correspondent in September 1826 and December 1827.
However, he was not elected until December 31, 1832.
 Appendix III 325

confirmed by the experiments of one of our most able physicists. 3 The numbers
supplied by the theory and observation agree to an extraordinarily high
degree. I eagerly seize this occasion to pray that you will be assured of my
highest regards.
A. L. Cauchy

Letter from Cauchy to the president of the Academie des Sciences,

July 14, 1831, Arch. Ac. Sci., Cauchy File
Geneva, June 29, 1833

Mr. President,

I should like to be able to present to my colleagues the study that I now have
the honor of addressing to you. However, the precarious condition of my poor
health has again obliged me to extend my leave of absence for a while; and I
must ask you to be so kind as to offer the Academie des Sciences the work to
which I refer. This paper is an extract from a more extended study that has as
its aim the investigation of various analytical methods; this work has been
published in volume 60 ofthe Bibliotheque Italienne. By means of methods that
are analogous to those, I developed lectures on differential calculus and its
main analytical applications and also on the calculus of variations. Later, I will
be honored to offer the Academie a second study that will present a new
application of the calculus of residues. Please be assured anew, etc.
Your very humble and obedient servant,
A. L. Cauchy

Letter from Cauchy to the president of the Academie des Sciences,

March 25, 1833; Bib!. lnst. M663 H*
Mr. President,

Permit me to ask you to be kind enough to offer the Academie, on my behalf,

the second part of my lithographed memoir. I hope that geometers will not
consider as unimportant, in this second part, the applications of this new
calculus, which I call the calculus of limits, to celestial mechanics, because it
serves to determine the limits of the errors in the series expansions of implicit
or explicit functions of one or several variables.

In the addition to this work, and in subsequent works that I will have the

3 The reference is to Savart.

326 Appendix III

honor of addressing to the Academie, I will present some new developments on

this topic and will show how, by means of this new calculus, it is possible to
determine the errors in the series expansions of functions that represent the
integrals of a system of differential equations, such as, for example, those that
give rise to the theory of planetary motion. Be assured, I beg you, of my highest
regards, etc.

A. L. Cauchy

Letter from Cauchy to Comte de l'Escarene, July 18, 1833, Musee

National de l'Education, Rouen, Ms. A 10822-1
My dear Count,

In agreement with your Excellency's views, I have told M. de Collegno of the

reasons that obliged me to hasten my departure. Many obstacles arose: on
Saturday, I conclude my last examination, and I am booked for the Geneva
post for Monday. But, an incident came up that obliged me to let the very
excellent M. d'Olry into our secret. I must cross Switzerland and Bavaria, and
I am carrying some mathematics books with me. It is very important that I do
not find myself forced to stop before crossing into Bavarian territory. This
compelled me to ask your Excellency to tell M. d'Olry of the position in which
I now find myself and to send him a few lines that I include in this letter. Also,
please accept assurances of the respectful and inalterable devotion of

Your very humble and very obedient servant,

A. L. Cauchy

Turin, July 18, 1838

I beg your Excellency be kind enough to send to M. d' Olry my passport, which
is included herein. I should like to have your response as soon as possible.

Copy of the official letter from Baron de Damas:

Toeplitz, June 22, 1833

Sir,

You could be very helpful in educating the Duc de Bordeaux, and the King has
asked me to write to you. His Majesty would like, if it is possible, that you
immediately begin to work with my pupil; you will be responsible for his
instruction in the sciences, a task that has heretofore been the responsibility of
M. Barande.
 Appendix III 327

M. Barande was removed for reasons that have greatly affected us; but, it did
not compromise either his honor or his considerateness. lowe you this
explanation because I do not want to give grounds for any unjust suspicions
that might come up at this time.

Please be assured of my steadfast attachment and special consideration.

Baron de Damas

A second letter, filled with expressions of goodwill for me, concludes as follows:
I do not know what your personal position is, but I do know your sentiments.
The heir of our King needs your services, and it is my duty to request them of
you.

Sir, please be assured of the deep attachment that I swore to you in times better
than these and that will not change.

Baron de Damas

P. S. The King desires that you should travel first to Prague and from there to
the countryside where his Majesty resides. As soon as possible, please write to
me telling the route you will be taking and the prospective date of your arrival.

The King will leave Toeplitz during the first days of July to return to
Buschtiehrad near Prague. His Majesty has good reason to be pleased with his
stay here: the waters do him good.

A. L. Cauchy
Professor of Sublime Physics at the University of Turin

Letter from Cauchy to Count de l'Escarene, September 24, 1834,

Musee National de l'Education, Rouen, Ms. A 10822~2
Buschtiehrad near Prague, September 24, 1833

Moved by the recognition of your kindness, as well as that of the King of

Sardinia, I impatiently await your letter at the very moment when the question
that was raised here several weeks ago shall have been completely solved.
Unfortunately, we are still operating on a provisional basis, so that it is
impossible to say definitively whether I will remain here or whether I will
return to Turin to teach the courses that his Majesty has entrusted to me. Be
all that as it may, the newspapers spread many errors and falsehoods about us.
The young heir of our King bears himself in a magnificent way, and his
sentiments, which are well worthy of a son of Saint-Louis, are those that you
328 Appendix III

will find expressed in a brochure which was published in Prague and of which I
am enclosing three copies. I beg you to be kind enough to present the first copy
to His Majesty the King of Sardinia as a testimonial of my gratitude, my
devotion, and my respect; the second is to be presented to Count de la Tour;
the third copy is for you, as is a new assurance of the respectful devotion that I
feel toward Your Excellency.

Your very humble and obedient servant,

A. L. Cauchy
P. S. Please give my respects to Madame Countess de l'Escarene and also
remember me to persons who have been kind enough to take an interest in me,
particularly M. Ie Chevalier d'Olry and R. P. Grassi. If I have not returned to
Turin by November 1, it would seem to me that R. P. Grassi could make very
fine suggestions to M. de Collegno as to the selection of a teacher for the
courses or for part of the courses that I was to teach. Both Father Moigno and
Father Lachaise would be excellent candidates. Father Moigno is perfectly
well acquainted with all areas of mathematical physics. M. d'Olry will receive
from Switzerland either one or several copies of the first edition of my
brochure.

Letter from Cauchy to Moigno written on June 12, 1837,

Notebook in the Sorbonne Library, Ms 1759 bis
June 12, 1837

My dear Abbe,

I only received your letter of May 17 this morning, and I am taking advantage
of the few remaining minutes to reply to it. Thus, I will not have time today to
examine the objections that MM. Sturm and Liouville have raised against the
theorem that concludes my 1833 study,4 as to whether or not it will be
necessary to make any changes or modifications. However, I want to look into
this closely when I have the time. Far be it from me to think that I am infallible,
but what really grieves me is that MM. Sturm and Liouville should imagine
that I would try to claim the glory for proofs that they published earlier. It
should be remembered that in my study on the theory of waves, I found and
presented a theorem that M. Fourier had given earlier in a work that had not
been published and the existence of which I was completely unaware;
but when I later found out about it, I hurried to insert in the Bulletin de la
Societe Philomatique an article in which I acknowledged M. Fourier's prior

4Ca/Cul des Indices des Fonctions,lithograph was published in Turin and reprinted in J.E.P. 15,
25th cahier 1837, pp. 176-226 (O.c., 2, 1, pp. 416-466).
 Appendix III 329

discovery.5 Similarly, when M. Sturm published his theorem on the number

of real roots, I was the first to congratulate him; and I did not write a word
mentioning that I had published, on this same subject, a work that had been
approved by the Academie and published in the Journal de ['Ecole Poly tech-
nique. 6 Meanwhile, M. Poisson gave a report that verified, as to an equation
of arbitrary degree, I was the first to have developed methods by which it is
possible to find rational functions with coefficients whose signs show the
number of real roots between given limits. Finally, in the 1833 study, I cited M.
Sturm's theorem. I am as sorry as your are, and I am very astonished, that you
have not been able to find in Paris a copy of the paper and of the article that
was published in the Gazette du Piedmont. 7 That by no means proves that they
cannot be found there; and, in particular, I remember that, at the moment
when I distributed to the Academie the article that was published in the
Gazette du Piedmont, one of my colleagues told me he regretted being unable to
read the article in Italian. In the meanwhile, I requested that the study be
deposited with the Bibliotheque de l'Institut. Perhaps someone borrowed it
and did not return it. That is all I can think of. At least, however, it can be found
in the hands of scholars here in Italy, especially in Turin, Milan, and Modena.
The only copy that I have in my hands is now in a two-volume work
containing various studies that were lithographed in Turin. While I am
waiting to see if you can find the paper, I will have a copy made for you and
send it. You will see the elementary proofs that you are interested in, especially
for the case that the equation

where u is a rational fraction in x and uo, U its values corresponding to x = X o

and x = X. It is from this equation that I deduced M. Sturm's theorem in the
study of 1833.

I am waiting with the greatest impatience for the news of the presentation to
the Academie and of the printing of the three letters, each 15 or 16 pages, that I
sent to you. These letters were dated May 6, May 13, and May 18. They
contain results that are as important as the proof of the theorem given at the
end of the 1833 paper. 8 I do not understand that the objectives raised by M.
Sturm should have suggested to you the thought of delaying the presentation
of the three letters. In fact, it is just one more reason for letting them appear
right away, unless you found some obvious errors, which you could then have

5 See Chapter 7, p. 113.

6 J.E.P., 10, 17th cahier, 1815, pp. 457-558 (O.c., 2,1, pp. 170-257). See Chapter 3, p. 39.
7 Gazetta piemontese, 113, September 22, 1832, p. 620. The text is reproduced by A. Terracini in

'Cauchy a Torino', Rendiconti del Seminario Matematico dell Universita e del Politecnico di
Torino, 16, 1957, pp. 178-179, but it does not appear in the Oeuvres Completes of Cauchy.
BSee O.c. 1,4, pp. 48-81.
330 Appendix III

eliminated. However, I do not think that they can be shortened. The theorems
stated are, for the most part, given with such clarity and precision that it will be
very easy for you to verify them; and, for those that might possibly constitute a
problem, I myself have taken care to make a corresponding note. The only
thing I fear is that this delay will give other persons the time to publish, before I
do, the proof of my theorem -and you can very well imagine how
disconcerting that would be, particularly since M. Libri and other geometers
have pressed me to send my proofs, which I did.

What I am requesting of you, then, is that you publish my letters as quickly as

possible, showing the date that they were postmarked. Please communicate
the following note to the Academie.

Note: The fundamental theorem on the integration of differential equations

stated in a letter of January 29 can be immediately deduced from the 1st
theorem stated in the letter of May 2 and the principles discussed in the
lectures for the second-year course at the Ecole Poly technique (see the copy in
the Bibliotheque de l'Institut of the first eleven lessions). 9 I will explain this in
more detail in another letter.

Please let me hear from you and be assured of my respect and unchanging
attachment.

Yours,

A. L. Cauchy

Letter from Cauchy to Moigno written December 15, 1837,

Melanges notebook of Madame de Pomyers, p. 107
Dear Abbe

I am honored to send to you a new note on the theory of light; it will be

followed by several others.10 In waiting until I had had these various notes
published, I request that they, as well as the preceding ones, be inserted in the
Comptes Rendus-or, if that should not be possible, that they be lithographed
or hand copied, as was done with some of my lectures at the College de France.
In this way, the expense will be small, and a copy can immediately be deposited
at the Bibliotheque de l'lnstitut after it has been properly noted in the minutes
of the Academie's meeting.

90n these lectures, see Chapter 5, p. 80.

lOThe reference here is to the 1837 study Recherches sur la tMorie de la lumiere, the first part of
which is entitled 'Polarisation rectiligne'. The Study is in the Melanges notebook. It has not been
published.
 Appendix III 331

Letter to Catalan, July 25, 1839, Bibliotheque Generale de

l'Universite de Liege, Correspondence of Catalan, Ms 1307C
My dear Friend and Colleague

You promised to come to Sceaux one day to dine with us, and I am asking
today if you would like it to be on next Tuesday, following the Academie
meeting. We dine at six o'clock, and I am looking forward to this opportunity.
As we should like M. Dirichlet to be with you, and since I do not know his
address, I am now mentioning it to you and am asking that you be kind
enough to forward the enclosed invitation to him.
Your good friend,
Baron Augustin Cauchy
Sceaux, July 25, 1839
Rue de Voltaire, No. 49

Letter to Dirichlet, July 25, 1839, Staatsbibliothek preussischer

Kulturbesitz, Berlin, Nachlass Dirichlet
Dear Friend and Colleague,

I am now asking that you please fulfill the promise that you made to me the
other day by coming to dine with us at Sceaux on next Tuesday at six o'clock.
M. Liouville should also be there. I hope that this does not present you any
inconvenience and that you will be assured that I will receive your
acknowledgment with the greatest pleasure.
Yours truly,
Baron Augustin Cauchy
Rue de Voltaire, No. 49

Letter from Cauchy to Moigno, October 26, 1842, Bibliotheque du

Centre des Fontaines, Chantilly, Moigno file

My dear Abbe

I have again thought about what we said yesterday and have become more and
more convinced that in free space there are no knots or bulges-at least so that
a reflected ray shall not interfere with an incident ray. For instance, if one
places himself behind a wall and if the sound reaches the ear of the observer
through an opening made in the wall, so that there is no reflection from
332 Appendix III

neighboring walls, there will be no knot, properly speaking, at which the

intensity of the second becomes zero. There will only be, as I said at the last
session, diffraction of the sound waves, and the points corresponding to the
maxima and minima are obviously located on parabolas whose parameters
form an arithmetical progression. I should like you to tell me tomorrow if this
absence of knots in free space is confirmed by experiments and, in particular, if
this agrees with the experiments of Monsieur M.ll

I would also like you to bring me last Wednesday's copy of Le National

tomorrow, as well as the medal claimed by Alicia. Finally, I cannot tell you too
often how much I believe in what was not replied to in last week's articles. The
objections that could be made fell flat in view of the actual facts. We count on
you tomorrow. If it should be impossible for you to come, you can send me, but
without the wrapper, the article from the National.

Yours,

Baron Augustin Cauchy

Sceaux, October 16, 1842

Letter to Saint-Venant, February 6, 1843, fonds Saint-Venant,

Archives E.P. not classified
Sir,

In a note that you sent to the Academie this morning, you mentioned the last of
the notes that I inserted earlier in the Comptes Rendus of the meeting before the
last session. 12 I said that the definition of pressure that you adopted led
precisely to the formulas that I first gave, if! am not mistaken, in Volume III of
Exercices de mathematiques. I added that my note was only an extract from a
study which should soon appear in the Exercices d'analyse··· (?)13 If in the
Compte Rendus of the session before last I did not give the proof even of the
proposition recalled below, it was simply that I was afraid that the article
would become too long.

On balance, this proof was precisely based-and I make this remark today-

llThe reference here is undoubtedly to Savart (see c.R. Ac. Sci., 15, pp. 761-762 and c.R. Ac. Sci.,
15, pp. 815-816).
l2'Note sur les pressions supportees, dans un corps solide ou fluide, par deux portions de surface
tres voisines, rune exterieure, I'autre interieur a ce meme corps', C.R. Ac. Sci., 16, January 23,
1843, p. 151 (O.c., 1,7, p. 252).
13 'Memoire sur les dilatations, les condensations et les rotations produites par un changement de
forme dans un systeme de points materiels', Ex. An. Phys. Math., 2, pp. 302-330 (O.c., 2, 12,
pp.343-377).
 Appendix III 333

on the consideration of the groups of molecules that are situated, pairwise, at

the extremities of parallel and equallines. 14

The proof that you gave in your notes uses the same argument. I observed, as
you do, that the groups of molecules that are situated pairwise, as I just said,
are always found on a right or oblique cylinder.

I should like to know if this proof is already found in the paper you presented
to the Academie at today's session. In this latter case, my note would be
pointless, and I would withdraw it. In any case, write a few lines to me here in
the country where I am now residing, that is, in Sceaux, on the Rue de Voltaire,
No. 41, Department de la Seine.

Yours truly,
Baron Augustin Cauchy

Letter to Saint-Venant, July 31, 1845, fonds Saint-Venant,

Archives E.P. not classified
Sceaux, July 31, 1845
Sir,
Yesterday, I presented a study to the Academie that had been initialled by the
Permanent Secretary and included a considerable number of formulas and
theorems on geometry.1S I discussed them with several of my colleagues, who
thought them to be new discoveries; I intended to present the main results to
the Academie itself next Monday. However, toward the end ofthe last session,
I had an opportunity to speak with M. Lame, and I found out from him that
M. Binet is presently in possession of one of your papers that probably
contains certain results that are similar to mine. 16 Accordingly, yesterday, I
unsuccessfully sought to get in touch with M. Binet who, as it turned out, was
not at the Academie. He had left for an outing in the country and, I am told,
will be out of town for several days. Nevertheless, I should still very much like
to know if there is any connection between your formulas and some of those
that I have come up with; and, thus, in order to satisfy myself on this point, I
am forced to ask for your assistance. There are, in particular, two points that I
hope can be clarified.

14'Memoire sur les pressions ou tensions interieures, mesurees dans un ou plusieurs systemes de
points materiels que sollicitent des forces d'attraction ou de repulsion mutuelles', c.R. Ac. Sci, 16,
February 6, 1843, p. 299 (O.c., 1, 7, pp. 252-260).
15 'Memoire sur de nouveaux theoremes de geometrie et, en particulier, sur Ie module de rotation
d'un systeme de !ignes droites menees par les divers points d'une directrice donnees', C.R. Ac. Sci.,
21, July 30, 1845, p. 273, and August 4, 1845, p. 305 (O.C., 1, 9, p. 253).
16Probably 'Memoire sur les !ignes courbes non planes', J.E.P., 18, 30th cahier, 1845, pp. 1-76.
334 Appendix III

In my study, I established the main properties of two systems of curves traced

out simultaneously on an arbitrary surface. These curves include, as a special
case, those that M. Lame found as orthogonal lines, as well as a considerable
number of others. They lead to curved coordinates formed by the intersection
of three systems of curved surfaces. From this, in order to arrive at the various
results I obtained, I used new properties of curves with double curvatures and
the special case of the modulus of rotation of a system of lines drawn through
various points of a given directrix. This modulus of rotation is simply the
quotient obtained by dividing an infinitely small arc of the directrix into the
infinitely small angle formed by two straight lines drawn from the endpoints of
the arc. The inverse of the modulus of rotation is a certain radius that reduces to
the radius of curvature when the system of lines is identified with the system of
tangents and, when the lines are perpendicular to the osculating plane, with
the radius of the flexion. Moreover, the sum of the squares of the moduli
corresponding to these two cases is precisely the square of the modulus of
rotation relative to the case in which the system of lines is reduced to the
system of the radius of curvature, etc.

Please be kind enough to see if there is something in common between this

theorem and those that you have obtained and reply to my letter as soon as
possible. I will be in the country-that is to say, at Sceaux-today (Thursday)
and tomorrow (Friday) as well as Sunday. On Saturday and on Monday, I will
be in Paris, so that if it is your intention to come and see me in the country, then
it would be better not to choose either Monday or next Saturday. Please
forgive my scrawl, but I do not have time to check and correct this letter, since
the mail will be picked up very shortly. I close by offering you, once more,
assurance of my best wishes.

Baron Augustin Cauchy

Letter to Leverrier, November 23, 1846?, Bibliotheque

de l'Institut, Ms 3710
Rue Serpente
Paris, November 23

Dear Colleague,

I have been working with determination, and I got up at six o'clock this
morning to complete your calculations. It could not have gone better. But, my
poor brain could not endure this sort of work. When I got ready to leave for the
Institute, I became quite ill and was forced to go to bed. Nevertheless, I have
quite a few things to discuss at the Academie, things that I should think will be
of great interest to you; since my new paper is ready, I do not want to wait until
next Monday. I fully appreciate the depth of our friendship, and so, in this
 Appendix III 335

extremity, I must call on you. I am not far away from either the Rue Saint-
Thomas or from the Academie. You would be of particular help to me if you
would allow me a few moments of conversation so that you would be able
today to inform the Academie of everything that I have done. I hope that you
can come.

Come ... (?)

Yours,

A. L. Cauchy

Letter to Herschel, undated, 1848, Archives of the

Royal Society, Ms 328
Sir,

Being called to America to direct the astronomical works at the Observatory

in Georgetown, Father Vieo-who is rightly respected by all true friends of
science I 7 -asked me to write a few lines to you on his behalf. I am now taking
advantage of his request in order to refreshen your memory of me.

Allow me, then, to use this opportunity to ask you to offer in my name to the
Royal Society of London my study on the theory of light. This work was
published in 1836, and I presented a copy of it to you when, on your trip to
Paris, I had a chance to meet with you for a few moments,IS I am also
enclosing a study, which I published in 1813, on the method for an a priori
determination of the number of real positive roots and· the number of real
negative roots of an equation of arbitrary degree. This method furnished the
first solution to the problem that has since been solved, by other means, by M.
Sturm's theorem. Aside from this, it is known that the method in question and
M. Sturm's theorem are found to be covered by certain more general theorems
dealing with arbitrary roots-real or imaginary-that I published in 1833.

Finally, I am enclosing with the 1836 study on light several reports that were
written in 1813; one of them contains the proof that I gave at that time of

17Franco de Vico was an Italian Jesuit astronomer. When the revolution of 1848 broke out,
Father de Vico left Rome to become Director of the Georgetown Observatory. He died on
November 15, 1848, in London while en route to his new appointment.
18 See C. R. Ac. Sci., 8, May 17, 1839, p. 38, for more on Daguerre's intervention: 'M. Cauchy, who
also saw M. Herschel when he [M. Herschel] passed through Paris, confirms M. Arago's remark.
He adds: M. Herschel declared that the attempts made in England are, in fact, mere child's playas
compared to M. Daguerre's methods. M. Talbot himself will soon come to share my opinion, for I
am going to write to him and ask him to come to see these wonders'.
336 Appendix III

Euclid's theorem. The last two opuscules, which I just mentioned, are in two
copies. I am asking that you will please be kind enough to offer one of the two
copies to the Royal Society, and I also hope that you will be assured of my
highest regards and deep respect.

Your devoted friend,

Baron Augustin Cauchy

Letter to Saint-Venant, October 5, 1850, fonds

Saint Venant, Archives E. P. not classified
Sceaux, October 5, 1850
Sir,
Upon returning from a short trip to Normandy, I read the letter that you were
so kind as to write me, and as you requested, I am hastening to send back to
you the paper that you presented to the Academie des Sciences on August 26. 19
As to the information that you would like to have relative to M. Dubuat, I
think that you can obtain it from M. Binet, who was Inspecteur des Etudes at
the Ecole Poly technique in 1816 and who continued in that capacity until
1830. 20 I pray that you give my highest regards to Mme de Saint-Venant and
that you will be assured of my highest estimation and my sincerest attachment.

Baron Cauchy
MMes. de Bure and Cauchy remember you fondly and ask me to bid you give
Mme. de Saint-Venant their compliments and greetings.

Letter to the Dean of the Faculte des Sciences,

May 14, 1852, Archives Ac. Sci. Cauchy file
Sceaux, May 14, 1852
My dear Dean,
I have received the letter that you so kindly sent me relating to the formality
prescribed by Article 14 of the Constitution. You understand, I am sure, the
caution and even-should it be necessary-the sacrifices that the position in
which I now find myself placed imposes upon me.

19'Formules nouvelles pour la solution des problemes relatifs aux eaux courantes' is a work for
which Cauchy served as commissioner. On October 25, Saint-Venant gave a sequel to this work.
2°Saint-Venant published the Notice sur La Vie et Les Travaux de Pierre Louis-Georges Comte du
Buat in 1865.
 Appendix III 337

When, in 1830, the Duke of Orleans ascended to the ruins of a throne that he
himself had worked to destroy, I renounced the three (professorial) chairs that
I then held and swore to myself to remain faithful to the oaths I had sworn. Can
I forget this promise, after the signal honors that Charles X bestowed on me by
requesting me as teacher to the heir of Saint-Louis, Henri IV, and Louis XIV?

If, in 1849, I did not hesitate to subscribe to the requests of certain of my

colleagues who pressed me to return to a career in teaching, it was not because
(and that is surely well known) I had any illusions about the dangers of the
anarchist doctrines and theories that threaten to bring desolation to our
homeland and overwhelm society everywhere in Europe. But, the political
oath came to be abolished, and I was happy to think that I might once more be
of use in guiding youth along toward the harmless study of a science that has
ever been the delight of my life. My efforts in this respect should be easier yet
under a government whose head has so loudly proclaimed his deep desire to
see growing and flourishing among us that true science that enlightens minds
without corrupting hearts.

Sir, I deeply regret the fact that the impossibility to subscribe to the oath
should separate me from my honorable colleagues and oblige me to renounce
my heartfelt hopes of working alongside them in serving my country in a field
in which I am able to do some good.

Please be assured, Sir, etc.

Augustin Cauchy

P.S. Please let me know if! should still hold my class session in mathematical
astronomy next Monday morning.
 Appendix IV
Unpublished Documents on Cauchy's
Two Candidacies to the
College de France (1843 and 1850-1851)

I. The Election of 1843

Letter from Liouville to Letronne June 6, 1843

(Archives of the College de France
C. XII, Liouville 1 A)

Paris, June 9, 1843

My Dear Mr. Administrator:

I am honored to ask you to present me as a candidate for the vacant chair in
mathematics at the College de France, a vacancy created by the death of your
venerable colleague, M. Lacroix. The members of the Committee are of the
opionion that the works I have published as a geometer and the students I
have taught as a professor entitle me to their consideration. I am no less certain
that I am worthy of such consideration on account of my enthusiasm and
scholarly standards. However, putting aside any discussion of matters not
having to do with the science, I nevertheless think that I ought to state here
clearly that if the Committee should choose M. Cauchy, I would-far from
regarding myself as having been hurt by that choice- be the first to applaud it.
Fairness to the superior merit of a fellow scholar is, at least to my thinking, a
sacred duty that one should observe at all times, and I can entertain no base
thoughts of personal interest and advancement that run contrary to that duty.

Mr. Administrator, I am
Your very humble and obedient servant,

J. Liouville
338
 Appendix IV 339

Letters from Cauchy to Letronne, June 11, 1843 (Archives

of the College de France, B-11, Mathematiques b - 1)

1. My Dear Mr. Administrator:

You yourself can bear witness to the promptness with which I sacrificed my
own candidature out of the desire to avoid knowingly giving anyone the
slightest pretext for seeing my efforts and aims in a bad light. However, today it
appears that the delicacy of this way of proceeding has not been appreciated in
full by everybody, contrary to what I had hoped. This situation, of course, no
longer allows me the liberty of remaining on the sidelines, for there are certain
points at which the very honor of the contestants-when the dispute is of a
scientific nature-will not allow them to withdraw from the fray. Under this
assumption, then, Mr. Administrator, I now find myself obligated to reenter
the lists. Accordingly, I request that you please be kind enough to read to the
honorable professors ofthe College de France not only the present note (which
I hope you will be kind enough to do in any event) but also the enclosed letter. I
also hope that the honorable professors will show me some indulgence relative
to the formalities that I ought to have satisfied by recognizing that time did not
allow me the customary opportunity of visiting each professor separately.
Please be assured, Mr. Administrator, that I have the honor of being

Your very humble and obedient servant,

Augustin Cauchy

2. Dear Mr. Administrator:

If the honorable professors of the College de France believed that my having

devoted my entire life, over the past forty years, to the study and teaching of the
physical and mathematical sciences should entitle me to lay claim to the chair
that became vacant upon the death of M. Lacroix, I would be happy to think
that I can still be of use, as a professor of differential and integral calculus, both
to science and to France, my homeland. The honorable professors may be of
the mind that those works of mine whose aim is the advancement of
infinitesimal analysis and the courses that I taught over all those years not only
at the Ecole Poly technique but also at the College de France, courses which, I
might add, were attended by many students, who today successfully occupy
chairs in mathematics in various countries and are distinguished members of
the Institut de France as well as various learned societies all over Europe-
give me special claim to a favor whose value I now realize.
340 Appendix IV

I hasten to seize this opportunity to assure you, Mr. Administrator, that I am

honored to be

Your very humble and obedient servant,

Augustin Cauchy

Letter from Cauchy to Letronne, undated, probably

a few days prior to June 18, 1843 (Archives of the
College de France, B-II, Mathematics b-1)

Mr. Administrator:

The honorable professors of the College de France already know that I am

presenting myself as candidate for the chair that became vacant upon the
death of M. Lacroix. My long years of work-the courses that I have taught
not only at the Ecole Poly technique and at the Faculte des Sciences but also at
the College de France, courses which were taken by outstanding scholars, both
French and foreign, and even by members of the Institut-justify, I should
think, my request, and they would appear to give me some title to such a
flattering distinction. The high regard and goodwill that the honorable
professors of the College de France have shown me, in particular, those
professors who are themselves in the physical sciences or in mathematics and
the publicly expressed esteem for my works, both by word of mouth and in
writing, not only by masters of the science whose loss sadden us all-scholars
such as MM. Lagrange, Laplace, Legendre, and Ampere-but also by
scholars now at the College de France-all this allows me to hope for success.
However, one objection has arisen that might give reason for doubt to the
professors. It might be, I am told, that my election may not be acceptable to the
government. However, I am morally convinced that the contrary is true.
Today, I believe that I am able to assure the professors that, in case they should
favor me, there will be no serious problems to fear from the government.
Moreover, I can assure them, if that be necessary, that the instructional
program in mathematics at the College de France will not be subjected to any
interruption.

Please be assured, Mr. Administrator, that I am honored to be

Your very humble and obedient servant,

A. L. Cauchy
 Appendix IV 341

Letter from Libri (the addressee is unknown,

but was probably Letronne), undated, probably
June 17, 1843 (Archives of the College
de France, Libri file)
Sir and Very Respected Colleague:
Allow me to add a few words to the official letter that I had the honor of
addressing to you the other day. M. Thenard has urged me to ask you to state,
on my behalf, to the Assembly of Professors-in case this is necessary-that if
I should be elected I will never become involved with nor accept any function
or position that would require me to be out of Paris, even temporarily, during
the school year....
I forgot to tell you in my first letter that before substituting for M. Lacroix, I
had filled in for M. Biot. I was a substitute Professor at the College de France
for eleven years. I ceased filling in for M. Biot, who paid me 2000 francs per
year, only in order to undertake (gratuitously) M. Lacroix's course, who could
not afford to pay a replacement.
An absurd rumor is now afoot, a rumor that has it that the Jesuits are pleased
by my candidature. This hateful trick seems to have made a big impression on
M. Michelet. I need not tell you, Sir, that nothing whatever can make me give
up my struggle against the Jesuits. My second letter is going to appear on the
15th ofthis month. If you could make it clear to everybody that if! am rejected,
the Jesuits will cry victory in their journals, then you will have done me an
outstanding service. Moreover, Sir, I ask you to please use this statement as
you think best in the eyes of your colleagues.
It is important to everyone that the election should be held as soon as possible.
If then, there is a second session held for the voting, I ask you to promptly get
an understanding with your colleagues so that this second session can take
place one day next week, preferably before Thursday, if at all possible. Sunday
should be rejected outright because nowadays everybody goes out of town on
that day....
G. Libri

Letter from Liouville to Letronne, June 19, 1843

(Archives of the College de France,
e. XII, Liouville, 1 B)
Mr. Administrator:

I am deeply humiliated, both as a man and as a geometer, by what took place

yesterday at the College de France. Accordingly, from now on it will be
342 Appendix IV

impossible for me to teach at that institution. Please receive and accept my

resignation as a substitute Professor there.
I have the honor, Sir, of being
Your very humble and obedient servant
J. Liouville

Letter from Cauchy to J. B. Dumas, June 20, 1843

(Arch. Ac. Sci. Libri file)
June 20, 1843

My Dear Mr. President:

I hasten to send to you what you honored me by requesting, namely, a

readable copy of the reflections that were presented to the Academie last
Monday. I take joy in the fact that at last, thanks to you, the question is going
to be squarely put and that there will be an official answer. After all the
testimonies of respect and esteem that my colleagues so kindly gave on my
behalf, it would be fitting neither to the Academie nor to me if I were to
renounce my candidature; for, at this time the battle is purely a scientific one.
This is, of course, the reverse of the situation wherein it would be convenient
neither for me nor the Academie that I should consent to remain a candidate if
the battle were on political grounds; in such a case, I would certainly withdraw.
As to the rest, I shall always have the same confidence and moral certitude that
I dared to express on Monday before the Academie. I am well convinced that
as soon as the honorable Minister reads my observations and remarks and has
a clear account of the present state of things (and, in particular, of the fact that
it has now come to pass that the chair in political economy at the College de
France has now been bestowed to a foreigner, a person who is not even
naturalized), as soon as learns the details of what a respected jurist consulted
by his colleagues of the Academie des Inscriptions et Belles-Lettres said at the
Institut a few days ago, he will not exclude a French geometer from a chair in
mathematics at the College de France by subjecting him, and him alone, to a
formality that has long since fallen into disuse at the College de France so that
it was decided to name as professors persons who are found in this obvious
incapability to fill positions. Thus, my dear Mr. President, thanks to you, the
Academie on the one hand and Ion the other, will both know what we have to
do. Thanks to you, the Academie will officially know whether or not I have
been defeated by a political burden that will never allow me to play either an
active part in a learned body or function as a professor of mathematics. Those
things are for me, I swear, a true delight; in a way of speaking, they are my
element and my life. I hope you will not be condemned to come to declare this
 Appendix IV 343

incapacitating burden before the Academie. I myself cannot conceive of how it

would be possible to even frame such a principle in any event without being
forced by the very logic ofthe situation to exclude me from the Academie itself.

I say no more. Your words and your thoughts will be much more easily
understood than mine; I am so certain of it that I beg you, my dear Mr.
President, to accept my thanks in advance with my best wishes.
Yours truly,
A. L. Cauchy
This June 20, 1843

Cauchy's statement to the Academie on June 19, 1843, together

with Libri's letter (Archives Ac. Sci., Libri file; rough draft in the
Bibliotheque des Fontaines, Chantilly, Moigno file)
Academie des Sciences. Meeting of June 19, 1843. A letter written by the
Minister of Public Instruction was read by the president of the Academie des
Sciences. Afterward, M. Cauchy took the floor and expressed the following
thoughts and reflections.

The letter that you have just heard imposes on me a duty to state the position
I take relative to the kind of competition opened by this letter to the
mathematicians in the Academie. My task is to explain what must be thought,
what I myself think of the hypothesis, which is accepted by certain persons and
rejected by others, that I am a candidate to fill the vacancy created by the death
ofM. Lacroix. I am pleased that I have nothing to say on this subject that will
not be satisfactory to the most sensitive susceptibilities, nothing to say that will
not be appropriate to the rejection of a subject that is such a source of
annoyance and disunion among scholars and, in particular, among the
members of the Academie. My disinterest, the moderation of my wishes, and
the frankness and loyalty of my character are so well known that the Academie
need have no fear that I shall say anything that any person can complain of, or
that I will say anything within these walls or even outside of these walls that
could be offensive to anyone.

As I had withdrawn to the country, I occupied myself by responding to the

confidence that the Academie itself had placed in me by appointing me
president of the committee responsible for evaluating works submitted in the
competition for the Grand Prix de MatMmatiques, and also being deeply
involved in the calculations that this examination required, I happened to
attend the session before the last at the Academie. I learned from the mouth of
my respected colleagues that there would be a question of replacing M.
344 Appendix IV

Lacroix. Several of them even pressed me to apply. I admit that I was not
insensitive to hopes of once more being useful, as a professor of mathematics,
both to the friends of science and to my homeland. I can recall, with real joy,
that I have seen gathered about my chair-at the Faculte des Sciences, at the
Ecole Poly technique, and at the College de France-outstanding scholars from
all over Europe. Of these, several have since become members of this very
academy. I believe my enthusiasm and my strengths will still permit me to
contribute to the advancement of the mathematical sciences. I am the more
disposed not to push my case myself since I know so well what the candidates
think of my works, what they have written about them, and what they have
recently said about them; and I was not at all surprised by the interest that, on
this occasion, was shown in an old professor by the candidates, who once
attended my courses at the College de France. Moreover, the goodwill with
which I was honored by the professors at the College, especially by those who
are involved in the physical and mathematical sciences, encouraged me to
follow through on their idea. But, a single objection has aFisen, an objection
that might be the source of doubts and questions in the minds of certain
persons. It was feared, I am told, that based on motives having nothing to do
with science, the authorization to teach courses would not be accorded to me
in case I should be appointed by the College or even by the Academie itself. I
believe the contrary is true. In France, even among the most opposite parties
and factions, justice has always been accorded to nobility of sentiments and to
open, loyal, honest conduct. We understand that sense of steadfastness in the
face of misfortune by which one imposes great sacrifices on himself rather than
knowingly give the slightest pretext for reproach or for the most blame. I recall
the anecdote often told about Professor Scarpa. 1 I also think of the words
addressed by one of the members of the Academie des Inscriptions et Belles-
Lettres by a Minister when the question arose over awarding the chair in
political economy at the College de France to a foreign scholar.

Am I under any illusions myself? I think not. I know only one thing, one thing
that certain other persons assume. Some people-in spite of my entreaties and
earnestness, in spite of a letter that I wrote and gave my thoughts on these
matters-have protested against the generally accepted hypothesis. Be that as
it may, in spite of my desire to be able-at least as long as my strength will
allow-to be of service as a professor of mathematics to the advancement of
science in France, my homeland, I will state once and for all that I will only
consent to my name being placed on the list of candidates for a chair in

1 Antonio Scarpa (1747-1832) was a famous Italian surgeon. In 1796, when the Italian republic
was established, Scarpa, who was devoted to the monarchy, refused to sit on the Committee of the
Juniori or to take a loyalty oath, and no one bothered him at all. As to the foreign scholar referred
to, the reference is to Count Rossi (1787-1848), who was appointed to a chair in political economy
in 1832 following the death of J. B. Say.
 Appendix IV 345

mathematics at the College de France if there are no serious nonscientific

obstacles against my candidacy whatsoever.

Letter from Sturm to J. B. Dumas, undated, undoubtedly a few

days before July 3, 1843 (Arch. Ac. Sci., Libri file)
Mr. President:

The two most senior members of the geometry section were of the view that
any section called upon to designate candidates for a vacant chair ought be
bound simply to making known the names of those persons who present
themselves (as candidates for the vacancy), if they are members of the
Academie. Monsieur Lame and I thought that we should cede on this point,
although the principle seems debatable to us.

However, we reserved the right to expose to the Academie our personal and
individual opinion as regards the incompetence of the only candidate who
presented himself to the section. M. Libri has shown by his public teaching
that he has little or no talent as a professor. It is to be feared that his reputation
as a geometer and as an academician has suffered because of it. Students do not
attend his classes, even though the professor has made them shorter. These
facts are publicly known.

We truly regret the voluntary or forced withdrawal of those two eminent

geometers, MM. Cauchy and Liouville. Both of these scholars were, at first,
presented as candidates for the chair in question and both were unquestion-
ably qualified. Since, in addition to them, several other geometers with known
success in teaching might be interested, we have no recourse except to declare
our formal opposition to M. Libri's candidature. Our protest has no goal
except for the advancement of the mathematical sciences and those who
cultivate them. In concluding, we think it should be pointed out to
the Academie that, by its very own report, the section does not present M.
Libri to the Academie's vote, it merely announces that M. Libri is presenting
himself.

M. Lame, being absent, has authorized me, by a letter (now) at may office,
to express his views to the Academie. They completely agree with my
own.

C. Sturm
346 Appendix IV

II. The Election of 1850-1851

Letter from Cauchy to the Administrator of the College de France,

November 18, 1850 (Archives of the College de France
B-II, Mathematiques, b-7)
My Dear Mr. Administrator and Honorable Professors:

Some of you have been of the opinion that it is once again time to allow the
voice of a certain person to be heard at the College de France, a person who in
1817 was called by a brilliant professor-a true master of the science-to
teach courses that, for many years, were assiduously studied by some of the
outstanding scholars who today are members of various academies through-
out Europe.

Some of you have been of the opinion that it is fitting to thus satisfy the wish
expressed on various occasions, and quite recently, too, by young professors as
well as by French and foreign scholars, who with such enthusiasm studied the
course on celestial mechanics that I offered at the Faculte des Sciences. These
scholars have also wanted to see developed in a special course the new
methods that I have been working on for 36 years, the solution of various
problems in mathematical physics and astronomy.

The sincerity and goodwill with which these thoughts were expressed to me,
and the sense of urgency with which they have been communicated over the
past few weeks by respected professors whom I hardly know, have convinced
me that I should follow these suggestions.
My former and present achievements are too well known by the friends of
science to need elaboration here. However, if it should be necessary that they
be recalled, I can call on the testimony not only of geometers and physicists
who presently hold positions among you, but also on the testimony of those
who would like your recommendations and votes.

Be assured then, Gentlemen, of the value I now attach to making myself ever
more worthy of having the title of your future colleague, a title that some of
you were kind enough to honor me with as you shook my hand.

Paris, November 18, 1850

Augustin Cauchy

Dear Mr. Administrator:

I am honored to submit my candidacy for the chair in mathematics at the

 Appendix IV 347

College de France. I should also like on this occasion to ask for your kind
support and that of your colleagues.

My reasons for this decision are stated in the note that is enclosed and is
addressed to the professors, and I hope you will be kind enough to read it to
them.

I pray you please be assured, Sir, of my best wishes.

Sceaux, November 18, 1850

Augustin Cauchy

Letters from Quatremere to Binet (Archives Ac. Sci. Binet file)

1. Letter of December 3, 1850

Sir and Dear Colleague:

I have just learned from M. Sedillot that during the meeting in which M.
Letronne was appointed Professor of Archaeology at the College de France
there were 20 votes cast, with M. Letronne receiving 10 voices since 10 blank
ballots were found in the urn. On this basis, it was decided that the candidate
had, in fact, received a majority of the votes. Thus, the College has adopted a
position that is contrary to the one that I had always assumed it to have. It
follows from this situation that M. Cauchy, having 11 votes as against 10, with
one blank ballot, should have been recognized as the College's candidate. I
have just written to M. Barthelemy-Saint-Hilaire in this sense and to protest
that I was led into error by adhering to the procedure that has always been
followed at the Academie des Inscriptions et Belles-Lettres.

My dear Colleague, I hope that you will accept my expressions of regret along
with my highest regards and sincere devotion.

Quatremere

2. Letter, undated probably on December 5, 1850,

or December 6, 1850

Dear Sir and Learned Colleague:

I am honored to communicate to you the reply that was sent to me by the

honorable Administrator of the College. As you can see, he circumvented the
348 Appendix IV

problem, but did not even try to resolve it. The example having to do with M.
Pelouze seems not to be well selected; for, if my memory serves me correctly,
M. PeIouze received the totality of the votes cast. Accordingly, his election
cannot really offer a basis for any discussion. I hasten to submit to you the
draft of a letter that we can put before the Minister. I urge you to read over this
rough draft and make whatever changes you think fitting. You may also
discuss it with M. Cauchy.

Dear Colleague, please accept my expression of deep attachment.

Quatremere

M. Cauchy is leaving my house; he should see you in the morning.

Letter from Desgranges, Quatremere, and Binet to the Minister

of Public Instruction, December 6, 1850 (Archives of the College
de France, C XII, Liouville, 5A)
Dear Mr. Minister:

We are honored to call your attention to a situation whose seriousness you will
readily appreciate. It is a situation presenting a very real problem that
demands a sure and invariable solution for the future as much as for the
present.

The chair in mathematics at the College de France is vacant, and the

professors were called together to designate the candidate who was to be
presented to the selection of the government. There were only two persons
who sought this position, M. Cauchy and M. Liouville, and both are members
of the Academie des Sciences. There were twenty-two votes cast, and on the
first ballot, M. Cauchy received eleven, while M. Liouville received ten, there
being one blank ballot cast. The honorable administrator was of the view that
as neither candidate had received an absolute majority of the votes, it would be
necessary to hold a second balloting. Needless to say, no objections were
raised at the time against this procedure, for there was not, within the memory
of any of us, a comparable situation. Accordingly, a new vote was held, and this
time M. Liouville received twelve votes and thus, in effect, became the
College's candidate.

Several of us subscribed, without qualms, to this decision, which appeared to

be quite regular. Nevertheless, some felt it important to know if a comparable
situation had ever arisen, and if so, whether the College de France had
followed the procedure that we used. Here are the results of our investigations
into this matter. At the meeting of November 27, 1837, the College was called
on to present a candidate for the chair in archaeology, and 21 professors were
assembled. One of these, however, declared that he would not vote, so that the
 Appendix IV 349

number was reduced to 20, and the majority was fixed at 11. On the first vote,
M. Letronne obtained 10 votes and his competitor 8, with two blank ballots. In
this situation, it was decided that as the number of voting professors had been
reduced to 18, M. Letronne had obtained the required majority. This situation,
as can readily be seen, is identical to the one that has just arisen.

In our desire to adhere, in our statements as well as in our conduct, to exact,

scrupulous, and impartial principles, we continued our inquiries by checking
the minutes and records of the College to convince ourselves that in other cases
a different procedure had not been followed. As far as we could determine,
nothing similar has arisen under the regulation of 1828. And this, to speak
frankly, was quite astonishing; for, it can hardly be presumed that so grave and
serious minded a body should, in identical situations, follow a legal rule that is
diametrically opposite to the one adopted in an earlier, similar case. It is thus
demonstrated that by the customs of the College de France the presence of
blank ballots [in a vote], by reducing the number of voting members, ought to
reduce, proportionally, the size ofthe majority required for the presentation of
a candidate.

This principle, which has been permanently adopted by the Academie des
Sciences, has been solemnly proclaimed by the Institut in a meeting of the five
academies. Finally, a decision by the Conseil d' Etat, which was approved by
the Emperor and thus carried the force oflaw, by the date of January 25, 1807,
prescribed in a formal manner that in electoral proceedings blank ballots were
to be meaningless and that the majority was to be based on the actual number
of valid votes cast.

By all the facts that have come to our attention, Mr. Minister, we are strongly
persuaded that M. Cauchy obtained a majority on the first ballot, that the
second ballot should be regarded as null and void, and M.Cauchy's right to the
title of candidate designated by the College should be recognized.

Please be assured, Mr. Minister, of our high and respectful regards.

Signed: Alix Desgranges, Quatremere, and Binet, Professor of Astronomy.
December 6, 1850

Letter from the Administrator to the Minister of

Public Instruction, December 10, 1850
(Archives of the College de France, C XII, Liouville, 3)
Dear Mr. Minister:

I think it fitting that I send you certain information on the last two meetings of
the assembly and supplement the minutes that I have transmitted to you.
350 Appendix IV

The College presented M. Liouville as its candidate for the chair III
mathematics. Two ballots were needed to decide the question.

At this first meeting, nobody spoke out against setting aside the first ballot, as
it appeared to everyone that the rule required that the blank votes be taken
into account and that the majority be based on the number of members
present at the deliberations and not on the number of votes cast.

At the following meeting, however, a discussion did take place relative to the
point determining whether or not blank ballots ought to be excluded, and this
time it was again understood that according to the terms of the regulation, a
majority could only be formed on the basis of the number of members who
were present.

We thought that all discussion on this matter would thus be unnecessary. But,
yesterday the debate was reopened, although only incidentally. Some
professors who had voted for M. Cauchy recalled that in 1837, in the election
of M. Letronne, the blank ballots had been disregarded as null and void; they
further claimed that it was fitting to do so in the present situation. The reply to
them was that in the election of M. Beudant the blank ballots had been
counted, and that from these two opposite precedents it was impossible to
draw any clear-cut guidelines. Moreover, it is contended that although the
regulation may indeed have been violated in the first instance, all efforts should
be made to avoid violating it a second time. Moreover, as to its formal terms,
there can be no possible doubt regardless of what the customary practices may
be as concerns other learned bodies or other deliberative bodies in general.

Mr. Minister, I would not bother you at all with a discussion of these debates,
seeing that no trace of them is to be found in the minutes of the meeting, and
that they therefore did not result in a formal proposition, if I did not fear that
certain efforts would be connected with the discussion. A remonstrance is in
the process of being drawn up and is to be signed by the professors who voted
for M. Cauchy, in the hope of placing in the balance a weight that, to my
thinking, ought not be there. I have thus sought, Mr. Minister, to warn you of
these efforts in order that you may more fully appreciate them.

One person at the College has even proposed to me an express change on

matters concerning the actual facts that I have just reported to you. The
decision that gave rise to the protestation has not been weakened at all, but
rather retains its full strength. One may later modify the regulation or its
interpretation, but art. 24 is perfectly clear; at the meeting where the
candidates presented themselves and at the following meeting, this article has
been unanimously understood to have the meaning that I understood it to
have, then and now. By common consent, the confrontation that took place
here yesterday can in no manner have the character of a new decision.
 Appendix IV 351

I have thought it my duty, Mr. Minister, to bring this matter to your attention,
since I am entrusted with the responsibility of applying the regulation. I should
be remiss in this obligations imposed on me were I to have kept quiet, and it is
only to protect the College's deliberative process against ill-conceived attacks
that I have thought it my duty to write this letter.
Please be assured, Sir, ...

Letter from Cauchy to the Minister of Public Instruction

(Arch. Nat. F17 15333)
My Dear Mr. Minister:
The problem has been submitted to you of determining precisely who the true
candidate is designated by the College de France for the chair in mathematics.
It is stated in writing that you now have before you that this question comes
down to the following: does a blank ballot have the power to annul and destroy a
majority?

General opinion, logic, custom, the bylaws of the College de France, of

deliberative assemblies, of academics, and finally even the very law itself (see
the Bulletin des Lois, 46th Series, no 104, Law 2178) all seem to point to the
resolution of this question in this sense: a blank ballot is a null and void ballot.

On the other hand, this question cannot be left undecided lest there be serious
inconvenience at the time in which the Academie is called on, in its own turn, to
present a candidate for the chair in mathematics.

Mr. Minister, I am certain that you do not want that I -a man who has
devoted 37 years to the advancement and teaching of the mathematical
sciences (as evidenced by the papers that you requested from me and that are,
today, before you) and who, beginning in 1819, has had as students some
outstanding scholars who are now members of various academies throughout
Europe-should find myself excluded from a career as a professor on account
of a blank ballot.

In the meanwhile, if I present myself to the votes of members of the Academie

des Sciences, it is because the scholars, the lawyers, and the professors, who,
after having supported my candidature, committed an involuntary error by
believing in the power of a blank ballot, seem today to be generally convinced
that my right to the title of candidate designed by the College de France cannot
be questioned.

Mr. Minister, I urge you to have the extreme kindness to tell me ifI am right or
wrong in my belief; and if, in your view, a blank ballot can annul and destroy a
352 Appendix IV

majority. If you have any doubts in this respect, it would be easy, it seems to
me, to throw light on this point since~at this very moment~you have at
your side on the Higher Council of Education scholars, magistrates, and
lawyers capable of clarifying this serious question.

Please be assured, Mr. Minister, of my high regards and consideration.

December 14, 1850

Augustin Cauchy

Letter from the Minister of Public Instruction to the Administrator,

December 21, 1850 (Archives of the College de France, C XII,
Liouville, 5B)
Dear Mr. Administrator:

I have just received a statement of protest dated December 6 from three of the
professors at the College de France. The protest has to do with what transpired
in the meeting of the Assembly of Professors that was called to consider the
presentation for the (professorial) chair in mathematics and resulted in M.
Liouville being designated as the candidate of the College.

I am pleased, Mr. Administrator, to transmit to you a copy of the letter.

I received, at the same time, a copy of the minutes of the proceedings of the
Assembly of Professors of the College that was held on December 8 and in
which I note that the discussion had been reopened for a second time on the
question of the vote for a candidate for the position of professor of mathe-
matics and that the meeting was then adjourned. Allow me to say to you, that
this new complaint, according to the statement of protest that I have received,
places in question, so to speak, the legality of the result of the vote of
November 15. It seems to be necessary that a formal decision be made and that
the College de France should inform me definitively as to precisely whom it
presents as its candidate.

Accordingly, I request that you, Mr. Administrator, kindly convoke the

Assembly of Professors right away to deliberate the questions that have been
raised by their colleagues' statement of protest.

Mr. Administrator, please be assured of my highest regards.

Parieu
Minister of Public Instruction and of Religious Affairs
 Appendix IV 353

Letter from the Administrator to the Minister of Public Instruction,

December 23, 1850 (Archives of the College de France, C XII,
Liouville, 6)
Dear Mr. Minister:

Before I agree to the desire that you expressed in your letter of the 21st of this
month, allow me to set before you some observations that I regard as essential.

The candidate presented by the College de France for the chair in mathematics
is M. Liouville. There can be no doubt on this point.

First of all, the minutes of the meeting confirm that this is so. After two votes
had been held, the College concluded that M. Liouville, having obtained a
majority, should be its candidate, and there was not the least complaint about
it.
In the following meeting, on December 1, 1850, nothing more was said about
it, only some members asked that there be a check of the circumstances of the
balloting. The Assembly acted on this request, and the minutes of the meeting
at which M. Liouville was presented were adopted without any changes.

Finally, at the last meeting, on December 8, there was not even a question
raised as to M. Liouville or his nomination. The question was not reopened,
insofar as it had been closed at the last two previous meetings, and the
discussion that took place had to do only with the interpretation of the
regulation in cases to come and with determining whether or not the blank
ballots should be counted toward the votes defining a majority.

In my opinion, Mr. Minister, the regulation is categorical in the sense that by

saying the majority ofmembers present, titular and honorary, it expressly means
the number of persons voting and not the number of votes cast. It is
particularly on this litigious point, a point that has only to do with the future,
that the debate has been focused. On the motion of M. Laboulaye, the
Assembly decided to adjourn once the nomination for the position in
mathematics had been made.

These, then, are the facts, Mr. Minister, and I urge you to convince yourself of
their validity by setting before you the minutes of our last three meetings.

Now, 3 professors out of 23 are asking that M. Liouville's nomination be set

aside as null and void and that M. Cauchy be accorded his place since he won
on the first ballot, as the blank ballots are to be disregarded.

This, Mr. Minister, is a question that you and you alone must determine. If you
should set aside the nomination of the College, there will be another meeting
354 Appendix IV

so that a ballot can be held at which, it would appear, no reversal will be

forthcoming. If, on the other hand, you accept the reasoning of MM.
Quatremere, Alix Desgranges, and Binet, there will be no need to consult the
College at all, for M. Cauchy would be the candidate, plain and simple.

Mr. Minister, I urge you to take these matters into account, and I will await
your reply before I convoke a special meeting of the professors, who, you may
be certain, regard this matter as being completely closed. The only point that,
in their view, remains to be settled, is to interpret the regulation so that in the
future there will be no possible delays.

Please be assured, Mr. Minister,

Letronne

Letter from the Minister of Public Instruction to the

Administrator, December 30, 1850 (Archives of the College de
France, C. XII, Liouville, 7)
Dear Mr. Administrator:

I hasten to reply to the observations you so kindly sent to me on the 23rd of

this month concerning the request that I put before you that you call a meeting
of the Assembly of Professors and submit to it the protest that has been
formulated by three of the professors against M. Liouville's selection as
candidate for the chair in mathematics.

You seem to think, Mr. Administrator, that this complaint has only to do with
future situations, that the election was completely in order relative to the time
it was held as well as in its confirmity to the regulation and the majority of
members present. Moreover, you added that the Assembly resolved that after
the nomination had been completed there should be a determination on
whether or not blank ballots should be counted in the total votes establishing a
majority.

A closer look at the regulation of the College de France, however, has

persuaded me, Mr. Administrator, that the question cannot be regarded in the
light in which you are considering it. Based on the information contained in
the minutes, the situation comes down to the following fact: neither applicant
was properly elected as candidate of the College and this can be concluded
from Article 24 of the regulation, which asserts:
On the day designated, the election of the candidate shall be effected by
vote. The balloting shall be repeated until one of the applicants shall
have obtained an absolute majority. This majority is to be based on the
 Appendix IV 355

number of eligible titular professors together with the honorary

professors present at the deliberation.
You must not ignore, Sir, the distinction to be made between an absolute
majority and a relative majority. Since the number of eligible titular professors
is 25, the true number for an absolute majority would be 13. As neither M.
Cauchy nor M. Liouville obtained this number, it is obvious that neither of
them can rightly be regarded as having been legally elected by the College de
France.

There are thus grounds, Mr. Administrator, for proceeding to an election

again, without going into the question of blank ballots as raised in the
statement of protest. I, therefore, urge you to call a special meeting of the
Assembly of Professors so as to elect, according to the sense of Article 24, a
candidate for the chair in mathematics.

Please be assured, Mr. Administrator, of my highest regards.

Parieu
Minister of Public Instruction and Religious Affairs

Letter from Cauchy to the Administrator and Professors of the

College de France, January 4, 1851 (Arch. Nat. F17 15333 and
Archives of the College de France, B II, MatMmatiques, b-8)
Dear Mr. Administrator and Professors:

Prompted by your kind advice, I presented myself, some weeks ago, as

candidate for the chair in mathematics at the College de France.

A new candidacy arose on the eve of the election, and, at the meeting of
November 20 last, the count of the votes on the first ballot was 11 votes in
favor of my candidacy, 10 for the opposing candidate, and 1 blank.

Based on this unarguable fact, I should be able to consider myself as the

candidate of the College de France and to present myself confidently for the
vote at the Academie des Sciences. This vote, especially when there is a
question of selecting between two colleagues, is generally determined by the
results from the election at the college.

However, the grievous mistake of allowing a blank ballot to have the power to
negate a majority apparently robs me of the candidacy of the College and, in
effect, that of the Academie as well.

This last consequence of an extraneous error can today no longer be set right.
356 Appendix IV

Before the vote at the Academie, I vainly urged the professors and the Minister
of Public Instruction to recognize this obvious error, which is contrary to the
customary of procedures and bylaws of the College de France, as well as to
those of academies and other deliberative assemblies and even to the law itself.
In spite of my efforts, however, I have only been able to obtain an assertion
that the question had been declared to be in doubt.

Today, I learned that it is proposed that there be an examination of the double

issue of the blank ballot and the candidacy itself and that the two issues might
perhaps be considered separately.

Sirs, such a decision hardly conforms to justice nor, I might add, is it worthy of
you. The vote of the College should be free and independent, it being precisely
for this reason that it is always held prior to the vote at the Academie. A new
election to come at a time when the issues are no longer considered as related,
and at a point at which the error committed appears to dominate the situation,
ill suits you.

The matter of the blank ballot and that of the candidature, which have been
closely connected to each other for quite some time, cannot now be separated
without compromising the diginity of your votes.

On Monday before last, it was determined, in conformity with the law and the
bylaws of the College de France, that the question of the blank ballot assures
me the double candidacy of the College and the Academie.

But, to decide today, in a sense, to set aside an acquired right, to decide on the
basis ofa new ballot the entire question of the candidacy in a contrary sense, to
subscribe to the setting aside ofthe blank ballots, and to set against me and me
alone an exception to the general rule, to deprive me of the candidacy for the
College after having deprived me, by a fatal blunder, of the candidacy for the
Academie, this, I am sure you understand, Sirs, constitutes an obvious
miscarriage of justice, an act that is contrary to the impartiality that should
always control your deliberations.

If, against my expectation, the College should decide to separate the two
questions and to call for a new vote, I hereby declare that I will withdraw from
the fight; which is improper both for my age and for my character and would be
completely unequal in light of a generally recognized error, the unhappy effects
of which would victimize me alone. This is not the first time that I have had to
take a courageous stand against an undeserved reversal. In order to justify the
absence of my candidacy, I ask only that this letter should be appended to the
minutes of the election meeting.
 Appendix IV 357

Please be assured, Sirs, of my highest regards.

Paris, January 4, 1851

A. L. Cauchy

P.S. I beg you, Sir, to be so kind as to have this letter read to the professors at
their next meeting.

Cauchy's lithographed note, Sur l'Influence Sou vent Exercee par des
a
Circonstances Etrangeres la Science dans la Solution des Questions
qui Paraissaient Purement Scientifiques, et sur Ie Pouvoir Attribue,
dans une Election Recente a un Billet Blanc, January 6, 1851 (Arch.
Ac. Sci. January 13, 1851)
I should like here to make a few remarks on how it is that considerations
having no relation whatsoever to science have frequently played a role in the
solution of problems that would seem to be purely scientific in nature.

In 1839, a certain member of the Institut was called by the Bureau des
Longitudes to fill a vacancy that had been created by the death ofM. de Prony.
However, the person to whom I now refer had sworn, in spite of some
altogether undeserved misfortunes that are well known, to remain firmly
attached to certain principles. These he would never desert. Thus, it happened
that after several years of struggle against persons who were audacious enough
to confer the title of geometer upon him without inquiring whether or not he
had accepted the current political situation, the government simply decided to
disregard the plain letter of the law that formally declares 'The Bureau des
Longitudes shall fill its own vacancies'.

In 1843 and 1850, the chair in mathematics became vacant. Certain members
of the Academie, professors at the College de France, invited a formal petition
of one of their former instructors to apply to fill the vacancy. The most serious
minded and earnest of the candidates did not dispute his qualifications or his
claims of right, and some of them were even so magnanimous as to support
them. Nevertheless, in 1843, the government threatened to leave the vacancy
unfilled unless a candidate was chosen who had its approval.

In 1850, shortly before the election, certain changes brought about a clever
scheme: a candidate who, on this occasion, had not even presented himself to
be placed on the lists. Moreover, a blank ballot, which was cast on the first
vote, was somehow invested with a power that had been altogether denied it
under the Empire [see Bulletin des Lois, 4th series, no. 134, law 2178J; it was, to
be sure, taken as possessing a power that, until now, it has never had in the
customary procedures of the College de France. Also, one of the candidates
358 Appendix IV

then declared, to the Academie in 1843 and to the College de France in 1851,
that he would be unable to accept the conditions imposed on him. Thus it came
to pass that what at first seemed to be a purely scientific matter was given a
solution which was strongly affected by political exigencies in 1843, while in
1850 and 1851, there were the combined influences of a scheme and a blank
ballot.
 Manuscript Sources and Bibliography

I. Manuscript Sources
A. Archives Nationales

Ecole nationale des Ponts et Chaussees, 1787-1855. Organisation,

reglement, personnel, concours, eleves.
F142187 2 Ponts et Chaussees. Individual files. Cauchy file.
F1422212 Idem. Egault file.
F142263 2 Idem. Lehot file.
F143374 Mouvement des ingenieurs des Ponts et Chaussees, livre de
mouvement pour l'annee 1816.
Rapport d'une commission des Ponts et Chaussees sur Ie canal de
I'Ourcq et ses travaux au 1er janvier 1816.
F1713551 Proces-verbaux de l'assemblee des professeurs du College de
France. Minutes.
Chaires du College de France. Sciences mathematiques, physiques
et nature lies.
a
F 17 20356 Professeurs lafaculte des Sciences. Individual files. Cauchy file.
F 1b1*531 Organisation des bureaux de ministere de l'Interieur par ordre
chronologique, 1792-1811.
Recompenses honorlfzques. Demande de Legion d'honneur pour
A. L. Cauchy, 1818-1819.
AA63(167) Letter from L. F. Cauchy to Coulomb, 14 fructidor, an XI.
AJ 16 25 Scolarite de la Faculte des Sciences, 1821-1843.
AJ 16 207 Nomination de Cauchy au titre de professeur adjoint de mecanique
(1823.
Registre des proces-verbaux des assemblees des professeurs de la
Faculte des Sciences. December 1821-December 1843.
Pieces annexes, 1809-1865.
Revue de marine. Cherbourg.
359
360 Manuscript Sources and Bibliography

M.CN. CXVIII, 640, 13 Octobre 1787, contrat de mariage entre Louis-

Franfois Cauchy et Marie-Madeleine Desestre.
LXXIII, 1260,4 Avrill818, contrat de mariage entre Augustin-Louis Cauchy et
Aloi·se de Bure.

B. Archives de l' Academie des Sciences

(1) Individual files, especially those of Cauchy, Binet, Libri.
(2) Ampere's papers.
(3) G. Bertrand's collection of autographs, Cauchy file, contains several letters
in Cauchy's hand.
(4) Meeting's packets (pochettes de seance) containing many manuscripts of
the studies or notes of presentation. The numbers for the studies are those
given in the nomenclature by R. Taton (documentary Appendix, O.c., 2,
15, pp. 589-595).
January 11, 1811 M 1
July 20, 1812 M 2
May 17, 1813 M 4
January 22, 1813 M 6
December 21, 1818 M 20
August 2, 1819 M 21
September 16, 1822 M 26 & M 27
January 27, 1823 M 30
November 17, 1823 M 34
August 9, 1824 M 37
February 14, 1825 M 41
February 28, 1825 M 42
November 7, 1825 M 9
December 26, 1825 M 44
February 13, 1826 M 45
February 27, 1826 M 46 & M 47
March 20, 1826 M 48
April 10, 1826 M 49
April 17, 1826 M 50
May 8, 1826 M 51
May 29, 1826 M 52
November 13, 1826 'Sur la nature des racines de quelques equations
transcend antes', Ex. Math., 1, pp. 297-338 (O.c., 2, 6, pp. 354-400).
December 26, 1826 M 55
January 15, 1827 M 56
March 12, 1827 M 59
April 16, 1827 M 61
April 28, 1827 De la differenciation sous Ie signe, Ex. math., 2, pp. 125-140
(O.c., 2, 7, pp. 160-176).
 Manuscript Sources and Bibliography 361

July 19, 1827 M 64

October 1, 1827 M 65 sealed message
November 5, 1827 M 66
December 10, 1827 M 67
December 17, 1827 M68 and 'Sur un memoire d'Euler qui a pour a titre Nova
methodus fractiones simplices resoluendi', Ex. Math., 2, pp. 315-316 (O.c.,
2, 7, pp. 363-365).
January 22, 1828 M 69
August 18, 1828 M 71, M 71bis, M 71ter sealed message
October 6, 1828 M 72, M 73, M 74, M 75, M 78, M 79
December 22,1828 M 75, M 78
February 16, 1829 M 79
April 6, 1829 M 82
April 27, 1829 M 82
May 4, 1829 M 83
June 1, 1829 M 85
September 14, 1829 M 88
January 18, 1830 Letter from Cauchy to the president of the Academie
published by R. Taton in the Revue d'Histoire des Sciences, 1971, 24,
pp.123-148.
April 12, 1830 M 94
May 10, 1830 M 95
June 14, 1830 Observations presentees par M. Cauchy.
June 21, 1830 M 98
July 5, 1830 M 101
July 4, 1831 M 103
October, 22, 1831 'Note sur Ie versement des voitures publiques'.
March 22, 1847 Sealed message: 'Sur Ie theoreme de Fermat'.
August 11, 1856 Sealed message: 'Note sur l'integration des equations
difl"erentielles qui representent les mouvements des astres dont se compose Ie
systeme solaire'.

C. Archives du College de France

Registre des proces-verbaux de l'assemblee des professeurs.
BU. Mathematiques. Chaire de mathematiques. Biot, Libri, and Liouville
files.

D. Archives de 1':Ecole Polytechnique

Registres des proces-verbaux du Conseil d'Instruction et du Conseil de
Perfectionnement.
Cauchy file.
VI2a2 Rapports de Prony.
XII C7 Registres d'instruction.
Fonds Saint-Venant, not classified.
362 Manuscript Sources and Bibliography

E. Bibliotheque Nationale
Fourier's papers.
Ms ffr 22516, p. 88, Note sur un memoire de Cauchy relatif aux racines de
eX - 1 = 0 et a la factorisation de la fonction eX - 1.
Ms ffr 22525, p. 207, Note relative aux integrales de la theorie de la chaleur.
Reponse a des objection de Cauchy.
Ms ffr 22 529, p. 127, Brouillon de lettre de Fourier aux secretaires perpetuels
de l'Academie. Reclamation a propos des fonction reciproques de Cauchy.

F. Bibliotheque de l'Ecole Nationale des

Ponts et Chaussees
Ms 1845 Memoire sur les roues de voiture ... , 15 fevrier 1808.
Ms 1982: 1. Memoire sur les moyens de perfectionner la navigation des
rivieres ... , 1809.2. Memoire sur les ponts en pierre. 3. Second memo ire sur
les ponts en pierre. Theorie des voutes en berceau.

G. Bibliotheque de l'Institut
Ms 3710. 3 letters from Cauchy to Leverrier.
fO AA/38. 4 autographs by Augustin Cauchy.

H. Bibliotheque de la Sorbonne
Ms 1759. Formules sur la resolution des equations (1837).
Ms 1759 bis. Notes et lettres a l'abbe Moigno (1837).
Ms 1760. Exercices d'arithmetique (1838).
Ms 1761. Exponentielles et logarithmes (undated).
Ms 1762. Recherches nouvelles sur la lumiere (1836).
Ms 1786. Resolution des equations (undated).
Ms 2057. Theorie des ondes, note XVI (1815).

I. Bibliotheque de la Faculte des

Sciences de Bordeaux
Cours de G. Lespiault, ancienne cote Ms 52.

J. Bibliotheque du centre des

Fontaines a Chantilly
Moigno file. Biographie du Clerge Contemporain 10, l'abbi: Moigno, by
G. Barbier, annotated by Father Cahier and 2 letters by Cauchy.

K. Family Papers
Cauchy's personal scientific papers are now destroyed.
 Manuscript Sources and Bibliography 363

When Cauchy's wife, Alolse de Bure, died in 1863, they became the property of
the couple's eldest daughter, Alicia, who was married to Felix de l'Escalopiei,
who kept them until he died in 1909. While his estate was being settled, Honore
Champion, a bookseller, surveyed Cauchy's papers. He found a number of
books filled with notes that he had put aside. With the consent of the notary in
charge of the estate settlement, he wrote to Painleve, and later to Darboux, the
Permanent Secretary of the Academie des Sciences, asking their advice as to
what should be done with these papers (see the letters by Honore Champion in
the Cauchy file at the Academie des Sciences and also in the archives of the
Honore Champion bookstore, Arch. Nat. AQ22). Darboux seems to have
replied that the Academie would be pleased to receive any papers that had
been written by Cauchy or that had to do with Cauchy. In the meantime, the
papers remained in the family's possession, in the hands of Madame de
Leudeville, Felix de Escalopier's youngest daughter. There were two large
trunks filled with notebooks and log books that had apparently belonged to
the scholar and were filled with numbers and calculations of all kinds. Not
knowing what to do with these items, the Leudevilles sent them to the
Academie des Sciences in 1936 or 1937. The Academie, however, immediately
sent them back. The decision was then made to destroy these unwanted
documents, and accordingly, everything was burned. The only items to have
escaped the general destruction were the notebooks that are kept in the
Sorbonne Library (See H.) and two handwritten notebooks that had been given
to Madame de Pomyers by her great uncle in rememberance of her ancestors.
1. Sur la theorie des ondes (1815-1821).
2. Melanges (1836-1837).
The correspondence between A. L. Cauchy and his family (1811-1812 and
1831-1837) was rediscovered in 1989. It is in the possession of the family de
Leudeville. Those personal letters have not been used for the biography.

II. Bibliography

A. Published Works of A. L. Cauchy

We refer the reader to A. L. Cauchy, Oeuvres Completes, Paris, Gauthier-
Villars, 1882-1974 and especially to the documentary appendix of volume
15 of the 2nd series, pp. 582-611.

1. Printed Scientific Treatises and Pamphlets

Expose Somma ire d'une Methode pour Determiner a Priori Ie Nombre des
Racines Reelles Positives et Ie Nombre de Racines Reelles Negatives d'une
Equation d'un Degre Quelconque, Paris, 1813 (O.c., 2, 15, pp. 11-16).
Cours d'Analyse de l'Ecole Royale Poly technique, lere Partie, Analyse Algebri-
que, Paris, 1821 (O.c., 2, 3).
364 Manuscript Sources and Bibliography

Resume des Lerons Donnees a l'Ecole Royale Poly technique sur Ie Calcul
Iriflnitesimal, 1st vol., Paris, 1823 (O.c., 2, 4, pp. 5-261).
Memoire sur les Integrales Definies Prises entre des Limites Imaginaires, Paris,
1825 (O.c., 2, 15, pp. 41-89).
Exercices de Mathematiques, Paris, 5 vol., 1826-1830 (O.c. 2, 6, 7, 8, and 9).
Lerons sur les Applications du Calcul Iriflnitesimai ala Geometrie, Paris, 2 vol.,
1826-1828 (O.c., 2, 5, pp. 5-403).
Memoire sur l'Application du Calcul des Residus a la Solution des Problemes de
Physique Mathematique, Paris, 1827 (O.c., 2, 15, pp. 90-137).
Lerons sur Ie Calcul Differentiel, Paris, 1829 (O.c., 2, 4, pp. 263-615).
Memoire sur la Theorie de la Lumiere, Paris, 1830 (O.c., 2,2, pp. 119-133).
Memoire sur la Dispersion de la Lumiere, Paris, 1830 (O.c., 2,10, pp. 195-220).
Resumes Analytiques, Turin, 1833 (O.c., 2, 10, pp. 9-184).
Nouveaux Exercices de Mathematiques, Prague, 1836 (O.c., 2, 10, pp. 189-
464).
Sur la Resolution des Equations de Degre Quelconque, Paris, 1837 (O.c., 2, 15,
pp. 448-482).
M emoire sur une Methode Generale pour la Determination des Racines Reelles
des Equations Algebriques ou meme Transcendantes, Paris, 1837 (O.c., 2,15,
pp.483-51O).
Recueil de Memoires sur la Physique Mathematique, Paris, 1839. 1
Exercices d'Analyse et de Physique Mathematique, Paris, 4 vol., 1841-1853
(O.c., 2, 11, 12, 13, and 14).
See also the proofs ofthe beginning of the Resume des Lerons Donnees a l'Ecole
Royale Poly technique (second year).
A. L. Cauchy, Equations Differentielles Ordinaires, Cours Inedit, Fragment,
Paris, Etudes Vivantes, 1981.

2. Lithographed Scientific Pamphlets

See the list of the pamphlets published in the documentary appendix of the
Oeuvres Completes, O.c., 2, 15, pp. 586-588.

3. Scientific Memoirs, Articles, and Notes in the Following Reviews

and Academic Collections
Annales de Mathematiques Pures et Appliquees of Gergonne.
Biblioteca Italiana.
Bulletin des Sciences of the Societe Philo mati que.
Bulletin des Sciences M athematiques of Ferussac.
Comptes Rendus Hebdomadaires des Seances de l'Academie des Sciences.
Correspondance sur l'Ecole Poly technique.

lThis pamphlet contains extracts from the Comptes Rendus Hebdomadaires des Seances de
a.c., 1, 4, no. 22, p. 112, and nos. 33, 34, 35, 36, 37, 38, 39,40, and 41,
I'Academie des Sciences (see
pp. 193-311).
 Manuscript Sources and Bibliography 365

Journal de l'Ecole Poly technique.

Journal de Mathematiques Pures et Appliquees of Liouville.
Journal des Mines.
Memoires de l'Institut de France.
Memoires Presentees a l'Academie des Sciences par Divers Savants.

4. Scientific Works of F. Moigno Directly Inspired by A. L Cauchy

F. Moigno, Lefons de Calcul DijJerentiel et de Calcul Integral Redigees d'apres
les Methodes et les Oeuvres Publiees ou Inedites d'A. L. Cauchy, 1 (Calcul
DijJerentiel), 2 (Calcul Integral), and 4 (Calcul des Variations), Paris,
Bachelier, 1840, 1844, and 1861.
F. Moigno, Lefons de Mecanique Analytique Redigees Principalement d'apres
les Methodes d'A. L Cauchy et Etendues aux Travaux les plus Recents, Paris,
Bachelier, 1868.

5. N onscientific Works
See the bibliography compiled by R. Taton in the documentary appendix of
the Oeuvres Completes of A. L. Cauchy, ~.C., 2, 15, pp 606-607. This should
be completed by the lectures published in the Bulletin de l'Institut
Catholique.
'Epitre d'un mathematicien a un poete, ou la le~on d'astronomie', poem read
on January 13, 1842, Bulletin de l'Institut Catholique, 1st instalment, pp. 14-
20.
'Sur quelques prejuges contre les physiciens et les geometres', Lecture of
March 3, 1842, Bulletin de l'Institut Catholique, 1st instalment, pp. 43-49.
'Sur la recherche de la verite', lecture of April 14, 1842, Bulletin de l'Institut
Catholique, 2nd instalment, pp. 18-29.
'Motif de regrets et d'esperance', lecture of April 8, 1843, Bulletin de l'Institut
Catholique, 3rd instalment, pp. 166-171.

B. Bibliography on the Life and Work of A. L. Cauchy

A. J. C. Barre de Saint-Venant, 'Historique abrege des recherches sur la
resistance et sur l'elasticite des corps solides', in Navier, De la Resistance des
Corps Solides, 3rd ed., Paris: Dunod, 1864.
B. Belhoste, 'Le cours d'analyse de Cauchy al'Ecole Poly technique en seconde
annee', Sciences et Techniques en Perspective, 9, 1984-1985, pp. 101-178.
B. Belhoste, Cauchy, un Mathematicien Legitimiste au XIXe siecle, Paris:
Belin, 1985.
B. Belhoste and J. Liitzen, 'J. Liouville et Ie College de France', Revue
d'Histoire des Sciences, 37, 1984, pp. 255-304.
J. Ben-David, 'The rise and decline of France as a scientific center', Minerva, 8,
1970, pp. 160-179.
G. Bertier de Sauvigny, La Restauration, Paris: Flammarion, 1963.
366 Manuscript Sources and Bibliography

G. Bertier de Sauvigny, Un Type d'Ultra-royaliste, Ie Comte F. Bertier de

Sauvigny et l'Enigme de las Congregation, Paris: Les Presses Continentales,
1948.
J. Bertrand, 'La vie et les travaux du baron Cauchy, par C. A. Val son', Journal
des Savants, 1869, pp. 205-215.
J. Bertrand, Eloges Academiques, 2, Paris: Hachette, 1902, pp. 10 1-120.
G. Bigourdan, 'Le Bureau des Longitudes, son histoire et ses travaux de
l'origine (1795) a ce jours', Annuaire du Bureau des Longitudes, from 1928 to
1932.
J. B. Biot, Melanges Scientifiques et Litteraires, 3, Paris: Michel Levy, 1858,
pp.143-160.
C. A. Bjerknes, Niels-Henrik Abel, Tableau de sa Vie et de son Action
Scientifique, Paris: Gauthier-Villars, 1885.
B. Boncompagni, 'La vie et les travaux du baron Cauchy', Bulletino di
Bibliografia e di Storia delle Scienze M atematiche et Fisiche, 2, 1869, pp. 1-
102.
L. Bucciarelli and N. Dworsky, Sophie Germain: An Essay in the History of the
Elasticity, Dordrecht, Boston: D. Reidel, 1980.
J. Buchwald, 'Optics and the theory of the punctiform aether', Archive for
History of Exact Sciences, 21, 1980, pp. 245-278.
H. Burkhardt, 'Entwicklungen nach oscillierenden Funktionen und In-
tegration der Differentialgleichungen der mathematischen Physik', J ahres-
bericht der deutschen M athematiker- Vereinigung, 10, 1904-1908, especially
pp.671-745.
J. Burnichon, La Compagnie de Jesus en France, Histoire d'un Siecle, 1814-
1914,4 vol., Paris: Beauchesne, 1914-1919.
J. P. Callot, Histoire de l'Ecole Poly technique, Paris: Les Presses Modernes,
1958.
G. Castella, 'Documents inedits pour un projet de fonder une Academie
Helvetique a Fribourg en 1830', Revue d'Histoire Ecclesiastique Suisse, 21,
1927, pp. 308-313.
M. Crosland (ed.), The Emergence of Science in Western Europa, London:
Macmillan, 1976.
A. Dahan-Dalmedico, Les Recherches Algebriques de Cauchy, 3rd cycle thesis,
typewritten, Paris: EHESS, 1979.
A. Dahan-Dalmedico, 'Les travaux de Cauchy sur les substitutions. Etude de
son approche du concept de groupe', Archivefor History of Exact Sciences,
23,1980, pp.279-319.
A. Dahan-Dalmedico, 'La matMmatisation de la tMorie de l'elasticite par
A. L. Cauchy et les debats dans la physique mathematique fran<;aise (1800-
1840)" Sciences et Techniques en Perspective, 9, 1984-1985, pp. 1-100.
A. Dahan-Dalmedico, 'Etude des methodes et des "styles" de mathematis-
ation: la science de l'elasticit6', in R. Rashed (ed.), Sciences a l'Epoque de la
Revolution Franfaise. Recherches Historiques, Paris: Blanchard, 1988,
pp. 349-442.
 Manuscript Sources and Bibliography 367

J. Dhombres, Nombre, Mesure et Continuo Epistemologie et Histoire, Paris:

Nathan, 1978.
J. Dieudonne (ed.), Abrege d'Histoire des Mathematiques, 1700-1900,2 vol.,
Paris: Hermann, 1978.
P. Dugac, Sur les Fondements de l'Analyse de Cauchy a Baire, thesis,
typewritten, Paris: Universite Paris VI, 1978.
P. Dugac, 'Histoire du theoreme des accroissements finis', Archive Internat-
ionale d'Histoire des Sciences, 30, pp. 86-101, 1980.
Ch. Dupin, Discours aux Funerailles de M. Augustin Cauchy Ie 25 mai 1857,
Paris: Gauthier-Villars, 1857.
J. B. Duroselle, Les Debuts du Catholicisme Social en France (1822-1870),
Paris: PUF, 1951.
J. B. Duroselle, 'Les "Filiales" de la Congregation', Revue d'Histoire Ecclesias-
tique, 50, 1955, pp. 867-891.
H. Edwards, Fermat's Last Theorem: A Genetic Introduction to Algebraic
Number Theory, New York: Springer-Verlag, 1977.
G. Fisher, 'Cauchy and the infinitely small', Historia Mathematica, 5, 1978,
pp.313-331.
A. Foucault, La Societe de Saint- Vincent-de-Paul, Paris: Editions 'Spes', 1933.
A. Fourcy, Histoire de l'Ecole Poly technique, Paris, 1828, reed. Paris: Belin,
1977.
H. Freudenthal, 'Cauchy, Augustin-Louis', Dictionary of Scientific Biography,
3, New York: Charles Scribner's Sons, 1971, pp. 131-148.
H. Freudenthal, 'Did Cauchy plagiarize Bolzano?', Archive for History of
Exact Sciences, 7, 1971, pp. 375-392.
J. P. Garnier, Charles X, Ie Proscrit, Paris: Fayard, 1967.
C. A. Geoffroy de Grandmaison, La Congregation (1801-1830), Paris: Plon-
Nourrit, 1889.
C. Gilain, Introduction to A. L. Cauchy, Equations DifJerentielles Ordinaires,
Cours Inedit, Fragement, Paris: Etudes Vivantes, 1981.
C. Gilain, 'Cauchy et Ie cours d'analyse 11 l'Ecole Polytechnique', Bulletin de la
Societe des Amis de la Bibliotheque de l'Ecole Poly technique, 5, July 1989,
pp.3-145.
H. Gouhier, La Jeunesse d'Auguste Comte et la Formation du Positivisme, 3
vol., Paris: Vrin, 1933-1941.
J. V. Grabiner, The Origins of Cauchy's Rigorous Analysis, Cambridge
(Massachusetts): MIT Press, 1981.
I. Grattan-Guinness, 'Bolzano, Cauchy and the "New analysis" of the early
nineteenth century', Archive for History of Exact Sciences, 6, 1970, pp. 372-
400.
I. Grattan-Guinness, The Development of the Foundations of Mathematical
Analysis from Euler to Riemann, Cambridge (Massachusetts): MIT Press,
1970.
T. Guitard, 'La querelle des infiniment petits 11 l'Ecole Poly technique au XIXe
siecle', Historia scientiarium, 30, 1986, pp. 1-61.
368 Manuscript Sources and Bibliography

A. d'Hautpoul, Quatre Mois ala Cour de Prague, Paris: Plon-Nourrit, 1912.

T. Hawkins, 'Cauchy and the spectral theory of matrices', Historia Mathemat-
ica, 2, 1975, pp. 1-29.
C. C. Heyde and E. Seneta, /.-J. Bienayme. Statistical Theory Anticipated, New
York: Springer-Verlag, 1977, pp. 71-76.
A. P. Iushkevich, Michel Ostrogadski et Ie Progres de la Science au XIXe
Siecle, Paris: Palais de la Decouverte, 1966.
A. P. Iushkevich, 'The concept of function up to the middle of the 19th
century', Archive for History of Exact Sciences, 16, 1977, pp. 37-85.
P. Jourdain, 'The theory of functions with Cauchy and Gauss', Bibliotheca
Mathematica, 3rd series, 6, 1905, pp. 190-207.
P. Jourdain, 'The origin of Cauchy's conceptions of a definite integral and of
the continuity of a function', Isis, 1, 1914, pp. 661-703.
M. Jullien, 'Quelques souvenirs d'un etudiant jesuite a la Sorbonne et au
College de France, 1852-1856', Les Etudes, 127, 1911, pp. 333-348.
M. Kline, Mathematical Thoughtfrom Ancient to Modern Time, New York:
Oxford University Press, 1972.
F. R. de Lamennais, Correspondance Generale, 1-6, Paris: A. Colin, 1970-
1978.
A. E. H. Love, A Treatise on the Mathematical Theory of Elasticity, 4th ed.,
Cambridge (England): University Press, 1927.
J. Liitzen, Joseph Liouville, 1809-1882. Master of Pure and Applied M athema-
tics, New York: Springer-Verlag, 1990.
J. Mandelbaum, La Societe Philomatique de Paris de 1788 a 1835, 3rd cycle
thesis, typewritten, Paris: EHESS, 1980.
L. Menabrea, Memorie, Firenze, Giunti: G. Barbera, 1971.
J. Michelet, Journal, 1, Paris: Gallimard, 1959.
F. Moigno, Preface to A. L. Cauchy, Sept Lerons de Physique Generale, Paris:
Journal 'Les Mondes', 1868.
C. de Montalembert, Eloge de Cauchy a la Seance des Cinq Academies du 17
Aout 1857, Paris: Firmin-Didot, 1857.
J. Paguelle de Fontenay, M. Teysseyrre, sa Vie, son Oeuvre, ses Lettres, Paris:
Poussielgue, 1882.
J. Peiffer, Les Premiers Exposes Globaux de la Theorie des F onctions de
Cauchy, 3rd cycle thesis, typewritten, Paris: EHESS, 1978.
M. Pensivy, 'Jalons historiques pour une epistemologie de la serie infinie du
binome', Sciences et Techniques en Perspective, 14, 1987-1988.
J. C. Pont, La Topologie Algebrique des Origines a Poincare, Paris: PUF, 1974.
K. Rychlik, 'Sur les contacts personnels de Cauchy et de Bolzano', Revue
d'Histoire des Sciences, 15, 1962, pp. 163-164.
L. Sedillot, Les Professeurs de Mathematiques et de Physique Generale au
College de France, Rome: Imprimerie des sciences math6matiques, 1869.
H. Sinaceur, 'Cauchy et Bolzano', Revue d'Histoire des Sciences, 26, 1973,
pp.97-112.
 Manuscript Sources and Bibliography 369

F. Smithies, 'Cauchy's conception of rigor in analysis', Archive for History of

Exact Sciences, 36, 1986, pp.41-61.
P. Stackel, 'Integration durch das imaginare Gebiet', Bibliotheca Mathemat-
ica, 3rd series, 1, 1900, pp. 111-121.
D. and R. Struik, 'Cauchy and Bolzano in Prague', Isis, 11,1928, pp. 364-366.
R. Taton, 'Sur les relations scientifiques d' Augustin Cauchy et d'Evariste
Galois', Revue d'Histoire des Sciences, 24, 1971, pp. 123-148.
A. Terracini, 'Cauchy a Torino', Rendiconti del Seminario M atematico
dell'Universita e del Politecnico di Torino, 16, 1957, pp. 159-203.
I. Todhunter, A History of the Theory of Elasticity and of the Strength of
Materials from Galilei to the Present Time, 1, Cambridge (England):
University Press, 1886.
C. Truesdell, 'The rational mechanics of flexible or elastic bodies, 1638-1788',
in Euler, Opera omnia, 2, 11, sect. 2, Zurich, pp. 7-435, 1960.
C. Truesdell, Essays in the History ofMechanics, Berlin, Springer-Verlag, 1968.
C. Truesdell, 'Rapport sur Ie pli cachete n° 126, paquet presente a I'Academie
des Sciences dans la seance du ler octobre 1827 par M. Cauchy et contenant
Ie memoire 'Sur I'Equilibre et Ie mouvement interieur d'un corps solide
considere comme un systeme de molecules distinctes les unes des autres',
Comptes Rendus de l'Academie des Sciences, 291, October 13, 1980, Vie
Academique.
C. A. Valson, La Vie et les Travaux du Baron Cauchy, 2 vol., Paris: Gauthier-
Villars, 1868.
E. T. Whittaker, A History of the Theories of Aether and Electricity, London,
New York: T. Nelson, 1951.
 Abbreviations

Ac. Sci. Academie des Sciences

Arch. Hist. Ex. Sci. Archive for History of Exact Sciences
Arch. Nat. Archives Nationales
Bull. Fer. Bulletin des Sciences M athematiques of Ferussac
Bull. Phil. Bulletin des Sciences de la Societe Philomatique
c.R. Comptes rendus Hebdomadaires des Seances de
l'Academie
E.N.P.e. Ecole Nationale des Ponts et Chaussees
E.P. Ecole Poly technique
Ex. An. Phys. Math. Exercices d'Analyse et de Physique Mathematique
Ex. Math. Exercices de M athematiques
J.E.P. Journal de l'Ecole Poly technique
J.L. Journal de M athematiques Pures et Appliquees of
Liouville
Mem. Ac. Sci. Memoires de l'Academie des Sciences
M em. I nstitut Memoires de l'Institut de France
Mem. Sav. Etr. Memoires Presentes par Divers Savants Etrangers
O.c. Oeuvres Completes of A. L. Cauchy (O.c., 1 is the 1st
Series and O.c., 2 is the 2nd series)
Rev. Hist. Sci. Revue d'Histoire des Sciences

370
 Index

Note: Subjects are italicized. References to the notes are italicized.

Abel, Niels-Henrik, 55, 57, 71, 254, Barande, Joachim, 159, 161, 164, 277,
257 326-327
Academie des Sciences, 29, 36-38,40- Barthelemy-Saint-Hilaire, Jules, 226-
42,45-47,53-60,87-88,97-98, 227, 347
138-140,144,155,174,185,188, Baunard, Louis, 282
191-194, 196-198,237,271-272, Bazaine, Pierre-Dominique, 10
284, 296-302, 325-326, 332-335, Beautemps-Beaupre, Charles- Fran~ois,
342-345, 357-358, 360-361 13
Affre, Denis-Auguste, Archbishop of Beauvais, Antoine-Thibaut, 2
Paris, 182, 187 Belhoste, Bruno, v, vi, 260, 290
D'Alembert, Jean Le Rond, 69, 71 Beluze, Eugene, 294
Ampere, Andre-Marie, 10,38,41-42, Berry, Duchess of, 160-161
44-45,48,50,53-54,61-63,65, Berry, Duke of, 72, 136
72, 74, 80-83, 85, 99, 141, 171, Berthollet, Claude, 12
194-195,243,250,251-253,258, de Bertier, Ferdinand, Count, 182,282
272,279-280,340,360 Bertier de Sauvigny, Guillaume, 272,
Andrieux, Fran~ois, 10,48 282, 288
Angouleme, Duchess of, 134, 166 Bertier de Sauvigny, Louis-Benigne, 1
Angouleme, Duke of, 63, 160, 166, 172 Bertrand, Gabriel, 248, 360
Arago, Fran~ois, 39, 42, 44, 54, 64-65, Bertrand, Jospeh, 46, 73, 156, 194,
81,175-176,178,194,258,335 196, 206, 230-231, 238,251, 257,
Arbogast, Louis-Fran~ois, 69 281,284-285,287,291,295
Argand, Jean-Robert, 232 Bessel, Friedrich, 177
Auger, Jean-Baptiste-Armand, 250 Beudant, Fran~ois, 180,350
Avogadro, Amedeo, 154 Bidone, Giorgio, 157, 276
Bienayme, Irenee-Jules, 237, 279
Biggs, Norman, 246
Bachelier, Jean-Jacques, 7 Billot,I61-164
Bader, Luigi, 280 Bigourdan, Guillaume, 248, 281, 283,
Bagge, Dominique, 271 289
Bailly, Etienne-Joseph, 141, 181 Binet, Jacques, 16,37-38,41-42,44-
Bailly, Jean-Sylvain, 1 45, 54, 62-63, 65, 81, 83, 134, 141,
371
372 Index

180, 185, 195, 197, 225-226, 236, 186, 204-206, 223, 281
238-239, 247, 254, 271, 284, 295, Bure, Jean-Jacques de, 132, 134
333, 336, 347-349, 354, 360 Bure, Marie-Jacques de, 132, 134, 271
Binet, Paul, 10, 61 Bure, Guillaume de, 132
Biot, Jean-Baptiste, 9, 28-29, 42, 48- Bure, Marguerite de, 134
49,54,83,175-176,192-194,200, Burkhardt, Heinrich, 260, 275
212,242,248,252,281,283,284,
286,288, 341, 361 Cachin, Joseph, 19-22,246
Bjerknes, Carl Anton, 231, 291 Cagniard-Latour, Charles, 42
Blacas, Pierre-Louis-Jean-Casimir, Cahier, Charles, Father, 362
Duke of, 161, 172 Calculus of algebraic keys, 215, 236-
Blanchet, Pierre-Henri, 203-204, 286 237,293
Bolzano, Bernard, 255, 280 Calculus of finite differences, 62, 178,
Bonald, Louis de, Viscount, 136 304
Bonald, Louis-Jacques-Maurice de, Calculus of indices of function, 153-
Archbishop of Lyon, 182 154, 292
Boncompagni, Baldassare, 275 Calculus of limits (method of
Bonnard, Augustin-Henri de, 41 majorants), 151-153, 166-168, 172,
Bonnet, Ossian, 196 199, 204-205, 207, 215, 325
Borchardt, Carl Wilhelm, 227 Calculus of probabilities, 219
Bordeaux, Duke of, 143, 159-166, 172, generating functions, 51
223, 277, 326 probability of testimonies, 51
Bosc, Louis, 41 Calculus of residues, ix, 39, 60, 122-
Bosq uet, Pierre-Joseph -F ran<;:ois, 131, 154,203,205,207,209,215,
Marshall, 238 221, 227, 232, 236, 269, 275, 292,
Bossut, Charles, 14, 40 301, 324
Bouchu, Fran<;:ois-Louis, Baron, 71- Calculus of substitutions, 32, 34, 199,
72 206-207, 215, 247, 287
Boulanger, Father, 179 Carlo Alberto, King of Sardinia, 150,
Bouquet, Jean-Claude, 228, 231, 234, 154-156, 327-328
237 Carlo Felice, King of Sardinia, 148-
Bourdier-Delpuits, Jean-Baptiste, 149, 154
Father, 14, 16 Carnot, Lazare, 28, 46, 69, 238, 248
Boussinesq, Joseph, 284 Caron, Jeanne, 282
Bravais, Auguste, 229 Castella, Georges, 273
Bredin, 250-252 Catalan, Eugene, 331
Breguet, Louis, 42, 46 Cauchy, Alexandre-Laurent, 3, 8, 31,
Brewster, David, 105, 171, 200 134-135,146, 148-149,239,242-
Briffault, Eugene, 289 243,273
Briot, Charles, 228, 231, 234, 237 Cauchy, Alolse, 132-134, 164,273,
Brisson, Barnabe, 51-54, 122-123, 277, 295, 363
126-127,268,299,301-302 Cauchy, Amedee, 273
Brunot, Andre, 245 Cauchy, Augustin-Louis,
Brute de Remur, Gabriel-Simon, 271 birth and family, 1-4
Bruyere, Louis, 12 charitable works, 135-136, 179-183,
Bucciarelli, Louis, 261 237-238
Buchwald, Jed Z., 264 death, 239-240
Bude, 232 education, 5-17
Bureau des liJngitudes, 36, 174-177, engineering, 12-31, 35-40
 Index 373

final years, 173-249 210-212,232,287

instructor, 44-86 theory of geometric quantities, 232,
marriage, 132-135 236,293
religious controversy, 136-142 Comte, Auguste, 213, 254
sojourn abroad, 144-173 Congregation, 14, 16-17, 63, 135, 140-
Cauchy, Eugene-Francois, 8, 135, 148, 142, 181, 242, 245, 271-272
243, 276, 295 Continuity, 65, 68-71,88-90,255-256,
Cauchy inequalities, 152, 168 303,309,316-317
Cauchy integral formula, 115, 151-152 definition, 63, 68-69, 255
Cauchy integral theorem, 109 117, 208 uniform continuity, 68, 77
Cauchy, Louis-Francois 1-8, 16,20, Coquand, Roger, 245
29-30, 39-41,43, 132, 134-135, Coriolis, Gustave-Gaspard, 61, 105,
148,224,230,241-245,247,249- 158, 180, 194, 198, 254, 274
250, 359-360 Cossa, Giuseppe, 150
Cauchy, Marie-Francois-Alicia, 134, Costaz, Louis, Baron, 30, 39, 249
295, 332, 363 Cotes, theorem of, 303, 306, 310
Cauchy, Marie-Madeleine, 2, 31, 172, Coue, Father, 295
241-243, 247, 360 Coulomb, Charles-Augustin, 14, 92,
Cauchy, Marie-Mathilde, 134 359
Cauchy, Marie-Therese, 243 Couplet, Claude-Antoine, 14
Cauchy's Turin theorem (on analytical Cousin, Victor, 176
functions), 71, 152, 174, 204-205, Cropelet, 144
207,224,274-275 Crosne, Louis Thiroux de, 1-2
Cauchy-Kovalevskaya theorem, 205 Cruveilher, Jean, 180
Cauchy-Riemann equations, 108-109 Cuvier, Georges, 7, 41
Cayol, Jean-Bruno, 180 Cyclotomic polynomials, 211-212
Champion, Honore, 363
Charles X, King of France, 136, 143, Daguerre, Jacques, 335
148-149, 159-162,223,337 Dahan-Dalmedico, Amy, x, 247, 254,
Chateaubriand, Fram;ois-Rene de, 161 259, 261, 287
Chevalier, Alexis, 294 de Damas d' Anlezy, Count, 277
Chladni, Ernst Florens Dietrich, 92 de Damas, Maxence, Baron, 159-160,
Clapeyron, Emile, 55 162-163, 277, 326-327
Clarion, Jacques, 149 Dambray, Charles-Henri, Viscount,
Clausel de Coussergues, Michel- 41,134
Amans, Abbe, 136 Darboux, Gaston, 292, 363
Clouet, Anne-Louis-Antoine, Baron, Debauve, Alphonse, 244
164 Dejean, Pierre-Francois-Marie, Count,
College de France, 48-49, 105, 113, 250-251
122, 127, 131, 184-186,225-227, Delambre, Jean-Baptiste-Joseph, 246,
251-252, 265, 266, 269, 282-283, 248
284, 290, 338-358, 359, 361 Delaunay, Charles, 224
Collegno, de, 326, 328 Delaunay, Dr., 241, 273
Collet-Descotils, Hippolyte-Victor, 41 Delavenne, Andre, 271
Combes, Charles, 64, 198 Delozanne, 247
Complex numbers, definition De Moivre's theorem, 62, 310
theory of symbolic expressions, 71, Demangeon, Alain, 245
210,232 Denoel, FranCois, 38
theory of algebraic equivalences, Deplace, Etienne, Father, 161
374 Index

Derivatives, 74, 77, 79, 306 92, 113, 131, 144, 146, 243~244,
derivative of a real function, 86, 122 250~252, 254,260
derivative of a complex function, 77, Edwards, Harold, 139, 288
86, 234 Egault, Pierre-Marie-Thomas, 12~13,
Desains, Paul, 229, 290 244
Descartes, Rene, 37 Eigenvalue problem, 83, 95, 104~ 105,
Desestre, Jean-Baptiste, 3 259
Desestre, Louis-Jacques, 3 Eisenmann, Joseph, 11
Desgarets, Nicolas-Jean, 183 Elasticity, 50, 60, 92, 96~97, 198, 215
Desgranges, Alix, 348~349, 352 continuum theory, 92~ 103, 106, 198,
Desjardins, Philippe-Jean-Louis, 215, 261 ~262, 265
Abbe, 134 molecular theory, 93, 98~ 103, 106, 156,
Destainville, Nicolas, 61 263~264
Determinants, 34~35, 88, 236, 247, 259, anisotropic media, crystals, 101 ~ 103,
308, 317 264
Devernon, Simon-Francois Gay, 243 isotropic media, 96, 98~99, 102~103,
Didot, Pierre, 132, 134 105~106, 169, 171, 229~230, 262~
Differentials, 76~77, 304, 306, 31O~31l, 263
318~319 Equations,
Differential geometry, 82~83, 259 application of the calculus of
Dinet, Charles, 9, 134, 243 residues, 130~ 131
Dirichlet, Gustav, Lejeune-, 131,271, number of roots, 37~39, 107, 114,
284, 331 153~ 154, 172, 195, 207, 306, 329,
Druilhet, Julien, Father, 161 ~ 162 335
Doubly periodic functions, 209, 227, fundamental theorem of Algebra, 63,
232~233 87,256,309
Dubuat, Pierre-Louis-Georges, Count, Sturm theorem, 154, 329, 335
336 Error theory, 40, 152, 169, 326
Dufriche-Desgenettes, Charles- I'Escalopier, Felix de, 295, 363
Elenore, Abbe, 136, 142 I'Escarene, Count of, 162, 326~328
Dugac, Pierre, x, 293 Ettlinger, H. J., 266
Duhamel, Jean-Marie-Constant, 55, Euclid, 26, 29, 66, 336
203, 237~238, 295 Euler, Leonard, vii, ix, 25~26, 64, 69,
Duhiiys, Marie-Charles, 10 77, 79~80, 89, 92, 97, 106~ 107,
Duleau, Alphonse, 92 210, 212, 215, 246, 257, 361
Dumas, Jean-Baptiste, 285, 342, 345
Dupin, Charles, 140, 219, 257 Faculte des Sciences of Paris, 48, 50,
Durand, Jean-Nicolas-Louis, 10 92, 120, 146, 224, 228, 230~237,
Durivau, Henri, 45, 250 252~253, 291~292, 336~337
Duroselle, Jean-Baptiste, 271 ~272, 282 Falloux, Frederic-Albert, de, 224, 238,
Duvernoy, Georges-Louis, 226 242, 281
Duvillard de Durand, Emmanuel- Faye, Herve, 228, 290
Etienne, 37~38, 41 Fermat, Pierre, de, 29, 32, 46, 210,
Dworsky, Nancy, 261 212, 288, 361
Fermat's last theorem, 46, 21O~2t2,
Ecole des Ponts et Chaussees, 12~ 14, 288
18, 35, 247, 362 Fermat's theorem on polygonal
Ecole Poly technique, ix, 8~ 11, 14~ 15, numbers, 29, 32, 46
41, 44~48, 50~51, 53~54, 61 ~86, Ferussac, Andre-Etienne de, 146, 364
 Index 375

First Class of the Institut, see Garnier, Jean-Guillaume, 10

Academie des Sciences Gaultier de Claubry, Henri-Francois,
Flesselles, Jacques de, 1 149, 180
Flourens, Pierre, 139 Gauss, Carl Friedrich, vii, ix, 32, 34,
Ferdinand, Emperor of Austria, 172 46, 64, 177, 210, 222, 240, 247
Fontanes, Louis de,S, 242 Gengembre, 42
Foresta, Marquis de, 161, 272 Genoude, Eugene de, 134
Fortier, Bruno, 245 Geoffroy de Grandmaison, Charles-
Fortoul, Hippolyte, 228 Alexandre, 245, 272
Fouche, Joseph, 43 Geoffroy Saint-Hilaire, Etienne, 138
Foullon, Joseph-Fram;:ois, 1 Gerbet, Philippe-Olympe, Abbe, 271,
Fouques-Duparc, Louis-Benoit, 22, 273
246 Gergonne, Joseph-Diez, 128,247,364
Fourcroy, Antoine-Francois, 2, 10, 51, Germain, Sophie, 46, 93, 158,261,276
54 Gilain, Christian, 255, 258, 364
Fourcy, Ambroise, 10, 243, 288 Gillis, James, Bishop of Edinburgh,
Fourier, Joseph, 54, 98, 113-114, 154, 182-183
157-158,194-195,262,267,328, Girard, Pierre-Simon, 12-13,35-36,
362 42,92,244-245
Fourier-series expansion, 119-121, 131, Gloria, Count, 154
170 Gontaut, Madame de, 161
Fourier transform, ix, 89-90, 92, 112- Gosin, Eugene, 283
113,119-120,130,202,299 Gossin, Jules, 188-189,283
Fourneyron, Benoit, 98 Gouhier, Henri, 251, 254
Francis II, Emperor of Austria, 160- Grabiner, Judith V., 256
161 Gramont, Antoine-Louis-Marie, Duke
Francoeur, Louis-Benjamin, 10,41 of, 161
Francois de Neufchateau, Nicolas- Grassi, Father, 328
Louis, 10 Grassman, Hermann, 232, 236-237,
Franqueville, Hippolyte de, 22, 246 293-294
Fraunhofer, Joseph von, 169 Grattan-Guinness, Ivor, x, 255, 257,
Frayssinous, Denis-Antoine, 274
Monseigneur de, 136, 141, 162- Gregory XVI, Pope, 155
164,253 Group theory, 32-35
Freddoni,324 Lagrange theorem, 32, 34
Fresnel, Augustin, 53, 92, 94, 98-99, theorem of Cauchy, 32, 34-35
103, 105-106, 171-172, 200, 248, Guignon, Pierre-Louis, 243
262, 264 Guitard, Thierry, 255
Freudenthal, Hans, 97, 255, 262, 275, Guizot, Francois, 183, 187,275
294 Guyon, Mayor of Sceaux, 294-295
Freycinet, Charles, de, 180 Guyton-Moreau, Louis-Bernard, 10,
Frizon,57 45
Function,
definition, 63, 68, 303, 306-307, 316 Hachette, Jean-Nicolas-Pierre, 10-11,
42
Gagarin, Jean-Xavier, Father, 238 Haller, Charles-Louis, 149, 216
Gall, Franz Joseph, 138, 140,271-272 Hamilton, William Rowan, 168, 278
Galois, Evariste, viii, 55, 57-58, 60, Hansen, Peter Andreas, 177
157-158, 186, 207, 254, 273, 283 Hassenfratz, Jean-Henri, 10, 45
376 Index

d'Hautpoul, Alphonse, General, 162- 240,288,290,293-294

163, 165, 258, 277-278
Kronecker, Leopold, 240
Hawkins, Thomas, 259
Kummer, Ernst Eduard, 211-212
Hennequin, Antonie-Louis-Marie, 16
Henri, 41 Labey, Jean-Baptiste, 44
Henri IV, King of France, 337 Laboulaye, Edouard de, 353
Henry, Charles, 276 La Bourdonnaye,Fran\!ois-Regis,
Hermite, Charles, 186,207,225,231, Count of, 143
234, 237-238, 292 Lacepede, Bernard-Germain-Etienne,
Herschel, John, 335 Count of, 7
Heyde, C. c., 279 La Cheze or La Chaise, Father, 157,
Hoefer, Ferdinand, 281 276, 320, 328
Holmboe, Bernst Michael, 56, 254 Lacombe, Hilaire de, 294
Hooke's law, 96 Lacordaire, Henri-Dominique, Father,
Houzel, Christian, 257, 292 187
d'Horrer, Marie-Joseph, 149 Lacroix, Sylvestre-Fran\!ois, 10, 40, 45,
Huygens, Christiaan, 104-105 48, 79, 175, 184, 194, 243, 257-
258,338-341,343-344
Ignatius of Loyola, 184 Lacuee, Jean-Girard General, 244
Imaginary exponentials, 62, 114 Laennec, Rene, 16
Imaginary logarithms, 208 La Fayette, Gilbert de, 22, 143
Implicit function theorem, 204 Laffitte, Jacques, 143
I1ifinitesimals, 62, 65-66, 68-70, 73- Laffitte, Pierre, 250, 254
74, 82-83, 85, 94, 199, 230, 255, Lagrange, Joseph-Louis, vii, viii, 6-7,
258, 306 18,25,32,34,36-38,42,46,54,
Integral, 40, 51 64, 74, 79-80, 85, 88-89, 93, 151,
extraordinary integral, 111, 130, 269 212,215,220,257-258,260,275,
imaginary integral, 78, 114-115, 299, 340
207-210, 269 Lamarle, Anatole, 207-208, 287
integral of a real function, definition, Lamarque, Maximilien, General, 155
107-108, 110-111, 115, 122-124, Lame, Gabriel, 46,55, 186, 195, 210-
310 211,236-237,284,288,333-334,
integral of a complex function, 345
definition, 107-108, 111, 113-114, Lamennais, Felicite-Robert de, 16,48,
117, 126, 128 135-137,142,187,271-273
singular integral, 39, 78, 109-116, Laplace, Charles-Emile, 9
118-119, 122, 126-128, 221, 258, Laplace, Pierre-Simon, vii, 5-6, 9, 12,
267,269,301-302 18, 25,40,42, 44-45, 48, 50-51,
Integral mean-value theorem, 77 54, 64, 79-80, 83-84, 87-90, 99,
Intermediate-value theorem, 63, 69 106-108,111,113-114,119,134,
Iushkevich, Adolf P., x, 254, 257, 269 151, 177, 215, 220, 248, 256, 258,
266, 302, 340
Jacobi, Karl Gustav Jakob, 64 La Provostaye, Frederic-Herve de,
Jamin, Jules, 229 229, 290
Janvier, Antide, 42 Larsonnier, Laurent, 2
Joachimsthal, Ferdinand, 227 Latil, Jean-Baptiste, Cardinal, 150
JoufTroy, Achille, Marquis de, 179 Latour-Maubourg, Marie-Charles-
Jourdain, Philip E. B., 256 Cesar Fay, Count of, 21-22, 162,
Jullien, Michel, 216, 229, 233, 236, 246,328
 Index 377

Laurent, Pierre-Alphonse, 196, 205, Louis-Philippe, King of the French,

229, 290 143, 149, 160, 223, 289, 337
Laurentie, Pierre-Sebastien, 136 Love, Augustus Edward Hough, 261
Lavigerie, Charles Martial, Cardinal, Liitzen, Jesper, x, 290
238, 294
Lebreton, Theodore, 2, 24, 250 MacLaurin formula, see Taylor and
Lefebure de Fourcy, Etienne, 45, 47, MacLaurin formulas
250 Magendie, Franc;ois, 139
Legendre, Adrien-Marie, vii, 26, 28- Magnin, Charles, 134, 271
29, 36, 40, 46, 57, 107, 111, 154, Maherault, Jean-Franc;ois-Regis, 7
194,248,251,266,340 Maistre, Joseph de, 137
Legris-Duval, Rene-Michel, Father, Malus, Etienne-Louis, 26, 243
134-135 Mandar, Charles-Franc;ois, 12
Lehot, Charles, 35-36, 248 Mandaroux-Vertamy, Jean-Baptiste-
Leibniz, Gottfried Wilhelm, 62, 123, Julien, 238
221 Mandelbaum, J., 247, 250-251, 266
Lenoir, Etienne, 42 Marechal, Christian, 251
Lenormant, Charles, 238 Marielle, Charles-Philippe, 251
Leon XIII, 294 Marie-Louise, Empress of France, 21
Lepelletier, 41 Marmont, Auguste-Frederic-Louis
Leroy, Charles-Franc;ois, 149, 178, 180 Viesse de, 143
Lespiault, Georges, 231, 291, 362 Martignac, Jean-Baptiste-Sievere Gay,
Letronne, Jean-Antoine, 226, 283, Count of, 141
338-341, 347-348, 350, 354 Mathieu, Claude-Louis, 14, 36, 238
de Leudeville, 247, 363 Mean-value theorem, 68, 77,116,118,
Leveque, Pierre, 40 152, 204
Leverrier, Urbain, 177, 196, 198, 205- Means of several quantities, 63, 66, 68,
206,224,228,286,290,334,362 307-308, 316
Lexell, Anders Johan, 286 Mechanics,
Lhuilier, Simon, 69 continuum mechanics, see Elasticity,
Libri, Guillaume, 184-186, 225, 227, continuum theory
271,279-280,282-283,321,323- fluid mechanics, 50, 87-94, 97, 99,
324, 330, 341-343, 345, 360-361 102, 260-262, 297-298, 302
Limit concept, 66-70, 74 solid mechanics, 50, 64, 94-95, 259-
Liouville, Joseph, 55, 175, 177-178, 260
184-186, 195-196,203-204,207, Melun, Armand, de, 238, 294
209,211-212,225-227,231,233, Menabrea, Louis-Frederic, 156, 276
282-284,287,290,292, 328, 331, Menjaud, Camille, 149
338,341-342,345,348-355,361, Meray, Charles, 231,236,240,293
365 Mersenne, Marin, Father, 46
Liouville theorem (on entire functions), Meynis, Dominique, 282
209 Meyraux, Pierre-Stanislas, 149
Livet, Jean, 10 Michelet, Jules, 183, 185, 283, 241
Lloyd, E. Keith, 246 Mickiewicz, Adam, 183
Lorion, Andre, 247 Mobius, August Ferdinand, 236
Louis XIV, King of France, 337 Modena, Duke of, 149
Louis XVI, King of France, 19,318 Moigno, Franc;ois-Napoleon-Marie,
Louis XVIII, King of France, 43, 132, 157, 164, 172, 177, 179, 185, 192,
136, 149 252-253, 260, 273, 276-278, 280,
378 Index

282,283,288,290,328,330-331, Painleve, Paul, 363

343, 365 Pardessus, Jean-Marie, 180, 188
Molard, Pierre, 42 Parieu, Marie-Louis Esquirou de,
Mole, Louis-Mathieu, Count, 18, 20, 226-227, 352, 355
36,244-245,247,249 Parseval-Deschenes, Marc-Antoine,
Monge, Gaspard, 10-11, 45-47, 54, 38,41
72, 245 Partial differential equations
Monod, Gabriel, 282 application of the Fourier
Montalembert, Charles, Forbes, Count transforms, see Fourier transforms
of, 142, 179, 187, 238 first order partial differential
Montbel, Guillaume-Isidore Baron, equations, 64, 304-305, 315
Count of, 164 linear partial differential equations,
Montlosier, Francois-Dominique de 49, 51, 92, 103, 107, 113, 115, 119·
Reynaud, Count of, 140 123, 130, 168,200-203,267-268,
Montmorency, Mathieu, de, 135-136 296-297, 316
Morin, Arthur, 198, 238, 295 Paupelin, Madeleine, 3
Multiformfunctions, 207-210, 233 Pearson, Karl, 261
indices of periodicity, 209-210 Peiffer, Jeanne, 274, 292
PeJouze, Jules, 348
Napoleon I, viii, 5, 7, 12, 16, 21, 40, Pelseneer, Jean, 250
42-43, 149,241,349,394 Permutations, see calculus of
Napoleon III, 228 substitutions
Navier, Claude-Louis, 42, 93, 97-99, Perreau, Abbe, 136
102, 194,261-263 Perturbation function, 153, 157,204-
Necker, Jacques, 1 206
Neuschwander, Erwin, 290 Petit, Alexis, 44-45, 65, 215
Neveu, Francois-Marie, 10 Picard, Emile, 247, 292
Newton, Isaac, 71, 138 Pierron, 4
Nicholas I, Czar of Russia, 149,237 Pius IX, 294
Noailles, Alexis de, 135 Plana, Jean, 151, 157, 276
Noel, Jean, 313 Poinsot, Louis, 25, 37-38,44--45,48,
54,61,73-74, 186, 194,250-251,
254,257,284,292
d'Olry, 326, 328 Poisson, Simeon-Denis, 29, 36-37,
Ordinary differential equations, 64, 68, 39-42,44,50-54,80,83-84,88,
78-81, 103-104, 108-109, 167- 90,93,99-100,102-103,106-107,
169, 278, 304-306, 313-315 113-114, 120, 122-123, 175, 177,
application of the calculus of 194,248,256,261-262,264,266-
residues, 121-123, 130-131,201, 268,281,298,329
209-210,234 Polignac, Auguste Jules Armand
Cauchy's problem, method of finite Marie, Count of, 134, 142-143
differences, 80-81 Polyhedra, 25-29, 37, 246
linear differential equations, 87, 95, Euclid's theorem on rigid
121-123, 130,200-201,252,315 polyhedra, 26, 29
Orleans, Duke of, see Louis-Phillippe Euler formula on polyhedra, 25-
Ostrogradski, Mikhail, 55, 60, 123- 26,64,80
124, 126, 131, 238, 254, 269, 302 regular polyhedra, 25
Ovaert, Jean-Louis, 257 Pomyers, Madame de, 261-262, 279--
Ozanam, Frederic, 179, 181-182,282 280, 297, 330, 363
 Index 379

Poncelet, Jean-Victor, 55-56, 58, 65, Roothan, Joannes, Father, 183, 282
195, 238, 254, 284, 295 Rossi, Pellegrino, Count, 344
Ponlevoy, Armand de, Father, 283 Rouche,55
Pont, Jean-Claude, 246 Rouge, Olivier-Charles-Camille-
Portalis, Joseph Marie, Count, 188 Emmanuel, Viscount of, 238
Primitive, 78-79, 109-110, 123,221, Ruffini, Paolo, 32, 253
255 Rychlik, Karel, 280
Principal function of a linear
differential system, 201-203 Saint-Hilaire, Auguste, de, 180
Principal value of an integral, 110-111, Saint-Hilaire, Barthelemy, see
115-118,221 Barthelemy-Saint-Hilaire
Prony, Marie-Riche de, 10-11, 14,35- Saint-Pol, Alfred de, 295
36, 54, 80, 83, 86, 174-175, 194, Saint-Venant, Adhemar Barre de,
262, 357, 361 195-196,198,229,236-237,261,
Prosper, 247 284, 287, 290, 292-294, 332-333,
Puiseux, Victor, 210, 231, 234, 237, 336
287 Saliere, Clementine Blanchet de la,
Puissant, Louis, 38, 41, 83 243
Salinis, Louis-Antoine de, Abbe, 136,
Quadrics, 83, 85, 95-96, 259 141-142
Quatremere, Etienne-Marc, 226, 347- Sansuc, Jean-Jacques, 257
349, 354 Sardinia, King of, see Carlo Alberto
Quincerot, Hippolyte d'Haranguier or Carlo Felice, King of Sardinia
de, 149 Saugrain, 132
Quinet, Edgar, 183-184 Savart, Felix, 106,325,332
Say, Jean-Baptiste, 344
Ravignan, Gustave-Xavier de, Father, Scarpa, Antonio, 344
187, 283 Sedillot, Louis-Pierre-Eugene-Amelie,
Raymond, Pierre, 257 347
Recamier, Joseph, 149, 180 Seguier, Antoine-Jean-Mathieu,
Reciprocal functions (inversion Baron, 144
formulas), 112,267,313, see also Semonville, Charles-Louis-Henri,
Fourier transform Marquis of, 40, 134
Rendu Ambroise, 179 Seneta, Eugene, 279
Residue theorem, 111, 114, 118-119, Senfft de, 271
122, 127-128, 130, 152-153,208- Series,
209 definition, 70
Reynaud, Antoine, 44-45, 134 convergence, 50, 62-64, 68, 70, 80,
Riancey, Henri de, 189,283 131,151-152,256,292,304,317
Riccati, equation of, 315 convergence tests, 70-71, 304, 317
Richerand, Marie Anthelmine, 243 imaginary series, 50, 310, 318
Riemann, Bernhard, 240 Laurent series, 205, 234
Rigor in analysis, 64, 66, 70-71, 80, Taylor series, 64, 79, 170, 234, 304,
128,215-216,220-222,256 306, 311, 319
Riviere, Charles-Francois de Serres, Augustin, 138
Riffardeau, Duke of, 136 Sganzin, Joseph-Mathieu, 10
Robespierre, Maximilien, 4 Silliman, Robert H., 264
Rochette, Raoul, 180 Sinaceur, Hourya, 255, 280
Roos, Prof., x, 291 Smithies, F., 257
380 Index

Societe philomatique, 28-29, 41-42, 97, reflection and simple refraction, 103,
247, 250-251, 262 105-106,171,200,229,291,298
Souton, Jean-Baptiste, 138 rotatory polarization, 193, 200
Speziali, Pierre, 284 Theory of vaulted arches, 14, 246
Statistics, 219, 272, 279 Thiers, Adolphe, 143
Stendhal, 139-140, 272 Thillaye, 41
Strain, 94-96, 98 Thouret, Michel-Augustin, 2
Strength of materials, 92-93 Timoshenko, Stephen, 261
Stress, 93-96, 98, 102, 261, 332-333 Tocqueville, Alexis de, 188
Struik, D. J., 280 Tortolini, Bamaba, 155
Struik, Ruth, 280 Trebuquet, Stanislas-Bamabe, Abbe,
Sturm, Charles, 55, 154, 175, 185-186, 164
195, 203, 259, 283-284, 328-329, Truesdell, Clifford, 261, 263
335, 345
Symmetric functions, 32, 34-36
Vaillant, Jean-Baptiste-Philibert,
Marshall, 228
Talbot, William Henry Fox, 335
Vallot, Simon, 22, 246
Taton, Rene, x, 58, 243, 254, 268, 273-
Valson, Claude-Alphonse, vii, viii, 3,
274, 284, 360-361, 365
6-7,20,29-30,53,134, 145, 175,
Taylor, Brook, 234, 292, 299
188,229,231,239-246.249-250,
Taylor's formula, see Taylor and
256,271,273,275-277,
MacLaurin formulas
280-284,286,289-295
Taylor and MacLaurin formulas, 78-
Vandermonde, Alexandre, 34
80,85-86,121,151-152,170
Vanneau, Louis, 144
Lagrange remainder, 78, 80, 85
Vemet, Joseph, 24
Cauchy remainder, 79, 86
Vico, Franco, de, 335
Terracini, Alessandro, 273-276, 278,
Villele, Jean-Baptiste-Guillaume-
329
Joseph, Count of, 136, 141
Tessier, Alexandre-Henri, 180
Villemain, Abel-Fran90is, 250
Teysseyre, Paul-Emile, 10, 14, 16,
Villeneuve, Pons-Louis-Fran90is,
134-135, 271
Marquis of, 166,278
Thenard, Jacques, 178, 341
Villeneuve-Bargemon, Jean-Paul-
Theory of functions, 131,231,234-235
Alban, Viscount of, 188
monodromic functions, 234
Villerme, Louis-Rene 188
monogenic functions, 234
Vitry,242
monotypical functions, 234
Voltaire, 138-139
synectical functions, 234
Theory of light, ix, 49, 54, 60, 87,
103-106,131,167,169-172,193, Wailly, de, Natalis, 238
199, 203, 228-230, 254, 264, 279- Wall on, Henri-Alexandre, 238
280, 300, 335 Wantzel, Frederic, 211
dispersion, 105-106, 169-171 Weierstrass, Karl 240
double refraction, 103, 105 Whittaker, Edmund Taylor, 264
polarization, 103, 105, 172, 200, 203, Wilson, Robin J., 246
280, 286 Woisard, Jean-Louis, 55