再别康桥原文-学生会宣传部

Alexandre-Théophile Vandermonde

Born: 28 Feb 1735 in Paris, France
Died: 1 Jan 1796 in Paris, France

Article by:
J J O'Connor
and
E F Robertson

March 2001
Edited by XiaJingbo, mail to: xjb@

Alexandre- Theophile Vandermonde's father was a medical doctor who was originally from
Landrices but had spent 12 years in the Orient. He had set up a medical practice in Paris and was
working there as a doctor when his son Alexandre-Théophile was born. He did not encourage his son to
follow a medical profession but rather encouraged him to take up a career in music. Certainly he was not
interested in mathematics when he was young. Alexandre-Théophile was awarded his bachelier on 7
September 1755 and his licencie on 7 September 1757.
His first love was music and his instrument was the violin. He pursued a music career and he only
turned to mathematics when he was 35 years old. It was Fontaine des Bertins whose enthusiasm for
mathematics rubbed off on Vandermonde. Perhaps surprisingly he was elected to the Académie des
Sciences in 1771 with little evidence of his mathematical genius other than his first paper which,
although he was not a member at the time, was read to the Academy in November 1770. However, he did
make quite a remarkable contribution to mathematics in this paper and three further papers which he
presented to the Academy between 1771 and 1772. These four papers represent his total mathematical
output and we will discuss their content below together with the views of a number of historians of
mathematics on his contribution.
Vandermonde's election to the Académie des Sciences did motivate him to work hard for the
Academy and to publish other works on science and music. In 1777 he published the results of
experiments he had carried out with Bézout and the chemist Lavoisier on low temperatures, in
particular investigating the effects of a very severe frost which had occurred in 1776. Ten years later
he published two papers on manufacturing steel, this time joint work with Monge and Bertholet. The aim

of this research was to improve the steel used for bayonets but experimenting with different mixtures
of iron and carbon. That he work closely with Monge reflected the fact that the two were very close
friends, in fact he so close that he was known as
femme de Monge.

In 1778 Vandermonde presented the first of a two part work on the theory of music to the
Académie des Sciences. The second part was presented two years later. This work
Système d'harmonie
applicable à l'état actuel de la musique
did not propose a mathematical theory of music as one might
have expected from someone who was an expert in both fields. On the contrary the aim of the work was
to put forward the idea that musicians should ignore all theory of music and rely solely on their trained
ears when judging music. As one might expect this proved a controversial work with musicians being
sharply divided as to whether they agreed with Vandermonde or not. Despite the opposition of many
musicians at first, the ideas put forward by Vandermonde gained favour over the years and by the
beginning of the nineteenth century the Académie des Sciences had moved music from the
mathematical area to the arts area. It is worth repeating that it is strange that a mathematician of the
highest rank should have argued against music as a mathematical art, a position it had held since the
days of ancient Greece.
Positions which Vandermonde held include director of the Conservatoire des Arts et Métiers in
1782 and chief of the Bureau de l'Habillement des Armées in 1792. In the same year of 1792 he sat
with Lagrange on a committee of the Académie des Sciences which had to examine the violon
harmonique, a newly invented musical instrument. He was involved with the École Normale, which was
founded in October 1794, and was on the team designing a course in political economy. His friend Monge
was also involved with the École Normale as were Lagrange and Laplace. However the establishment only
operated for six months after it opened in the Muséum d'Histoire Naturelle in January 1795 before
being closed down.
Like Monge, Vandermonde was a strong supporter of the Revolution which began with the storming
of the Bastille on 14 July 1789. The politics of Revolution in France long before this event had been so
exciting for Vandermonde that it diverted him from a possible longer mathematical and scientific
career. However the truth of the matter is that he suffered from poor health all his life and, but for
this, he might well have been able to be highly involved in politics yet continue with mathematical and
scientific activities.
Perhaps the name of Vandermonde is best known today for the Vandermonde determinant. While it
is certainly true that he made a major contribution to the theory of determinants, yet nowhere in his
four mathematical papers does this determinant appear. It is rather strange, therefore, that this
determinant should be named after him and several authors have puzzled over the fact for some time.
Lebesgue's conjecture in [3] (first published in 1940) that it resulted for someone misreading
Vandermonde's notation, and therefore believing that this determinant was in his work, seems the most
likely.
Vandermonde's four mathematical papers, with their dates of publication by the Académie des
Sciences, were
Mémoire sur la résolution des équations
(1771),
Remarques sur des problèmes de
situation
(1771),
Mémoire sur des irrationnelles de différens ordres avec une application au cercle

(1772), and
Mémoire sur l'élimination
(1772).

The first of these four papers presented a formula for the sum of the
m
th powers of the roots of
an equation. It also presented a formula for the sum of the symmetric functions of the powers of such
roots. Neither of these were new having appeared in Waring's work shortly before but, although he was
aware of this Vandermonde claimed, rightly in my [EFR] opinion, that his approach was sufficiently
different to make publication of these results for a second time worthwhile. The paper also shows that
if
n
is a prime less than 10 the equation
x
n
- 1 = 0 can be solved in radicals. Jones writes in [1]:-
... Vandermonde's real and unrecognised claim to fame was lodged in his first paper, in which he
approached the general problem of the solubility of algebraic equations through a study of functions
invariant under permutations of the roots of the equation.

Kronecker claimed in 1888 that the study of modern algebra began with this first paper of
Vandermonde. Cauchy states quite clearly that Vandermonde had priority over Lagrange for this
remarkable idea which eventually led to the study of group theory.
In his second paper Vandermonde considered the problem of the knight's tour on the chess board.
This paper is an early example of the study of topological ideas. Vandermonde considers the
intertwining of the curves generated by the moving knight and his work in this area marks the beginning
of ideas which would be extended first by Gauss and then by Maxwell in the context of electrical
circuits.
In his third paper Vandermonde studied combinatorial ideas. He defined the symbol
[
p
]
n
=
p
(
p
-1)(
p
-2)(
p
-3)...(
p
-
n
+1 )
and
[
p
]
-
n
= 1(
p
+1)(
p
+2)(
p
+3)...(
p
+n
).
He gave an identity for the expansion of [
x
+
y
]
n
and also proved that
π2 = [
1

2
]
12
.[-
1

2
]
-12

It is interesting to note that at this time no notation existed for
n
! yet with his notation
Vandermonde had defined something more general. Clearly
[
n
]
n
=
n
!
The final of Vandermonde's four papers studied the theory of determinants. Muir [4] claims that
because of this paper Vandermonde was:-
The only one fit to be viewed as the founder of the theory of determinants.

The reason for this strong claim by Muir is that, although mathematicians such as Leibniz had
studied determinants earlier than Vandermonde, all earlier work had simply used the determinant as a
tool to solve linear equations. Vandermonde, however, thought of the determinant as a function and gave

properties of the determinant function. He showed the effect of interchanging two rows and of
interchanging two columns. From this he deduced that a determinant with two identical rows or two
identical columns is zero. Finally he gave a remarkably clever notation for determinants which has not
survived.


References for Alexandre-Theophile
Vandermonde

1. P S Jones, Biography in Dictionary of Scientific Biography (New York 1970-1990).
Articles:
2. H Lebesgue, L'oeuvre mathématique de Vandermonde, Thales, recueil des travaux de
l'Institut d'histoire des sciences IV (1937-39), 28-42.
3. H Lebesgue, L'oeuvre mathématique de Vandermonde, Enseignement Math. (2) 1 (1956),
203-223.
4. T Muir, History of determinants Volume 3 (1920).
5. J H Przytycki, History of the knot theory from Vandermonde to Jones, in XXIVth National
Congress of the Mexican Mathematical Society (México City, 1992), 173-185.
6. J J Tattersall, Who put the 'C' in A-T Vandermonde?, Historia Math. 15 (4) (1988),
361-367.
7. J J Tattersall, Vandermonde's contributions to the early history of combinatorial theory,
Eleventh British Combinatorial Conference, Ars Combin. 25 (1988), C, 195-203.