— Let us put E=A[X1,…,Xn] and S=A[X1,…,Xn]sym.
a) The A-algebra S of symmetric polynomials is generated by s1,…,sn.
b) The elements s1,…,sn of E are algebraically independent over A (IV, p. 4).
c) The family of monomials Xν=X1ν(1)…Xnν(n) such that 0≤ν(i)<i for 1≤i≤n is a basis of the S-module E. In particular, E is a free S-module of rank n!.
We shall prove the theorem by induction on n, the case n=0 being trivial. Let us write B=A[Xn] and denote by sk′ the elementary symmetric polynomial of degree k in X1,…,Xn−1; we thus have B[X1,…,Xn−1]=A[X1,…,Xn]. If we replace n by n−1 and A by B in the statement of Th. 1, we can formulate the induction hypothesis as follows:
(A) The B-algebra S′ of polynomials f∈A[X1,…,Xn] invariant under all permutations of X1,…,Xn−1 is generated by s1′,…,sn−1′.
(B) The elements s1′,…,sn−1′, of E are algebraically independent over B.
(C) The family of monomials X1ν(1)…Xn−1ν(n−1) such that 0≤ν(i)<i for 1≤i≤n−1 is a basis of the S′-module E.
We have the obvious relation
(5)
sk=sk′+sk−1′Xn(1≤k≤n−1),
from which we obtain by induction on k
(6)
sk′=(−1)kXnk+i=1∑k(−1)k−isiXnk−i(1≤k≤n−1).
We have S⊂S′, hence s1,…,sn belong to S′; by (A) and formula (6) the B-algebra S′ is therefore generated by s1,…,sn−1′.
By (B) there exists an endomorphism μ of the B-algebra S′ such that
By (5), we have u(sk)=u(sk′)+u(sk−1′)Xn, whence u(sk)=sk′ by an easy calculation. Let P∈B[Y1,…,Yn−1]; then from P(s1,…,sn−1)=0 we deduce
0=u(P(s1,…,sn−1))=P(s1′,…,sn−1′),
whence P=0 by (B). It follows that the B-algebra S′ is generated by the algebraically independent elements s1,…,sn−1. This property may be reformulated as follows: (D) The A-algebra S′ is generated by the algebraically independent elements s1,…,sn−1,Xn.
We can thus identify S' with the polynomial ring C[Xn], where C is the sub-A-algebra of E generated by s1,...,sn−1.
To prove a), let f∈S be a homogeneous symmetric polynomial of degree m. We have f∈S′=C[Xn], so there exists an element g=P(s1,...,sn−1) of C, homogeneous of degree m in X1,...,Xn, such that f−g is divisible by Xn. Since f−g is symmetric, each of its terms is also divisible by X1,...,Xn−1, hence f−g is divisible by sn=X1...Xn. In other words, there exists h∈S such that f=g+hsn, whence degh<m. By induction on m it follows that f belongs to
C[sn]E=A[s1,...,sn−1,sn]E.
Thus we have shown that the A-algebra S is generated by s1,...,sn.
Next we prove b). If we substitute X, for X in (4), we find
(−1)n+1sn=Xnn+k=1∑n−1(−1)n−ksn−kXnk;
in other words, (−1)n+1sn is a monic polynomial in X, of degree n and with coefficients in C. By IV, p. 11 we thus have the following property : (E) The C-algebra homomorphism φ of C[T] into C[Xn]=S′ which maps T to sn is injective, and S′ is a free module over the image of φ, with basis (1,Xn,...,Xnn−1).
The elements s1,...,sn−1 of C are algebraically independent over A, by (D); thus the injectivity of φ means that s1,...,sn−1,sn are algebraically independent over A, whence b).
To prove c), the image of φ equals C[sn]E=S, hence by (E), S′ is a free module over S on the basis (1,X1,...,Xnn−1). Now the assertion c) follows from the induction hypothesis (C) and Prop. 25 of II, p. 222. This completes the proof.
Let f be a symmetric polynomial in X1,...,Xn, homogeneous of degree m. By Th. 1 (IV, p. 62) there exists a polynomial Q∈A[Y1,...,Yn] such that f=Q(s1,...,sn). The preceding proof provides a means of explicit calculation of Q, by a double induction on n and m. For as we have seen, there exists a polynomial P∈A[Y1,...,Yn−1] and a polynomial h symmetric in X1,...,Xn which is homogeneous of degree m−n, such that
f=P(s1,...,sn−1)+snh.
For every polynomial u∈A[X1,...,Xn] we put
u′(X1,...,Xn−1)=u(X1,...,Xn−1,0)
Then s1′,...,sn−1′ are the elementary symmetric polynomials in X1,...,Xn−1 and formula (8) implies
f′=P(s1′,...,sn−1′)
The determination of P is thus reduced to a calculation of symmetric polynomials in n−1 indeterminates, and h is obtained from (8).
2) Again let n=3 and put p=X13+X23+X33. We have p(X1,0,0)=X13=s1(X1,0,0)3. Writing q=p−s13, we obtain
q=−3f−6X1X2X3=−3s1s2+3s3
and finally
p=s13−3s1s2+3s3.
Let S1,...,Sn be indeterminates. We shall equip the polynomial algebra A[S1,...,Sn] with the graduation of type N for which Sk is of weight k for 1≤k≤n (IV, p. 3), and we equip A[X1,...,Xn] with the usual graduation. For 1≤k≤n the elementary symmetric polynomial sk,n in X1,...,Xn is homogeneous of degree k. By Th. 1 (IV, p. 62) the mapping g↦g(s1,n,...,sn,n) is thus an isomorphism of graded algebras
φn:A[S1,...,Sn]→A[X1,...,Xn]sym
Let m be an integer such that 0≤m≤n. For every integer k such that 1≤k≤m we have
sk,m(X1,...,Xm)=sk,n(X1,...,Xm,0,...,0)
by definition (IV, p. 61, formula (1)) of sk. Hence the diagram