Bourbaki

Theorem 2

Tag 00MW, permanent. Cite it as [Bourbaki, Tag 00MW].

Algebra, VIII, p. 385, § 20. Linear Representations of Algebras

Let A\mathscr{A} be a generating subset of the K-vector space A. The homomorphism χA:RK(A)→\chi_{\mathscr{A}}: R_K(A)\rightarrow (1 + TK[[T]])A^{\mathscr{A}} defined by the relation χA(x)=(χa(x))a∈A\chi_{\mathscr{A}}(x) = (\chi_a(x))_{a\in\mathscr{A}} is injective.

Let xx be an element of RK(A)R_K(A) such that χA(x)=1\chi_{\mathscr{A}}(x) = 1. By (11), we have θ(x)(a)=0\theta (x)(a) = 0 for every a∈Aa\in \mathscr{A}, and therefore θ(x)=0\theta (x) = 0 because θ(x)\theta (x) is a K-linear form on A and A\mathscr{A} generates the K-vector space A. If the characteristic of K is zero, then this implies x=0x= 0 (VIII, p. 384, Corollary of Proposition 6), and therefore the result in this case. Suppose from now on that the characteristic pp of K is nonzero.

Let us first treat the case when K is algebraically closed. By the above and loc. cit., we then have x∈pRK(A)x\in pR_K(A). Let y∈RK(A)y\in R_K(A) be such that x=pyx=py. For every element aa of A\mathscr{A}, we have χa(y)p=χa(py)=χa(x)=1\chi_a(y)^p=\chi_a(py) =\chi_a(x) = 1. So we have (χa(y)−1)p=0(\chi_a(y)-1)^p= 0, and therefore χa(y)=1\chi_a(y) = 1 because the ring K[[T]] is an integral domain. So yy belongs to the kernel of the endomorphism χA\chi_{\mathscr{A}}. It follows by induction that xx belongs to pnRK(A)p^n R_K(A) for every integer n≥1n \geq 1. Since RK(A)R_K(A) is a free Z\mathbf{Z}-module, this implies x=0x = 0, and therefore the injectivity of χA\chi_{\mathscr{A}} in this case.

If KK is no longer supposed to be algebraically closed, then we choose an algebraic closure K‾\overline{K} of KK and consider the diagram of groups and group homomorphisms

RK(A)→χA(1+TK[[T]])A∣u∣iRK‾(A(K‾))→χ‾A(1+TK‾[[T]])A, \begin{array}{ccc} R_K(A) & \xrightarrow{\chi_{\mathscr{A}}} & (1 + TK[[T]])^{\mathscr{A}} \\ |_u & & |_i \\ R_{\overline{K}}(A_{(\overline{K})}) & \xrightarrow{\overline{\chi}_{\mathscr{A}}} & (1 + T\overline{K}[[T]])^{\mathscr{A}}, \end{array}

where uu is the homomorphism deduced from the extension of scalars from KK to K‾\overline{K} (VIII, p. 195), ii is the canonical injection, and χ‾A\overline{\chi}_{\mathscr{A}} is the homomorphism z↦(χ1⊗a(z))a∈Az \mapsto (\chi_{1 \otimes a}(z))_{a \in \mathscr{A}}. By formula (12) of III, §9, No. 1, p. 542, diagram (12) is commutative. By the above, the homomorphism χ‾A\overline{\chi}_{\mathscr{A}} is injective. Since uu is injective (VIII, p. 195, Theorem 1), the homomorphism χA\chi_{\mathscr{A}} is injective.

Bourbaki sets the proof directly after the statement, and the corpus does not separate them, so what is above is the statement and its proof together.

Cited by

Cites

Transcribed from the printed edition, 2023, Springer Nature. Not translated.