Bourbaki

Lemma 2

Tag 0AL8, permanent. Cite it as [Bourbaki, Tag 0AL8].

Lie Groups and Lie Algebras, chapter III, § 10. The automorphism group of a Lie group

Let K′K' be a non-discrete closed subfield of K, A a K′K'-manifold, B and C KK-manifolds and ff a K′K'-analytic mapping of A×BA \times B into C. Suppose that, for all

LIE GROUPS

a∈Aa \in A, the mapping b↦f(a,b)b \mapsto f(a, b) of B into C is K-*analytic*. Then, for all t∈TAt \in TA, the mapping u↦(Tf)(t,u)u \mapsto (Tf)(t, u) of TB into TC is K-*analytic*.

We fix t∈TAt \in TA and write g(u)=(Tf)(t,u)g(u) = (Tf)(t, u). Clearly gg is K'-analytic. By Differentiable and Analytic Manifolds, R, 5.14.6, it suffices to prove that the tangent mappings to gg are K-linear. It can be assumed that A, B, C are open neighbourhoods of 0 in complete normable spaces E, F, G over K', K, K and that tt is tangent to A at 0. We identify TA, TB, TC, with A×E,B×F,C×GA \times E, B \times F, C \times G and tt with an element of E. Then for all (x,y)∈TB=B×F(x, y) \in TB = B \times F.

g(x,y)=(f(0,x),(D1f)(0,x)(t)+(D2f)(0,x)(y)). g(x, y) = (f(0, x), (D_1f)(0, x)(t) + (D_2f)(0, x)(y)).

We identify T(B×F)T(B \times F) with (B×F)×(F×F)(B \times F) \times (F \times F) and T(C×G)T(C \times G) with (C×G)×(G×G)(C \times G) \times (G \times G). Then, for all

((x,y),(h,k))∈T(B×F)=(B×F)×(F×F), ((x, y), (h, k)) \in T(B \times F) = (B \times F) \times (F \times F),

(Tg)((x,y),(h,k))=((a,b),(c,d))(Tg)((x, y), (h, k)) = ((a, b), (c, d)), where

a=f(0,x),b=(D1f)(0,x)(t)+(D2f)(0,x)(y),c=(D2f)(0,x)(h),d=(D2D1f)(0,x)(t,h)+(D2D2f)(0,x)(y,h)+(D2f)(0,x)(k). \begin{align*} a &= f(0, x), \\ b &= (D_1f)(0, x)(t) + (D_2f)(0, x)(y), \quad c = (D_2f)(0, x)(h), \\ d &= (D_2D_1f)(0, x)(t, h) + (D_2D_2f)(0, x)(y, h) + (D_2f)(0, x)(k). \end{align*}

We now fix (x,y)∈B×F(x, y) \in B \times F. We need to prove that the mapping (h,k)↦(c,d)(h, k) \mapsto (c, d) of F×FF \times F into G×GG \times G is K-linear. As the mapping x↦f(0,x)x \mapsto f(0, x) of B into C is K-analytic, the mappings

(h,k)↦(D2f)(0,x)(h),(h,k)↦(D2D2f)(0,x)(y,h), (h, k) \mapsto (D_2f)(0, x)(h), \quad (h, k) \mapsto (D_2D_2f)(0, x)(y, h),
(h,k)↦(D2f)(0,x)(k)(h, k) \mapsto (D_2f)(0, x)(k)

are K-linear. On the other hand

(D2D1f)(0,x)(t,h)=lim⁡λ∈K′,λ→0λ−1((D2f)(λt,x)(h)−(D2f)(0,x)(h)) (D_2D_1f)(0, x)(t, h) = \lim_{\lambda \in K', \lambda \to 0} \lambda^{-1}((D_2f)(\lambda t, x)(h) - (D_2f)(0, x)(h))

and, for fixed λ\lambda, the mapping x↦f(λt,x)x \mapsto f(\lambda t, x) is K-analytic, so that the mapping h↦(D2f)(λt,x)(h)h \mapsto (D_2f)(\lambda t, x)(h) is K-linear.

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Cites

Nothing.

Transcribed from the printed volume. Not translated.