Bourbaki

Proposition 6

Tag 09M9, permanent. Cite it as [Bourbaki, Tag 09M9].

Integration, chapter VIII, § 5. The space of closed subgroups

Suppose that the locally compact group G is generated by a compact neighborhood of e. Then the set Γc0\Gamma_c^0 of measures α∈Γ0\alpha \in \Gamma^0 such that G/HαG/H_\alpha is compact is open in Γ0\Gamma^0, and the restriction to Γc0\Gamma_c^0 of the mapping α↦∥μα∥\alpha \mapsto \| \mu_\alpha \| is vaguely continuous.

With the notations of Prop. 5 of No. 2, we have, for g∈K+(G)g \in \mathscr{K}_+(G),

Γ0(g)⊂Γc0. \Gamma^0(g) \subset \Gamma_c^0.

For, if K is the support of gg, the relation ∫g(xs) dα(s)≥1\int g(xs)\, d\alpha(s) \geq 1 for all x∈Gx \in G implies KHα=GKH_\alpha = G, the integral obviously being zero on the complement of KHαKH_\alpha, therefore G/Hα=πα(K)G/H_\alpha = \pi_\alpha(K) is compact. Given a measure α∈Γc0\alpha \in \Gamma_c^0, it will therefore suffice to define a function g∈K+(G)g \in \mathscr{K}_+(G) such that Γ0(g)\Gamma^0(g) is a neighborhood of α\alpha in Γ0\Gamma^0. Since G/HαG/H_\alpha is compact and the canonical mapping f↦fαf \mapsto f_\alpha of K+(G)\mathscr{K}_+(G) into K+(G/Hα)\mathscr{K}_+(G/H_\alpha) is surjective (Ch. VII, §2, No. 2), there exists a function g∈K+(G)g \in \mathscr{K}_+(G) such that ∫g(xs) dα(s)=2\int g(xs)\, d\alpha(s) = 2 for every x∈Gx \in G. Let K be the (compact) support of gg, L a symmetric compact neighborhood of e in G that generates G; the mapping β↦g∗β\beta \mapsto g * \beta of M+(G)\mathscr{M}_+(G) into C(G)\mathscr{C}(G) being vaguely continuous (§ 4, No. 2, Remark 1), there exists a neighborhood W of α\alpha in Γ0\Gamma^0 such that

(g∗β)(x)=∫g(xs) dβ(s)≥1 (g * \beta)(x) = \int g(xs)\, d\beta(s) \geq 1

for all β∈W\beta \in W and x∈LKx \in LK. The first member of (7) being equal to zero outside KHβKH_\beta, the relation β∈W\beta \in W implies

LK⊂KHβ, LK \subset KH_\beta,

from which one deduces, by induction on nn, that LnK⊂KHβL^n K \subset KH_\beta for every integer n>0n > 0; since L generates G, we therefore have G=KHβG = KH_\beta for every measure β∈W\beta \in W, which proves that W⊂Γc0W \subset \Gamma_c^0. On the other hand, the first member of (7) being invariant on the right under HβH_\beta, the inequality (7) is also valid for x∈LKHβ=Gx \in LKH_\beta = G; therefore W⊂Γ0(g)W \subset \Gamma^0(g), which proves the proposition.

Finally, (iii) will be a consequence of the following proposition:

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

Cited by

Cites

Transcribed from the printed volume. Not translated.