(a) The restriction π0 of π to L(X) is a derivation of L(X). (b) For every integer n≥1 and all u in Ln(X), π(u)=n.u.
(a) Let E be the endomorphism algebra of the module L(X) and θ the homomorphism of A(X) into E such that θ(x)=adx for all x∈X. The restriction of θ to L(X) is a Lie algebra homomorphism of L(X) into E, which coincides on X with the adjoint representation of L(X), whence
θ(u).v=[u,v]for u,v in L(X).
Let a be in A(X) and b in A+(X); then
π(a.b)=θ(a).π(b).
It suffices to consider the case a=x1…xp, b=xp+1…xp+q with p≥0,q≥1 and x1,…,xp+q in X; but then (5) follows immediately from (3) since θ(x)=adx for x∈X.
Let u and v be in L(X); by (4) and (5),
π0([u,v])=π(uv−vu)=θ(u)⋅π(v)−θ(v)⋅π(u)=[u,π(v)]−[v,π(u)]=[u,π0(v)]+[π0(u),v],
hence π0 is a derivation of L(X).
(b) Let π1 be the endomorphism of the module L(X) which coincides on Ln(X) with multiplication by the integer n≥1. The formula [Ln(X),Lm(X)]⊂Ln+m(X) shows that π1 is a derivation (Algebra, Chapter III, § 10, no. 3, Example 6). The derivation π1−π0 of L(X) is zero on X and, as X generates L(X), π0=π1, whence (b).
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.