Bourbaki

Corollary

Tag 05NN, permanent. Cite it as [Bourbaki, Tag 05NN].

Algebra, chapter II, § 2. Modules of linear mappings. Duality

*Let M, N be two supplementary submodules in an A-module E and p:E→M,q:E→Np : E \to M, q : E \to N the corresponding projectors; then tp+tq:M∗⊕N∗→E∗^tp + ^tq : M^* \oplus N^* \to E^* is an isomorphism and tp^tp (resp. tq^tq) is an isomorphism of M∗M^* (resp. N∗N^*) onto the submodule of E∗E^* orthogonal to NN (resp. MM). Moreover, if i:M→Ei : M \to E and j:N→Ej : N \to E are the canonical injections, tp∘ti^tp \circ ^ti and tq∘tj^tq \circ ^tj are the projectors E∗→tp(M∗)E^* \to ^tp(M^*), E∗→tq(N∗)E^* \to ^tq(N^*) corresponding to the decomposition of E∗E^* as the direct sum of tp(M∗)^tp(M^*) and tq(N∗)^tq(N^*).

p∘i=1M,q∘j=1N,p∘j=q∘i=0,i∘p+j∘q=1Ep \circ i = 1_M, \quad q \circ j = 1_N, \quad p \circ j = q \circ i = 0, \quad i \circ p + j \circ q = 1_E, whence, by transposition (no. 5, formulae (16), (17) and (18)), ti∘tp=1M∗, tj∘tq=1N∗, tj∘tp=ti∘tq=0. tp∘ti+tq∘tj=1E∗^ti \circ ^tp = 1_{M^*}, \ ^tj \circ ^tq = 1_{N^*}, \ ^tj \circ ^tp = ^ti \circ ^tq = 0. \ ^tp \circ ^ti + ^tq \circ ^tj = 1_{E^*} and the proposition follows from § 1, no. 6, Corollary 2 to Proposition 6.

Under the hypotheses of the Corollary, M∗M^* (resp. N∗N^*) is often identified with the orthogonal tp(M∗)^tp(M^*) (resp. tq(N∗)^tq(N^*)) of NN (resp. MM) in E∗E^*, thus identifying every linear form uu on MM (resp. NN) with the linear form on EE extending uu and which is zero on NN (resp. MM).

When an A-module EE admits a basis (et)t∈T(e_t)_{t \in T}, it has been seen that giving this basis defines canonically an isomorphism u:As(T)→Eu : A_s^{(T)} \to E. By virtue of Proposition 10 and no. 3, Proposition 5, the dual of As(T)A_s^{(T)} is canonically identified with the product AdTA_d^T; consider the contragredient isomorphism u~:AdT→E∗\tilde{u} : A_d^T \to E^*. If, for all t∈T,ftt \in T, f_t is the element of AdTA_d^T all of whose projections are zero with the exception of that of index tt, which is equal to 1, and if we write et∗=u~(ft)e_t^* = \tilde{u}(f_t), the elements et∗e_t^* of E∗E^* are, by (19) and (22), characterized by the relations

⟨et,et′∗⟩={0for t′≠t1for t′=t. \langle e_t, e_{t'}^* \rangle = \begin{cases} 0 & \text{for } t' \neq t \\ 1 & \text{for } t' = t. \end{cases}

(23)

It amounts to the same to say that, for all x=∑t∈Tξtet∈E, et∗(x)=ξtx = \sum_{t \in T} \xi_t e_t \in E, \ e_t^*(x) = \xi_t; also et∗e_t^* is called the coordinate form of index tt on EE. It follows from (23) that (et∗)(e_t^*) is a free system in E∗E^*.

In particular, if TT is finite, the et∗e_t^* form a basis of E∗E^*, the ftf_t then forming the canonical basis of AdTA_d^T. Hence:

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 edition, 1998, Springer. Not translated.