Every artinian A-module is artinian as a A-module.
Let M be an artinian A-module; every element of M is annihilated by a power of mA, hence by a power of mA. By Lemma 2 of No. 2, the sub-A-modules of M are its sub-A-modules, hence M is artinian as an A-module.
Let us now fix a Matlis A-module I. For every A-module M, let us denote by DA(M) the A-module
DA(M)=HomA(M,I)
The A-module DA(A) is canonically identified with I, the A-module DA(I) with A (n∘2, prop. 3), and the A-module DA(κA) with I1 (loc. cit.).
For every A-linear mapping f:M→N, we shall denote by
DA(f):DA(N)→DA(M)
the A-linear mapping HomA(f,1I). Since the A-module I is injective, the sequence (DA(g),DA(f)) is exact for every exact sequence (f,g) of A-linear mappings.
We shall apply these definitions to the ring A endowed with the Matlis module I (lemma 3, a)) ; for every A-module P, DA(P) is therefore the A-submodule HomA(P,I) of DA(P). It amounts to the same thing to say that P is artinian as an A-module or as a A-module (lemma 5) ; if this is the case, one has DA(P)=DA(P) (loc. cit.).
Let M be an A-module. For m∈M, the mapping f↦f(m) of DA(M) into I is A-linear ; let us denote it by αM(m). Thus one defines an A-linear homomorphism
αM:M⟶DA(DA(M))
We denote by αM:A⊗AM⟶DA(DA(M)) the A-linear mapping deduced from αM.
Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.