The endomorphism ring of an isotypic module is simple #
A semisimple module M all of whose simple submodules are isomorphic to one another
(IsIsotypic R M) and which is finite over R is a finite power Sⁿ of a single simple module,
by Mathlib's IsIsotypic.linearEquiv_fun. Its endomorphism ring is therefore the matrix ring
Matₙ(End_R S) over the division ring supplied by Schur's lemma, and a matrix ring of nonzero
size over a division ring is simple. So
IsSimpleRing (Module.End R M)
for every nonzero such M, with no hypothesis on R beyond what makes M semisimple.
This is the shape in which Artin--Wedderburn uniqueness uses endomorphism rings: over a semisimple
ring the isotypic components of the regular module are exactly the modules this applies to, so
End_R R is a product of simple rings indexed by them, one factor per Wedderburn block. See
TauCeti/RingTheory/Semisimple/Wedderburn/Blocks.lean.
Main results #
TauCeti.isSimpleRing_moduleEnd_of_isIsotypic: the endomorphism ring of a nonzero finite isotypic semisimple module is simple.
Implementation notes #
Nonzero-ness of M is what makes the matrix size n nonzero, and it cannot be dropped: for
M = 0 the ring Module.End R M is trivial and a trivial ring is not simple.
References #
This is the module-theoretic engine behind the Layer 2 uniqueness statements of the semisimple algebras roadmap. See T. Y. Lam, A First Course in Noncommutative Rings, GTM 131, §3, or C. W. Curtis and I. Reiner, Representation Theory of Finite Groups and Associative Algebras, §25.
The endomorphism ring of a nonzero finite isotypic semisimple module is simple.
Writing M as a power Sⁿ of a simple module presents End_R M as the matrix ring
Matₙ(End_R S) over the division ring End_R S of Schur's lemma, and n ≠ 0 because M ≠ 0.
The ring R is arbitrary: only the module is asked to be semisimple, isotypic and finite. Over a
simple Artinian ring every module is isotypic, which is the specialization
TauCeti.IsSimpleRing.moduleEnd.