On , define the permutation and diagonal actions by
The inverse in the first formula gives a left action. Applying to every tensor factor and then reordering produces the same tensor as reordering and then applying . Thus the two actions commute. The Schur algebra is
It is the commutant of an operator algebra of the permutation action.
To identify this commutant, use the multilinear isomorphism
Conjugation by a permutation reorders the factors on the right. Therefore the invariant subspace is the space of symmetric tensors of degree in . It is spanned by . Indeed, the polarization identity
expresses every symmetrized elementary tensor as a linear combination of pure powers. Those symmetrized elementary tensors span the invariant subspace in characteristic zero.
One may restrict to invertible endomorphisms without changing the span. For fixed , the vector-valued polynomial has degree at most . Choose distinct values of away from the finitely many roots of . Polynomial interpolation expresses its constant term as a linear combination of those invertible tensor powers. Consequently
The span is an algebra, since products are .
The image of is semisimple, as a quotient of a semisimple algebra. The double-centralizer theorem for semisimple operator algebras gives . Since the preceding computation says that is the span of the general linear action, the two actions are mutual commutants. More explicitly, semisimple module decomposition gives
Here is the multiplicity space, with its natural general linear action. The commutant algebra is the product of the full endomorphism algebras of these multiplicity spaces. Hence each nonzero is irreducible, and distinct spaces have nonisomorphic general linear representations, because the operators span that commutant. A primitive picks a one-dimensional factor from , so is an equivalent realization of as a Schur module.
The range of shapes is exact. If has more than rows, a column antisymmetrizes more than vectors and its action vanishes by . Conversely, for at most rows place the th basis vector in every tensor position belonging to row of . Row symmetrization multiplies this tensor by . Column antisymmetrization is nonzero, because all vectors within each column are distinct, and its different permutations give distinct basis tensors. Thus . This proves the length bound for a Schur module and the stated Schur–Weyl duality. For , the empty partition and give the trivial version.