4.1.5 Tensor Multilinear Map Algebraic Role

Tensor multilinear maps play a foundational role in algebra by generalizing linear transformations and enabling the study of multilinear relationships in tensor algebras.

Tensor Multilinear Map Algebraic Role is the position occupied by the multilinear map interpretation of tensors within the broader algebraic system of tensor algebra, encompassing how it underlies the vector space structure of tensor spaces, mediates the operations of tensor product, contraction, and pullback, and connects tensor algebra to the theory of algebras over a field through the tensor algebra construction.

Multilinear Maps as the Generating Mechanism of Tensor Spaces

Vector Space Structure Inherited from Multilinearity

The set of all multilinear maps of a fixed type (p, q) on V forms a vector space under pointwise addition and scalar multiplication, and this vector space is what is meant, algebraically, by the space of type (p, q) tensors. The algebraic role of the multilinear map picture is thus foundational: it is the definition from which the linear structure of tensor spaces is derived, rather than an alternative description added afterward.

(T + S) (args) = T (args) + S (args)

Basis Independence as an Algebraic Guarantee

Because the multilinear map is defined without reference to a basis, algebraic identities proved about tensors, such as distributivity of the tensor product over addition, or the interaction of contraction with the tensor product, hold automatically in every basis. The multilinear map role is precisely to supply this basis-free algebraic footing, with the component picture serving only as a computational tool derived from it.

Role in the Tensor Product Operation

Universal Property as an Algebraic Statement

The multilinear map structure is what defines the tensor product algebraically, via its universal property: the tensor product V ⊗ W together with the canonical bilinear map ⊗: V × W → V ⊗ W is characterized by the fact that every bilinear map out of V × W factors uniquely through ⊗. This makes the multilinear map role central rather than incidental to the very existence and uniqueness of the tensor product as an algebraic object.

Associativity and the Tensor Algebra

Iterating the tensor product on a single vector space V produces the graded tensor algebra

T (V) = \oplus_{k = 0}^{\infty} V^{\otimes k}

with multiplication given by concatenation of tensor factors. The multilinear map role guarantees this multiplication is well-defined and associative, since concatenation of multilinear maps corresponds exactly to forming a multilinear map on the concatenated argument list, and the associativity of argument-list concatenation transfers directly to associativity of the algebra's product.

Role in Contraction and Trace

Contraction as a Canonical Linear Map

Contraction, the operation pairing one covector slot with one vector slot and summing, is algebraically a linear map from the space of type (p, q) tensors to the space of type (p-1, q-1) tensors. Its definition relies entirely on the multilinear map role: contraction is defined by inserting the canonical evaluation pairing V* × V → ℝ into the chosen pair of slots, making it a composition of multilinear maps rather than an ad hoc rule on index arrays.

Trace as a Special Case

The trace of a type (1, 1) tensor, viewed as a linear operator, is the contraction of its two slots, and algebraically it is the unique linear functional on End(V) that is invariant under conjugation, a fact whose proof proceeds directly from the multilinear map description of contraction and its compatibility with change of basis.

Role in Functorial Operations

Pullback as an Algebra Homomorphism on Covariant Tensors

For a linear map f: V → W, pullback of covariant tensors, f*: T^0_q(W) → T^0_q(V), defined via precomposition of the multilinear map with f in every slot, is not merely linear but also respects the tensor product,

f^{*} (S \otimes T) = f^{*} S \otimes f^{*} T

which makes pullback an algebra homomorphism between the covariant tensor algebras of W and V. This homomorphism property is a direct algebraic consequence of the multilinear map role, since precomposition with f commutes with the argument-list concatenation that defines the tensor algebra's product.

Functoriality and Composition

The assignment f ↦ f* is contravariantly functorial, satisfying (g ∘ f)* = f* ∘ g* for composable linear maps f and g, a fact whose proof is a one-line consequence of unwinding the multilinear map definition of pullback on both sides. This functorial role is what allows tensor algebra to interact coherently with the broader categorical structure of vector spaces and linear maps.

Role in Distinguishing Tensor Types

Algebraic Classification by Slot Type

The multilinear map role also serves a classificatory algebraic function: it is what distinguishes a symmetric tensor algebra, an exterior algebra, and the full tensor algebra as three different algebraic quotients or substructures built from the same underlying multilinear map data, differing only in which symmetry constraints are imposed on the argument slots. This shared foundation is why results proved at the level of multilinear maps, such as universal properties or dimension counts, transfer uniformly across all of these related algebraic constructions.