1.7.1 Algebraic Tensor Interpretation

Algebraic Tensor Interpretation explains how tensors generalize multilinear relationships using structured algebraic frameworks.

Algebraic Tensor Interpretation is the understanding of a tensor as an element of a vector space constructed purely from other vector spaces by an algebraic procedure, the tensor product, together with the equivalent understanding of a tensor as a multilinear map satisfying a universal factorization property. This interpretation treats tensors as citizens of ordinary linear algebra, built and manipulated using vector space operations, quotients, and universal properties, without appeal to geometry, physics, or any particular basis.

Constructing the Tensor Product Space

The Free Vector Space on Pairs

Given vector spaces V and W, the tensor product V ⊗ W is constructed by first forming the free vector space generated by all ordered pairs (v, w) with v in V and w in W, an enormous space in which every pair is treated as an independent basis element with no relations imposed yet.

Imposing Bilinearity by Quotient

The tensor product is then obtained by quotienting this free vector space by the subspace generated by all relations that bilinearity demands, such as (v_1 + v_2, w) - (v_1, w) - (v_2, w) and (αv, w) - α(v, w), so that the equivalence class of (v, w), written v ⊗ w, behaves linearly in each slot by construction.

V \otimes W = Free (V \times W) / R

where R denotes the subspace of bilinearity relations imposed on the free vector space generated by pairs from V and W.

The Universal Property

Statement of the Property

The tensor product space is characterized, up to a unique isomorphism, by the requirement that every bilinear map out of V × W into any vector space X factors uniquely through V ⊗ W by way of a linear map.

β : V \times W \to X \Rightarrow \exists! \tilde{β} : V \otimes W \to X with \tilde{β} (v \otimes w) = β (v, w)

Why This Property Defines the Interpretation

The algebraic tensor interpretation regards this universal property, rather than the explicit quotient construction used to build it, as the essential content of what a tensor product is: any vector space satisfying the same universal property is canonically isomorphic to V ⊗ W, so the property alone determines the algebraic role tensors play.

Tensors as Multilinear Maps: The Dual Algebraic Route

Identifying Tensors with Multilinear Functionals

Applying the universal property with V^* in place of V and repeating the construction identifies the tensor product V^{⊗p} ⊗ (V^*)^{⊗q} with the space of multilinear maps taking p covectors and q vectors to a scalar, giving the algebraic tensor interpretation two equivalent faces: elements of a tensor product space, and multilinear functionals.

V^{\otimes p} \otimes {(V^{*})}^{\otimes q} ≅ Mult ({(V^{*})}^{p} \times V^{q}; F)

Consequence for Finite-Dimensional Spaces

In finite dimensions this identification is an isomorphism of vector spaces, and it is what justifies treating the multilinear map definition of a tensor and the tensor product element definition as interchangeable within the algebraic interpretation, rather than as two different mathematical objects that happen to correspond.

The Algebraic Operations Native to This Interpretation

The Tensor Product as Multiplication

The algebraic interpretation supplies the tensor product operation itself, ⊗, as a bilinear multiplication combining a tensor in one space with a tensor in another to produce a tensor in the tensor product of those spaces, giving the collection of all tensors over V the structure of a graded algebra.

Contraction as a Canonical Linear Map

Contraction arises algebraically as the canonical linear map obtained by applying the evaluation pairing between V and V^* to one chosen upper and one chosen lower slot of a tensor product space, a construction available purely from the vector space structure of V and V^*, requiring no additional geometric data.

ev : V^{*} \otimes V \to F, ω \otimes v \mapsto ω (v)

Position Within the Family of Interpretations

Algebraic Versus Component Interpretation

The algebraic interpretation is basis-free by construction, in contrast to the component array interpretation, which requires a chosen basis to produce numbers; the algebraic view explains why component arrays obey the transformation law they do, since that law is exactly what is required for an array to represent a well-defined element of the basis-free tensor product space.

Algebraic Versus Geometric Interpretation

Where a geometric interpretation attaches meanings such as direction or physical magnitude to a tensor, the algebraic interpretation strips away any such meaning, treating a tensor purely as an element of a vector space equipped with the tensor product operation, a foundation onto which geometric or physical meaning can later be layered.

Diagrammatic Summary

The diagram traces the algebraic tensor interpretation from its construction, a free vector space collapsed by bilinearity relations, to its defining consequence, the universal property through which every bilinear map out of V × W factors uniquely through the tensor product space.