5.22.2 Tensor Product Symmetry Isomorphism

Tensor Product Symmetry Isomorphism is the canonical linear isomorphism τ_{V,W} : V ⊗ W → W ⊗ V defined on simple tensors by τ_{V,W}(v ⊗ w) = w ⊗ v, identifying the tensor product of two vector spaces taken in one order with the tensor product taken in the reverse order. This isomorphism is the precise algebraic statement that the tensor product is commutative up to canonical identification, complementing the associativity isomorphism and together with it giving the category of vector spaces the structure of a symmetric monoidal category.

Definition of the Isomorphism

Action on Simple Tensors

The symmetry isomorphism is the unique linear map determined by

for all v ∈ V, w ∈ W, swapping the two tensor factors while leaving each factor's content unchanged.

Existence via the Universal Property

The map (v, w) ↦ w ⊗ v is bilinear as a function V × W → W ⊗ V, since it is linear in v for fixed w and linear in w for fixed v; the universal property of V ⊗ W therefore induces a unique linear map τ_{V,W} : V ⊗ W → W ⊗ V satisfying the displayed formula on simple tensors, and extended to all of V ⊗ W by linearity.

Bijectivity and Self-Inverse Property

The Symmetry Isomorphism Is Its Own Inverse

Applying the analogous construction with V and W swapped gives τ_{W,V} : W ⊗ V → V ⊗ W, and composing,

shows this composite equals the identity on simple tensors, and hence, by spanning, on all of V ⊗ W. Thus τ_{W,V} = τ_{V,W}^{-1}, and applying the swap twice returns the original space and the identity map, an involution property characteristic of a symmetry isomorphism.

Diagram of the Symmetry Isomorphism

Naturality of the Symmetry Isomorphism

The Naturality Square

For linear maps f : V → V′ and g : W → W′, the symmetry isomorphism satisfies the naturality condition

meaning that swapping factors before or after applying independent transformations to each factor gives the same result, provided the order of the maps f and g is swapped correspondingly on the other side.

Hexagon Coherence with Associativity

When both associativity and symmetry isomorphisms are present, a further coherence condition, the hexagon identity, governs how τ interacts with α across three factors, ensuring that reordering and regrouping three tensor factors gives consistent results regardless of the sequence of symmetry and associativity moves used to achieve a given permutation and grouping.

Matrix Description in Finite Dimensions

Symmetry as a Permutation of Basis Indices

With bases {eᵢ} of V and {fⱼ} of W, the symmetry isomorphism sends the basis element eᵢ ⊗ fⱼ of V ⊗ W to the basis element fⱼ ⊗ eᵢ of W ⊗ V; relative to the corresponding orderings of these two bases, τ_{V,W} acts as the permutation matrix that transposes the roles of the two index sets, matching the transpose operation familiar from swapping the row and column indices of a matrix.

Distinction from Associativity

A Separate Structural Isomorphism

The symmetry isomorphism addresses the order of factors, V ⊗ W versus W ⊗ V, whereas the associativity isomorphism addresses the grouping of factors, (U ⊗ V) ⊗ W versus U ⊗ (V ⊗ W); these are independent structural features of the tensor product, and neither one is derivable from the other, though both are needed together to fully describe how tensor products of several factors can be freely reordered and regrouped.

Significance of the Symmetry Isomorphism

Formalizing Commutativity Up to Isomorphism

The symmetry isomorphism captures precisely in what sense the tensor product is commutative: not as a literal set-theoretic equality V ⊗ W = W ⊗ V, but as a canonical, natural, involutive isomorphism between the two orders, a distinction essential to careful algebraic and categorical treatments of the tensor product.

Structural Component of the Symmetric Monoidal Category

Together with the associativity isomorphism and the unit isomorphisms F ⊗ V ≅ V ≅ V ⊗ F, the symmetry isomorphism equips the category of vector spaces with the structure of a symmetric monoidal category, the standard categorical framework in which the full coherence theory of tensor products — governing associativity, commutativity, and their mutual compatibility — is developed.