2.21.3 Tensor Isomorphism Dimension Preservation

Tensor Isomorphism Dimension Preservation is the principle that any isomorphism between vector spaces, and every tensor construction built functorially from such an isomorphism, leaves dimension invariant, together with the specific counting formulas that show exactly how the dimension of a tensor space is determined by the dimensions of its constituent factors. Dimension is the complete isomorphism invariant for finite-dimensional vector spaces over a fixed field, so tracking how dimension behaves under duals, tensor products, direct sums, and symmetric and exterior powers is what allows the isomorphism type of any tensor space to be computed in advance, without constructing an explicit isomorphism.

Dimension as a Complete Invariant

Sufficiency of Dimension for Isomorphism

For finite-dimensional vector spaces V and W over the same field F, V ≅ W if and only if dim(V) = dim(W). Sufficiency follows from the explicit basis-matching construction described in the isomorphism linear map entry: any bijection between a basis of V and a basis of W, extended linearly, produces an isomorphism whenever the two bases have equal cardinality. Necessity follows because an isomorphism carries a basis of V to a basis of W, so the two spaces must have equally many basis vectors.

What Dimension Preservation Does and Does Not Guarantee

Because dimension is a complete invariant, confirming that two tensor spaces have equal dimension is sufficient to know they are abstractly isomorphic as vector spaces, but it says nothing about whether a natural, structure-respecting isomorphism exists between them if they arise from different constructions. Two spaces of equal dimension built by different tensor operations, such as a space of symmetric bilinear forms and a space of alternating bilinear forms of matching dimension, are isomorphic merely as vector spaces but carry no natural correspondence to one another; dimension preservation is necessary for a construction-respecting isomorphism to exist, but never sufficient on its own.

Dimension Formulas for Tensor Constructions

Dual Space

dim(V*) = dim(V), a direct consequence of the dual basis construction, which produces exactly n dual basis covectors from n basis vectors, for n = dim(V).

Tensor Product

Dimension is multiplicative under tensor product:

since a basis for V ⊗ W is given by all products e_i ⊗ f_j of a basis element of V and a basis element of W, and there are dim(V) × dim(W) such pairs. Applied repeatedly, this gives the dimension of a general (p, q)-tensor space, dim(T^p_q(V)) = n^{p+q} for n = dim(V), since T^p_q(V) is built from p + q tensor factors, each of dimension n.

Direct Sum

Dimension is additive under direct sum:

since a basis for V ⊕ W is the disjoint union of a basis of V and a basis of W, embedded in the two summands.

Dimension of Symmetric and Exterior Powers

Symmetric Power Dimension

The space Sym^k(V) of symmetric tensors of rank k built from V, used to model symmetric forms and symmetric tensor products, has dimension given by a combination with repetition count:

for n = dim(V), since a basis for Sym^k(V) corresponds to the multisets of size k drawn from an n-element basis of V, counted by the standard combinations-with-repetition formula.

Exterior Power Dimension

The space Λ^k(V) of alternating tensors of rank k, used to model volume forms and orientation, has dimension given by an ordinary binomial coefficient:

since a basis for Λ^k(V) corresponds to the k-element subsets, without repetition, of a basis of V, reflecting the antisymmetry that forces any repeated basis vector in a wedge product to vanish. In particular Λ^k(V) = 0 once k > n, a dimension collapse with no analogue in the symmetric or general tensor product case.

How an Isomorphism of the Base Space Propagates Dimension Preservation

Induced Maps Preserve Dimension by Construction

Whenever φ : V → W is an isomorphism, every induced map, φ^* on duals, φ^{⊗k} on tensor powers, the restriction of φ^{⊗k} to Sym^k or Λ^k, is automatically an isomorphism on spaces that already have matching dimension by the counting formulas above; the induced maps do not need to separately verify dimension equality, since equal dimension is guaranteed in advance once dim(V) = dim(W) is known and the same construction is applied to both sides.

Dimension Preservation as a Consistency Check

Because every tensor construction has an explicit dimension formula, dimension preservation also functions as a sanity check on proposed isomorphisms: if two tensor spaces claimed to be isomorphic do not have matching dimension under these formulas, no isomorphism between them can exist, regardless of how plausible the claimed correspondence appears. This makes dimension counting the first and most economical test applied before attempting to construct or verify any tensor space isomorphism.