15.21.1 Tensor Symmetric Multilinear Pattern Boundary

Tensor Symmetric Multilinear Pattern Boundary is the identification of exactly how far the defining multilinear pattern of symmetric tensors, permutation invariance among identically typed index slots drawn from a single fixed vector space, extends before the pattern either becomes vacuous, ceases to apply meaningfully, or must be replaced by a genuinely different construction.

The Pattern in Its Native Range

What the Pattern Requires

The Component Constraint, requiring invariance of a tensor's components under permutation of its index positions, presupposes that every one of the d index positions plays an identical structural role: each must be an index into the same underlying vector space, with the same variance (all covariant, in the sense discussed under the Tensor Symmetric Component Transformation), so that permuting the positions is a meaningful operation in the first place. This requirement defines the native range of the pattern, and every result established throughout Tensor Symmetric Decomposition Structure, from the Rank Relation to the Alexander-Hirschowitz classification, is stated within this native range.

Boundary at Mixed-Valence Tensors

A tensor built from a mix of vector and covector slots, such as one representing a linear map or a more general multilinear map between different spaces, does not admit a permutation-invariance condition of the type used for symmetric tensors, because swapping a covariant slot with a contravariant one is not a well-defined operation without an auxiliary identification, such as an inner product, between the space and its dual; the symmetric tensor pattern simply does not extend across this boundary without first fixing such an identification, at which point one is studying a different, though related, structure rather than a genuine symmetric tensor in the sense used throughout this material.

Boundary at the Extremes of Order

Order Zero and Order One as Degenerate Cases

At order zero, a "symmetric tensor" is simply a scalar, and the permutation group acting on zero index slots is trivial, so the Component Constraint imposes no condition at all; at order one, the permutation group acting on a single index slot is likewise trivial, so every tensor of order one, meaning every vector, is automatically symmetric. These two boundary cases are consistent with, but essentially vacuous instances of, the general pattern, included in the graded structure of the symmetric algebra chiefly for completeness rather than because they carry any nontrivial symmetry content.

Genuine Content Beginning at Order Two

The Component Constraint first imposes a nontrivial condition at order two, where it reduces to the familiar requirement that a matrix equal its own transpose, discussed throughout the Matrix Case; this marks the practical beginning of the pattern's substantive range, with orders zero and one serving only as a boundary convention needed to keep the symmetric algebra's grading complete and its dimension formulas valid at every degree.

Boundary at Infinite Dimension

Finite Multilinear Algebra as the Assumed Setting

Every construction described throughout this material, the symmetrization operator, the catalecticant matrix, and the secant variety dimension counts underlying the Alexander-Hirschowitz theorem, implicitly assumes that the underlying vector space V is finite-dimensional, since these constructions rely on explicit, finite sums over bases, permutations, or monomials.

What Changes in the Infinite-Dimensional Setting

When V is infinite-dimensional, such as a Hilbert space, the multilinear pattern can still be defined, and the symmetric tensor power extends to a construction known as the symmetric Fock space in functional analysis and mathematical physics, but several finite-dimensional facts no longer transfer automatically: dimension counts become cardinalities requiring separate justification, convergence of infinite sums replacing finite averages must be addressed using the topology of the space, and the secant variety geometry underlying rank classification must be replaced by more delicate analytic notions of approximate or infinite decomposition. The Multilinear Pattern Boundary marks precisely this transition, beyond which the combinatorial and algebraic-geometric machinery developed for finite-dimensional symmetric tensors requires substantial reformulation rather than direct application.

Boundary at Small Characteristic

Where the Pattern's Consequences, Not the Pattern Itself, Break Down

Unlike the mixed-valence and infinite-dimensional boundaries, which concern whether the Component Constraint can be formulated at all, the characteristic boundary discussed under Tensor Symmetric Type Preservation concerns a case where the pattern remains perfectly well-defined but certain of its consequences, such as the clean separation of symmetry types via averaging projectors, fail; this is a boundary of consequence rather than of definition, and it is included here to distinguish it clearly from the definitional boundaries at mixed valence and infinite dimension.

Significance of Marking the Boundary

Preventing Overextension of the Theory

Explicitly identifying the Multilinear Pattern Boundary prevents the inadvertent application of finite-dimensional, characteristic-zero results, such as specific secant variety dimension formulas or explicit Waring rank bounds, to settings where the underlying assumptions no longer hold, ensuring that the rich decomposition theory built throughout Tensor Symmetric Decomposition Structure is invoked only within the range where its proofs and classifications remain valid.