15.6.5 Tensor Symmetric Multilinear Form Role

Tensor Symmetric Multilinear Form Role is the generalization of the symmetric bilinear tensor role to arbitrary rank n, describing the function a totally symmetric rank-n tensor performs when it is used to act on n vector arguments and return a single scalar, invariant under any rearrangement of those arguments. In this role, the tensor is not merely an array of numbers subject to an equality constraint but an operational device: it consumes an unordered collection of n vectors and yields a measurement of that collection, with the total symmetry of the tensor guaranteeing that the measurement does not depend on any artificial ordering imposed on the inputs.

Understanding this role clarifies why totally symmetric tensors of higher rank arise naturally whenever a quantity depends jointly on several vectors in a way that treats those vectors on equal footing, generalizing the rank-2 case where a symmetric tensor measures a pairwise, order-independent relationship between two vectors to situations involving three, four, or more vectors evaluated jointly and symmetrically.

The Tensor as an n-Ary Symmetric Map

Multiple Interchangeable Input Slots

In its multilinear role, a totally symmetric tensor T of rank n defines a map B taking n vector arguments and producing one scalar output, required to be linear in each argument when the other n minus one arguments are held fixed, and required by total symmetry to be invariant under every permutation of the n argument slots, not merely under interchange of any two of them.

The Role Generalizes, Rather Than Replaces, the Bilinear Role

Setting n equal to two in this general role recovers exactly the symmetric bilinear tensor role, so the multilinear form role is not a separate concept but the same underlying idea, an order-independent measurement built from a tensor, extended from two arguments to an arbitrary number of arguments.

Distinguishing the Multilinear Role From Other Rank-n Roles

Roles Requiring Fewer Symmetric Arguments

A rank-n tensor can be used in roles that consume fewer than n vector arguments, for instance by contracting some indices against fixed vectors while leaving others as free output indices, producing a lower-rank tensor rather than a scalar; such partial-evaluation roles do not require, and do not by themselves exhibit, the full argument-order invariance associated with the multilinear form role described here.

Total Symmetry as a Precondition for This Specific Role

The multilinear form role, in the fully symmetric sense discussed here, specifically requires the tensor to satisfy the total symmetry component constraint; a rank-n tensor lacking this constraint still defines a multilinear map on n arguments, but that map lacks the permutation-invariance property that defines the symmetric multilinear form role.

Consequences of the Role for Interpretation

Aggregating Rather Than Ordering Information

Because the role produces a scalar from an unordered collection of n vectors, it is naturally suited to represent aggregate or joint measurements, such as a coupling strength among several directions simultaneously, where no single vector among the n is distinguished as first, second, or otherwise ordered relative to the rest.

The Diagonal Evaluation as a Single-Vector Summary

Evaluating the tensor in its multilinear role on n copies of one vector produces the associated homogeneous degree-n polynomial, offering a single-vector summary of the tensor's behavior; this summary is complete, by the polarization identity, meaning no information about the tensor's role on distinct argument tuples is lost by focusing on this diagonal evaluation alone.

The Role Within the Broader Symmetric Tensor Framework

Rank-n Instance of a General Correspondence

The multilinear form role, together with its generalized slot-exchange operator, generalized quadratic relation, and generalized matrix-like array representation, together establish that a totally symmetric tensor of any rank can be equivalently understood either as a constrained array of components or as a permutation-invariant multilinear map, with the two viewpoints connected by exactly the same correspondence machinery regardless of rank.

Increasing Richness at Higher Rank

While the rank-2 instance of this role reduces to the well-understood theory of symmetric bilinear forms and symmetric matrices, the same role at rank 3 or higher retains all the same defining properties, permutation invariance, multilinearity, and a faithful associated homogeneous polynomial, but the classification and structural theory of such higher-rank symmetric multilinear maps is considerably less complete, reflecting the greater combinatorial complexity of measuring several vectors jointly and symmetrically rather than only two.