7.1 Tensor Component Scope

Tensor Component Scope is the precise delimitation of what the word "component" refers to when discussing a tensor: a single scalar value, drawn from the base field, obtained by evaluating the tensor against a specific choice of basis vectors and dual basis covectors for each of its slots, and nothing broader than that. Establishing this scope prevents the term from being used loosely to mean the tensor itself, a sub-tensor, or a vector-valued piece of a tensor, all of which are related but distinct notions that fall outside what "component," properly used, denotes.

What Falls Inside the Scope

A Single Scalar Per Full Index Assignment

A component of a tensor T of type (p,q) is the scalar obtained by supplying one basis index to each of the p + q slots simultaneously — writing Tⱼ¹ᵢ¹... requires every slot to be assigned a specific value before the result is a component. Supplying only some of the slots and leaving others free produces not a component but a lower-order tensor-valued object, which lies outside the scope of the term.

Scalars, Not Vectors or Sub-Tensors

A component is always an element of the base field F (typically the real or complex numbers), never a vector or a smaller tensor; this is what distinguishes "the components of T" from "a slice of T," where the latter refers to fixing only some of the indices and leaving a genuinely tensor-valued (not merely scalar) object as the result.

What Falls Outside the Scope

The Tensor as a Whole Is Not a Component

Although casual speech sometimes uses "the tensor's components" to refer to the tensor's coordinate representation collectively (the whole array), the scope of "a component," singular, is strictly one entry of that array; conflating the entire array with a single component is a scope error that can lead to confusion about, for instance, how many components a tensor has versus what a single component's numerical value is.

Basis-Independent Quantities Are Not Components

A component, by its dependence on a chosen basis, is never itself a basis-independent quantity; invariants computed from the components — such as the trace or determinant of a (1,1) tensor's matrix — lie outside the scope of "component" even though they are computed from components, because these invariants remain fixed under a change of basis while any individual component generally does not.

Diagram of the Scope Boundary

Why the Scope Distinction Matters

Precision in Stating Formulas

Formulas such as the component count relation N = d^(p+q) are, by their nature, statements about how many components exist; misapplying "component" to mean the whole tensor would make such a formula nonsensical (a tensor is not d^(p+q) "tensors"), so respecting the scope of the term is a precondition for these formulas to even be correctly stated.

Clarity When Discussing Transformation

Because it is individual components, not the tensor as a whole or its invariants, that change value under a change of basis, keeping the scope of "component" narrow and precise is essential to correctly stating transformation laws: the transformation rule describes how each component's numerical value changes, while the tensor itself, understood as the basis-independent object, does not change at all.