8.2.3 Tensor Dummy Index Area

Tensor Dummy Index Area is the conceptual domain of tensor index notation concerned specifically with dummy (or bound) indices: those index letters occurring exactly twice in a single term, once as a superscript and once as a subscript, and therefore summed away under the Einstein summation convention rather than surviving into the result. It covers the rules governing when an index is dummy, how such indices may be renamed, and how they interact with contraction, as distinct from free indices, which are never summed.

Identifying a Dummy Index

The One-Up-One-Down Rule

An index is dummy within a term precisely when the same letter appears exactly twice in that term, once in an upper (superscript) position and once in a lower (subscript) position; this repetition triggers the implicit summation convention, and the index is fully summed over its entire range before the term is combined with anything else.

Dummy Indices Do Not Survive Into the Result

Because a dummy index is summed over its full range, no single value of it labels the outcome; once the sum is carried out, the letter itself disappears from the expression entirely, in contrast to a free index, which remains attached to the result as one of its labeling slots.

Renaming Within the Dummy Index Area

Any Unused Letter Will Do

Since the meaning of a dummy index depends only on the pattern of its repetition — one occurrence up, one occurrence down, within the same term — and not on which specific letter is used, a dummy index may always be renamed to any other letter not already in use elsewhere in that term, without changing the value or meaning of the expression.

Why Renaming Is Necessary in Practice

Renaming a dummy index becomes a practical necessity when two separate expressions, each already using a particular letter as a dummy index, are combined into one larger expression; one of the two must first be renamed to an unused letter, since otherwise the two originally independent summations would be incorrectly merged into a single, unintended contraction over both original ranges at once.

Diagram of the Dummy Index Area

Dummy Indices and Contraction

Contraction Is Performed Through Dummy Indices

Every contraction — the operation of pairing one superscript slot with one subscript slot and summing — is expressed in the notation precisely by making the paired slots into a dummy index; the dummy index area and the operation of contraction are, in this sense, two views of the same underlying mechanism: contraction is what a dummy index does, and a dummy index is the notational signature that a contraction is taking place.

Multiple Dummy Indices in One Term

A single term may contain several independent dummy indices simultaneously, each summed separately, as in T^{ij}_{k} S^{k}_{ij}, where i, j, and k are all dummy; the total operation performed is a triple sum, with all three sums nested or combined depending on how the term is evaluated, and the result carries no free indices at all, reducing to a scalar.

Distinguishing the Dummy Index Area from the Free Index Area

Opposite Roles, Shared Notation

The same superscript-and-subscript notation is used for both free and dummy indices; the distinction between the two areas is made entirely by counting occurrences within a single term, not by any difference in how the letters themselves are written. An index letter's role as free or dummy is therefore local to the term in which it is examined, and the identical letter could in principle be dummy in one term of an expression and free in a different, unrelated term.

Complementary Contributions to a Valid Expression

A well-formed tensor expression relies on both areas working correctly together: free indices must balance consistently across every term of an equation, while dummy indices must each be properly paired, one up and one down, within their own term; an expression that gets either of these two requirements wrong fails to represent a valid tensor operation.