1.2.17 Dual Basis Definition

Dual Basis Definition is the characterization of the dual basis, associated with a chosen basis of a vector space, as the unique basis of the dual space consisting of covectors defined so that each dual basis covector returns the value one when applied to its corresponding original basis vector and the value zero when applied to every other original basis vector. It supplies the precise construction linking a basis of a vector space to a corresponding basis of its dual, and it is the device that makes the natural pairing between vectors and covectors computable in explicit coordinates.

Constructing the Dual Basis

Given a finite-dimensional vector space with a chosen basis, the dual basis is defined by specifying, for each basis vector, a corresponding covector that acts as a coordinate-extraction functional: applied to that particular basis vector, it returns one, and applied to any other basis vector in the chosen set, it returns zero. This condition, sometimes called the biorthogonality condition, uniquely determines each dual basis covector, since a linear functional on a finite-dimensional space is completely determined once its value on every basis vector is specified.

The expression above defines the dual basis using the Kronecker delta, which equals one when its two indices coincide and zero otherwise, capturing precisely the biorthogonality condition that determines the dual basis from the original basis.

The Dual Basis Is Itself a Basis

The set of covectors constructed by this biorthogonality condition is not merely a convenient collection of functionals; it forms a genuine basis of the dual space, meaning it is linearly independent and spans the entire dual space. Because the original vector space is finite-dimensional, its dual space has the same finite dimension, and the dual basis, having exactly as many elements as the original basis, satisfies both requirements of a basis precisely because of how it was constructed from the biorthogonality condition.

Extracting Coordinates with the Dual Basis

One of the most immediate uses of the dual basis is coordinate extraction: applying the i-th dual basis covector to an arbitrary vector returns exactly the i-th coordinate of that vector relative to the original basis. This follows directly from the biorthogonality condition combined with the linearity of the dual basis covectors, since expanding the vector as a linear combination of basis vectors and applying the dual basis covector kills every term except the one matching its index, leaving only the corresponding coefficient.

This property explains why the dual basis is sometimes described informally as a set of coordinate-reading devices: each dual basis covector, applied to any vector, reads off precisely one coordinate of that vector, without requiring any other computation.

Dependence on the Original Basis

Just as a vector space admits many different bases, each choice of basis for the original space produces, through the biorthogonality condition, its own corresponding dual basis for the dual space. Changing the original basis therefore changes the dual basis as well, and the two changes are linked by a precise rule: if the original basis vectors transform according to a certain set of coefficients under a change of basis, the dual basis covectors transform according to the inverse of those same coefficients. This inverse relationship is the direct source of the distinction between contravariant components, which transform with the inverse coefficients relative to the basis vectors, and covariant components, which transform with the same coefficients as the basis vectors themselves.

Role in Tensor Algebra

The dual basis provides the concrete computational apparatus needed to express covectors, and more generally any tensor carrying covariant indices, in terms of explicit numerical components once a basis has been fixed. Every tensor of mixed type, when expressed relative to a basis and its associated dual basis, decomposes into components indexed partly by the original basis and partly by the dual basis, and the tensor transformation law governing how these components change under a change of basis is derived directly from the transformation behavior of the basis and dual basis together.