Adjoint of a Matrix: Definition, Examples & Formula

The adjoint of a matrix, also called the adjugate, is a special matrix built from the cofactors of a square matrix. If A is an n × n matrix, its adjoint adj(A) is the transpose of the matrix of cofactors of A. It is the key ingredient in the formula for the inverse of a matrix: A⁻¹ = adj(A) / det(A).

The adjoint matters because it turns the difficult problem of inverting a matrix into a mechanical recipe: compute the cofactors, transpose them, and divide by the determinant. It sits behind the matrix method for solving linear equations, Cramer's rule, and many computer calculations where inverses of small matrices are needed repeatedly.

Key Concepts

• Minor Mᵢⱼ: the determinant of the smaller matrix left after deleting row i and column j of A.
• Cofactor Cᵢⱼ: the signed minor, Cᵢⱼ = (−1)^(i+j) × Mᵢⱼ. The sign is positive when i + j is even and negative when i + j is odd.
• Cofactor matrix: the n × n matrix whose entry in row i, column j is Cᵢⱼ.
• Adjoint adj(A): the transpose of the cofactor matrix, so the (i, j) entry of adj(A) is Cⱼᵢ.
• Square matrices only: minors, cofactors and the adjoint are defined only for square matrices.

Formulas

• Definition: adj(A) = [Cⱼᵢ], the transpose of the cofactor matrix. This works for every square matrix, even when det(A) = 0.
• Fundamental identity: A × adj(A) = adj(A) × A = det(A) × Iₙ, where Iₙ is the identity matrix. The adjoint "cancels" the original matrix down to a multiple of the identity.
• Inverse formula: A⁻¹ = (1 / det(A)) × adj(A), valid only when det(A) ≠ 0.
• Scalar multiple: adj(kA) = k^(n−1) × adj(A) for an n × n matrix and any number k.

Worked Example

Find adj(A) for A = [[2, 1], [3, 4]].

1. Compute the four cofactors. C₁₁ = +(4) = 4. C₁₂ = −(3) = −3. C₂₁ = −(1) = −1. C₂₂ = +(2) = 2.
2. Arrange them into the cofactor matrix: [[4, −3], [−1, 2]].
3. Transpose it to get the adjoint: adj(A) = [[4, −1], [−3, 2]].
4. Verify with the fundamental identity: A × adj(A) = [[2, 1], [3, 4]] × [[4, −1], [−3, 2]] = [[8 − 3, −2 + 2], [12 − 12, −3 + 8]] = [[5, 0], [0, 5]] = 5 × I₂.
5. Since det(A) = 2 × 4 − 1 × 3 = 5, the check passes, and the inverse is A⁻¹ = (1/5) × [[4, −1], [−3, 2]].

Common Misconceptions

• The adjoint is not just the transpose. It is the transpose of the cofactor matrix. Skipping the cofactor step, or skipping the transpose, is the most common mistake.
• The adjoint exists even when the inverse does not. If det(A) = 0, adj(A) is still perfectly defined; only the inverse formula breaks down.
• Sign errors in cofactors. Remember the checkerboard of signs: any position with an odd i + j gets a minus sign.

Real-World Applications

• Matrix inversion: the standard formula for inverting 2 × 2 and 3 × 3 matrices in engineering and physics software.
• Solving linear systems: the matrix method X = A⁻¹B uses the adjoint behind the scenes.
• Cramer's rule: another method of solving linear equations, built directly from cofactors.
• Computer graphics: inverting transformation matrices that position and rotate objects in 3D scenes.

The adjoint of a matrix is the transpose of its cofactor matrix, and it is the heart of the inverse formula A⁻¹ = adj(A) / det(A). Master the sign pattern of cofactors, never skip the transpose, and verify with A × adj(A) = det(A) × I.