commuting operators

Two operators commute when applying them in either order gives the same result: AB = BA. Most pairs of matrices do not commute — order matters in matrix multiplication — so commuting is a special, structurally meaningful relationship rather than the norm.

The structural payoff is invariance of eigenspaces. If AB = BA and v is an eigenvector of A with A v = lambda v, then A (B v) = B (A v) = lambda (B v), so B v is again in the eigenspace E_lambda. In words: B maps each eigenspace of A into itself. This lets you analyze B one eigenspace at a time, which is exactly what makes a shared eigenbasis possible.

Commuting is therefore the key that unlocks simultaneous diagonalization: a commuting family of diagonalizable operators has a common eigenbasis. The idea reaches far beyond linear algebra — commuting observables in quantum mechanics, commuting symmetries that can be diagonalized together, and the entire theory of how a set of operators can be simplified at once all rest on this single relation.