An eigenvalue of a linear operator A on a vector space over a field is a scalar λ for which there exists a nonzero vector v (an eigenvector) satisfying A v = λ v; equivalently λ lies in the point spectrum of A when A−λI fails to be injective.

Principle

Eigenvalues arise from the algebraic condition det(A−λI)=0 in finite dimensions and reflect invariant one-dimensional subspaces; they determine spectral decomposition in diagonalizable cases and are central to stability, modal analysis, and representation of linear actions.

Demonstration

A symmetric real matrix has real eigenvalues and an orthonormal basis of eigenvectors; a rotation in R^2 about the origin with no fixed directions has no real eigenvalues but two complex conjugate eigenvalues on the unit circle; nilpotent matrices have eigenvalue 0 with algebraic multiplicity equal to matrix size in the extreme case of a single Jordan block.

Misapplication

Assuming every linear operator is diagonalizable or that geometric multiplicity equals algebraic multiplicity; treating singular values or pseudospectrum as interchangeable with eigenvalues; or inferring eigenvectors exist over the real field when eigenvalues are complex without extending scalars.

Consequence

Eigenvalues and associated eigenvectors enable diagonalization, spectral decompositions, modal analysis of differential equations, principal component methods in statistics, and precise characterization of invariant dynamics and resonances.

Reversal

Viewing the complementary concept—singular values—measures norms of action on vectors without requiring invariant directions and yields a different, norm‑sensitive spectral picture; an operator can have no eigenvalues in a field while having nonzero action measured by singular values.

Boundary

Defined for linear maps on vector spaces; in infinite-dimensional settings eigenvalues form only part of the spectrum (point spectrum), and operators may have continuous or residual spectrum without eigenvalues. Field of scalars, domain, and closure properties affect existence and algebraic multiplicities.

Semantic Tension

Tension with singular values and the full operator spectrum: eigenvalues capture algebraic invariant directions and may be unstable under perturbation for nonnormal operators, while singular values are robust norm quantities and the full spectrum includes non-point parts relevant in infinite dimensions.

Synthesis

An eigenvalue is a scalar characterizing an invariant direction of a linear map where action reduces to scalar multiplication; together with eigenvectors and multiplicities it informs diagonalization, stability, and modal structure, but must be distinguished from singular values and broader spectral concepts in infinite dimensions.