A complete vector space over the real or complex numbers equipped with an inner product whose induced norm makes the space complete for Cauchy sequences; supports orthogonality, projections, and unitary evolution operators.

Principle

The inner product ⟨·,·⟩ induces a norm ‖v‖=√⟨v,v⟩ and completeness (every Cauchy sequence converges) guarantees limits exist for sequences defined by orthogonal expansions, ensuring analytic and spectral techniques apply.

Demonstration

L^2(Ω), the space of square-integrable functions on a domain Ω with inner product ⟨f,g⟩=∫_Ω f* g dx, is a Hilbert space; orthonormal bases allow expansion of elements as convergent series and define orthogonal projections onto closed subspaces.

Misapplication

Treating a pre‑Hilbert space (an inner product space that is not complete) as if it were complete leads to erroneous conclusions about convergence and existence of limits—e.g., finite sequences are dense but not complete until completion is taken.

Consequence

Hilbert space structure permits results such as existence of orthogonal projections onto closed subspaces, expansion in orthonormal bases, and well‑posed definitions of adjoint and unitary operators relevant to quantum theory and functional analysis.

Reversal

A Banach space may be complete under a norm but lack an inner product; absence of an inner product removes canonical notions of orthogonality and projection by inner‑product minimization.

Boundary

Requires a positive‑definite inner product and metric completeness; excludes indefinite inner product spaces (Krein spaces), non‑complete inner product spaces, and purely algebraic inner product structures without a topology.

Semantic Tension

Finite-dimensional Euclidean spaces are Hilbert spaces with trivial completion, but infinite-dimensional Hilbert spaces introduce subtleties (bases, separability, compact operators) absent in finite-dimensional linear algebra.

Synthesis

A Hilbert space is a complete inner product space whose geometry of angles and lengths plus completeness enable orthogonal decompositions, projection operators, and the analytic machinery of spectral and unitary operator theory.