The Born Rule

Four ingredients, all derived from or established by the axioms:

• The pre-state is a complex, separable Hilbert space H with dim(H) ≥ 3 (AP09 + AP10). Separability (countable orthonormal basis) is satisfied by all physical Hilbert spaces: L²(R³) is separable by construction.
• Actualisation targets pure states (Proposition 1, from Axiom R).
• The probability of each outcome is a Gleason non-contextual function of the state and the projector (Proposition 2, from Axiom R + Axiom B).
• The probability measure is orthogonally additive and normalised (Lemma 1, from Axioms S + B + R).

Conditions (2)–(4) together define a frame function on H: a non-negative, orthogonally additive function μ on projection operators, defined on projectors alone (not on context), with ∑ μ(P_i) = 1 for every resolution of the identity.

This is exactly the object to which Gleason’s theorem applies.

Gleason’s Theorem (A. M. Gleason, 1957).

Let H be a separable Hilbert space over C with dim(H) ≥ 3. Let μ be a frame function on H: a non-negative, orthogonally additive function on projection operators such that for any complete set of mutually orthogonal projectors {P_i} with ∑ P_i = I, we have ∑ μ(P_i) = 1. Then there exists a unique density operator ρ such that μ(P) = Tr(ρ · P) for all projectors P.

Gleason’s theorem is a mathematical result about probability measures on Hilbert spaces. It is not a physical postulate.

It is in the same category as the spectral theorem or Noether’s theorem: a piece of mathematics that the axioms use to derive consequences.

The axioms provide the physical hypotheses (complex Hilbert space, definite records, non-contextuality, orthogonal additivity, normalisation). Gleason provides the mathematical conclusion.

Theorem 1 (The Measure). The unique probability measure on outcomes of actualisation, consistent with the axioms, is the Born rule: μ(P) = Tr(|ψ⟩⟨ψ| · P).

For rank-1 projectors P_a = |a⟩⟨a|, this gives μ(P_a) = |⟨a|ψ⟩|².

Proof. The pre-state is a complex, separable Hilbert space H with dim(H) ≥ 3 (AP09, AP10). By Proposition 1, actualisation targets pure states |ψ⟩.

By Proposition 2, the probability assigned to projector P is Gleason non-contextual: it depends only on |ψ⟩ and P, not on which resolution of the identity P belongs to.

By Lemma 1, the probability measure is orthogonally additive. By normalisation (Axioms S + B), probabilities for a complete set of outcomes sum to 1. These conditions define a frame function on H.

By Gleason’s theorem, the unique frame function on a complex, separable Hilbert space of dimension ≥ 3 is μ(P) = Tr(ρ · P).

Since the state being actualised is a pure state |ψ⟩ (Proposition 1), ρ = |ψ⟩⟨ψ| and therefore μ(P) = Tr(|ψ⟩⟨ψ| · P). For rank-1 projectors, this is |⟨a|ψ⟩|². This is the Born rule. ∎

You have just watched four axioms force the most mysterious equation in quantum mechanics. Not choose it. Not postulate it. Force it. There is no other measure.

Gleason’s theorem does not say the Born rule is a good choice. It says the Born rule is the only choice. The probability of quantum outcomes was never a postulate.

It was a consequence — waiting in the structure of Hilbert space for the axioms to land on it.

Intuition: why the Born rule is forced. The elementary core of Gleason’s result (restricted to rank-1 projectors) can be sketched without the full Gleason machinery.

Non-contextuality and normalisation together imply that the weight function f (assigning probabilities to rays) is additive: f(s + t) = f(s) + f(t) for overlaps of orthogonal subspaces. Non-negativity implies monotonicity.

A monotone additive function on [0,1] with f(0) = 0 and f(1) = 1 is the identity: f(t) = t. Therefore μ(P_a) = |⟨a|ψ⟩|² — the Born rule. This is the Cauchy functional equation route.

Gleason’s theorem generalises this to arbitrary projectors (not just rank-1) and to the full lattice structure, which is why the paper uses Gleason rather than the Cauchy argument directly.

Example: Spin measurement in full Hilbert space. Consider an electron with spin. The full Hilbert space is H = L²(R³) ⊗ C², where L²(R³) encodes spatial degrees of freedom and C² encodes spin.

This space is infinite-dimensional (dim ≥ 3 satisfied). It is separable (L²(R³) has a countable orthonormal basis).

Prepare the spin in state |ψ⟩ = α|↑⟩ + β|↓⟩ with |α|² + |β|² = 1. Measure spin along z. The projectors are P_↑ = |↑⟩⟨↑| and P_↓ = |↓⟩⟨↓| (rank-1).

By Theorem 1: μ(P_↑) = |⟨↑|ψ⟩|² = |α|². μ(P_↓) = |⟨↓|ψ⟩|² = |β|². These are the Born rule probabilities. They sum to 1.

The spin-only Hilbert space C² has dimension 2, where Gleason’s theorem does not apply directly.

But Theorem 1 is proved on the full physical Hilbert space H = L²(R³) ⊗ C², where dim ≥ 3 is satisfied.

The Born rule on C² is inherited by restriction: the probability measure on the full space, restricted to the spin subspace, gives the same Born rule probabilities.

The dim ≥ 3 requirement applies to the full pre-state, not to effective subspaces used in calculations. For effective two-dimensional systems treated in isolation, the POVM extension of Gleason’s theorem (Busch, 2003) reaches the same measure; the composite-space argument above is sufficient for the physical Hilbert space.

Uniqueness is crucial. Gleason’s theorem does not merely say the Born rule is consistent with the axiom-derived conditions — it says the Born rule is the only measure that satisfies them. There is no alternative.

Any non-contextual, orthogonally additive, normalised probability assignment on a complex, separable Hilbert space of dimension ≥ 3 is the Born rule. The probability structure of actualisation is not a choice. It is forced.