Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-adaptedPipeline-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Analyzer projections of a two-dimensional polarisation space and their overlaps

Statement

Let V be a two-dimensional complex inner product space with inner product linear in the first argument (Real and complex inner product spaces, with the inner product linear in the first argument, Real and complex inner-product spaces and their induced length), and let (e1,e2) be an orthonormal basis of V (Orthonormal families, complete orthonormal systems and Hilbert bases). For θ∈R put eθ:=cos⁡θ e1+sin⁡θ e2,eθ⊥:=−sin⁡θ e1+cos⁡θ e2, and for x∈V define Pθx:=⟨x,eθ⟩ eθ,Pθ⊥x:=⟨x,eθ⊥⟩ eθ⊥. Then:

(a) for every θ the maps Pθ and Pθ⊥ are orthogonal projections on V with ranges Ceθ and Ceθ⊥ (The Hilbert orthogonal projection onto a closed subspace, Hilbert projections are linear, self-adjoint and contractive), one has Pθ+Pθ⊥=I and PθPθ⊥=0, and PθPθ=Pθ=Pθ∗,PθPθ⊥=Pθ⊥Pθ=0;

(b) for all real φ,θ the unit vectors ψφ:=cos⁡φ e1+sin⁡φ e2 satisfy ⟨eθ,ψφ⟩=cos⁡(θ−φ),⟨ψφ,Pθψφ⟩=∥Pθψφ∥2=∣⟨eθ,ψφ⟩∣2=cos⁡2(θ−φ), and ⟨ψφ,Pθ⊥ψφ⟩=sin⁡2(θ−φ);

(c) the assignments θ↦Pθ and θ↦Pθ⊥ are 2π-periodic, Pθ+π=Pθ and Pθ+π/2=Pθ⊥, and for every z∈C and x∈V one has Pθ(zx)=zPθx and ∥Pθ(zψφ)∥2/∥zψφ∥2=cos⁡2(θ−φ) whenever z≠0; in particular the number cos⁡2(θ−φ) depends only on the ray [ψφ], not on the representative;

(d) the pair (Pθ,Pθ⊥) is a two-element family of mutually orthogonal projections on V with Pθ+Pθ⊥=I, the spectral values 1 and 0 of the bounded self-adjoint operator Pθ have projections Pθ and Pθ⊥, respectively. These two alternatives may be labelled +1 and −1 in a declared analyser; those labels are not the spectral values of Pθ. The eigenspace of Pθ at eigenvalue 1 is Ceθ and the eigenspace at eigenvalue 0 is Ceθ⊥.

No physical interpretation of V, of the pair (e1,e2) or of the projections Pθ is asserted.

Facts & Assumptions

Given: the two-dimensional complex inner product space V with inner product linear in the first argument, the orthonormal basis (e1,e2) of V, the vectors eθ,eθ⊥,ψφ∈V and the maps Pθ,Pθ⊥:V→V.

[F1]

Inner product and norm. For all x,y∈V and a,b∈C, ⟨ax+by,z⟩=a⟨x,z⟩+b⟨y,z⟩ and ⟨x,ay+bz⟩=a‾⟨x,y⟩+b‾⟨x,z⟩; ⟨x,x⟩≥0 with equality only for x=0; and ∥x∥2=⟨x,x⟩ (Real and complex inner product spaces, with the inner product linear in the first argument, Real and complex inner-product spaces and their induced length).

[F2]

Orthonormality. ⟨ei,ej⟩=δij for i,j∈{1,2} (Orthonormal families, complete orthonormal systems and Hilbert bases).

[F3]

Finite expansion. For an orthonormal basis (u1,u2) of V, write x=a1u1+a2u2. Taking inner products with uj gives aj=⟨x,uj⟩ by [F1] and orthonormality, and expanding ⟨x,x⟩ gives ∥x∥2=∣a1∣2+∣a2∣2. Thus the finite expansion and norm identity hold without invoking the countable-choice Hilbert-space version of Parseval.

[F4]

Finite-dimensional orthogonal projection. The projection onto a subspace M of V assigns the unique M-component of x in the decomposition V=M⊕M⊥; equivalently, its value lies in M and the residual lies in M⊥. This is the finite-dimensional agreement clause of The Hilbert orthogonal projection onto a closed subspace.

Proof

technique · direct; compute the overlaps of the displayed vectors from the orthonormality of $(e_{1},e_{2})$ and verify the projection identities

Given: The data of the statement and [F1]-[F4].

1.1F1F2algebra

[The rotated pairs are orthonormal] By [F1] and [F2], ⟨eθ,eθ⟩=cos⁡2θ+sin⁡2θ=1, ⟨eθ⊥,eθ⊥⟩=sin⁡2θ+cos⁡2θ=1 and ⟨eθ,eθ⊥⟩=−cos⁡θsin⁡θ+sin⁡θcos⁡θ=0; hence (eθ,eθ⊥) is an orthonormal family of the two-dimensional space V, therefore an orthonormal basis of V.

1.2F1F2algebra

[Overlaps and the cosine identity] By [F1] and [F2], ⟨eθ,ψφ⟩=cos⁡θcos⁡φ+sin⁡θsin⁡φ=cos⁡(θ−φ) and ⟨eθ⊥,ψφ⟩=−sin⁡θcos⁡φ+cos⁡θsin⁡φ=sin⁡(φ−θ). Then ⟨ψφ,Pθψφ⟩=⟨ψφ,⟨ψφ,eθ⟩eθ⟩=∣⟨ψφ,eθ⟩∣2=cos⁡2(θ−φ) because cos⁡(θ−φ) is real, and ∥Pθψφ∥2=∣cos⁡(θ−φ)∣2∥eθ∥2=cos⁡2(θ−φ); similarly ⟨ψφ,Pθ⊥ψφ⟩=sin⁡2(θ−φ). This is (b).

2.1F1F3step 1.1F4algebra

[The projections are orthogonal projections] For x∈V, Pθx=⟨x,eθ⟩eθ, so Pθ(Pθx)=⟨⟨x,eθ⟩eθ,eθ⟩eθ=⟨x,eθ⟩⟨eθ,eθ⟩eθ=Pθx by [F1] and step 1.1: Pθ2=Pθ. For the adjoint, ⟨Pθx,y⟩=⟨⟨x,eθ⟩eθ,y⟩=⟨x,eθ⟩⟨eθ,y⟩=⟨x,⟨y,eθ⟩eθ⟩=⟨x,Pθy⟩ for all x,y, so Pθ∗=Pθ; the same computation with eθ⊥ in place of eθ gives Pθ⊥ 2=Pθ⊥=Pθ⊥ ∗. For either unit vector u=eθ or u=eθ⊥, Pux=⟨x,u⟩u lies in Cu and ⟨x−Pux,u⟩=0, so the residual is orthogonal to every multiple of u. Hence [F4] identifies Pu as the orthogonal projection onto Cu; its range is exactly that line because Puu=u. By the finite norm identity in [F3], ∥Pux∥2=∣⟨x,u⟩∣2≤∥x∥2, so it is bounded. The line is closed: if aru→v, then ar=⟨aru,u⟩→⟨v,u⟩ (use the finite coordinate norm identity), and therefore v=⟨v,u⟩u.

3.1F1F3step 1.1step 2.1algebra

[Complementarity] By step 1.1 and [F3] applied to the orthonormal basis (eθ,eθ⊥), every x∈V satisfies x=⟨x,eθ⟩eθ+⟨x,eθ⊥⟩eθ⊥=Pθx+Pθ⊥x; hence Pθ+Pθ⊥=I. Applying Pθ to the same expansion and using Pθeθ⊥=⟨eθ⊥,eθ⟩eθ=0 gives PθPθ⊥=0, and likewise Pθ⊥Pθ=0. Since Pθ=I−Pθ⊥ and Ceθ∩Ceθ⊥={0} by orthogonality of the nonzero vectors eθ,eθ⊥, the eigenspace of Pθ at eigenvalue 1 is Ceθ and at eigenvalue 0 is Ceθ⊥; this proves (a) and (d).

4.1F1step 1.2algebra∎

[Periodicity and phase invariance] Since eθ+π=−eθ and eθ+π/2=eθ⊥, the identities Pθ+π=Pθ and Pθ+π/2=Pθ⊥ follow from the definition of Pθ and [F1]; in particular Pθ is 2π-periodic. For z≠0 and the unit vector ψφ one has ∥zψφ∥2=∣z∣2, and Pθ(zψφ)=zPθψφ by linearity of x↦⟨x,eθ⟩ in the first argument of [F1]; hence ∥Pθ(zψφ)∥2/∥zψφ∥2=∥Pθψφ∥2=cos⁡2(θ−φ) by step 1.2. This proves (c) and completes the proof.

Remarks

  • What the lemma is not. The lemma is finite-dimensional linear algebra: no dynamics, no measurement postulate and no physical identification of eθ with an apparatus direction is asserted. The physical reading of the projections as the outcome structure of a polarisation analyser is fixed in The fixed-mode polarisation model of the framework.
  • The identity cos⁡2(θ−φ) is classical and quantum at once. The same projection formula governs the classical transverse field amplitude and the quantised polarisation amplitude; the comparison, and the reason that the coincidence of the numbers does not establish field quantisation, is the subject of the companion examples page of this pair, which is not a prerequisite of this lemma.

Depends on

Dependency tree · 2 levels

Sources