Alphabeta Math
Pipeline-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.

Analytic double fibrations and Bolker visibility — Examples and Counterexamples

1 · Prerequisites

2 · Logical flowchart

3 · Definitions, theorems and proofs

None yet.

4 · Examples, counterexamples and false statements

ExampleConstruction: Literature-sourcedVerification: AI-adaptedverified 2026-10-22 (gpt-6.1-sol)Open item page →

A missing-angle covector has no data-side canonical pair

Example

Let Θ⊂S1 be a nonempty proper open set of line normals, invariant under θ↦−θ, and suppose S1∖Θ‾ contains an open arc. Restrict the Euclidean Radon incidence relation to

ZΘ={(θ,s,x)∈(Θ×R)×R2:s=x⋅θ}.

Give ZΘ the standard nowhere vanishing analytic total density ∣dθ ds dt∣, with x=sθ+tθ⊥, and constant weight 1. This is a properly supported analytic double fibration. For any x0∈R2 and any nonzero covector ξ0∈Tx0∗R2 with ξ0/∣ξ0∣∉Θ, there are no (θ,s)∈Θ×R and no data covector η such that (θ,s,η;x0,ξ0)∈CZΘ. In particular, this object covector has no data-side canonical pair for the restricted-angle transform.

Facts & Assumptions

Given: The open antipodally symmetric set Θ, the incidence submanifold ZΘ, and the invisible covector (x0,ξ0).

[F1]

The Euclidean Radon transform on R²: the line with normal θ and signed offset s is given by s=x⋅θ.

[F2]

Visible covectors for limited-angle data: a nonzero plane covector is visible exactly when its unit direction lies in Θ; Θ is antipodally symmetric.

[F3]

Distributions, conormal kernels, and cotangent relations: (θ,s,η;x,ξ) belongs to the incidence relation exactly when (θ,s,x;η,−ξ) is a nonzero conormal to ZΘ.

[F4]

Incidence-coordinate conormal kernel and canonical relation: if F(θ,s,x)=0 defines the incidence hypersurface, its conormal covectors are scalar multiples of dF.

[F5]

Analytic double-fibration transform and canonical relation: both incidence projections must be analytic submersions and the support condition is properness over compact object supports.

[F6]

Real analytic inverse and implicit functions: a regular zero of a real-analytic scalar function is locally an analytic hypersurface.

Proof

Given: The data in the example.

1.1F1F6givenalgebra

The defining function F(θ,s,x)=s−x⋅θ is real analytic and has ∂sF=1. Thus dF is surjective at every point of ZΘ, and [F6] makes ZΘ an analytic hypersurface of dimension 3.

1.2F1F3F4algebra

Use a local angular coordinate α on S1, so ∂αθ=θ⊥. The defining function F(θ,s,x)=s−x⋅θ has differential dF=ds−(x⋅θ⊥)dα−θ⋅dx. By [F4], a conormal covector is τdF for a nonzero scalar τ. Matching its data components and object component with the sign convention [F3] gives ηs=τ,ηα=−τx⋅θ⊥,ξ=τθ. Thus any object covector in the relation is parallel to the measured normal θ.

2.1F5step 1.1algebrachoose

A tangent vector satisfies s˙−x˙⋅θ−x⋅θ˙=0. Given any data tangent pair (θ˙,s˙), choose x˙=(s˙−x⋅θ˙)θ; then πG has a tangent lift and is a submersion. Given any object tangent vector x˙, choose θ˙=0, s˙=x˙⋅θ; hence πX is a submersion. Both maps are analytic restrictions of product projections and 4>3>2≥2. The displayed analytic total density is nowhere zero in these incidence coordinates. For compact K⊂R2 and compact L⊂Θ×R, the preimage of L under πG∣πX−1(K) is closed in pr⁡1(L)×K, which is compact because pr⁡1(L)⋐Θ. Thus the restricted projection is proper and [F5] applies.

3.1F2step 1.2given∎

If (θ,s,η;x0,ξ0)∈CZΘ, step 1.2 implies ξ0/∣ξ0∣=θ or −θ. Since Θ=−Θ, this would place ξ0/∣ξ0∣ in Θ, contradicting the hypothesis. No such data covector exists.

Geometric diagram

The diagram chooses Θ to be two small antipodal arcs around the horizontal normals. The vertical covector ξ0 lies in the open missing arc, so it cannot be parallel to a measured normal. The proof uses the conormal equation above, not the drawing.

»1»2£»0=j»0junmeasurednormaldirection
ExampleConstruction: Literature-sourcedVerification: AI-adaptedjudge pass (gpt-6-luna)verified 2026-10-22 (gpt-6.1-sol)Open item page →

A conjugate antipode breaks global Bolker injectivity

Example

Let X=G=S2 be the unit round sphere with its standard real-analytic structure, and give Z the density induced by the product round metric. Let

Z={(n,x)∈S2×S2:n⋅x=0},

so n∈G labels the oriented great circle Cn={x∈X:n⋅x=0}. The transform uses the analytic density induced on Z and the constant analytic weight 1. Then G←Z→X is an analytic double fibration, and the geodesic from x to its antipode −x along Cn has conjugate endpoints. Moreover, for every (n,x)∈Z, the nonzero data covector x♭∈Tn∗G is paired by the canonical relation with both (x,−n♭) and (−x,n♭). Thus the left projection of the canonical relation is not injective, so the Bolker condition fails globally.

Facts & Assumptions

Given: The unit round sphere, the incidence hypersurface Z, and the round arclength density on its great-circle fibers.

[F1]

Analytic double-fibration transform and canonical relation: both incidence projections must be analytic submersions, and the required support condition is properness over compact object sets.

[F2]

Bolker condition for a double-fibration transform: Bolker requires the left cotangent projection to be globally injective.

[F3]

Incidence-coordinate conormal kernel and canonical relation: for a defining function F, the conormal bundle is generated by dF and the relation uses (η,−ξ).

[F4]

Riemannian metric and riemannian manifold: the unit round metric identifies tangent vectors and cotangent vectors by v↦v♭.

[F5]

A Euclidean sphere is a regular level set with tangent hyperplanes: TxS2=x⊥ for every x∈S2.

[F6]

Real analytic inverse and implicit functions: a regular zero of a real-analytic scalar function is locally an analytic graph.

[F7]

Geodesic variation: a smooth family whose parameter curves are affinely parametrized geodesics is a geodesic variation.

[F8]

Jacobi field: a Jacobi field satisfies Dt2J+R(J,γ˙)γ˙=0.

[F9]

Conjugate points along a geodesic and their multiplicity: endpoints are conjugate when a nonzero Jacobi field vanishes at both ends.

[F10]

Great circles as round-sphere geodesics: constant-speed great-circle parametrizations are geodesics of the round sphere.

[F11]

Variation field of a geodesic variation is a Jacobi field: the variation field of a smooth geodesic variation is Jacobi; this implication uses no choice axiom.

Proof

Given: The incidence Z and its round geometry.

1.1F5F6givenalgebra

Put F(n,x)=n⋅x on S2×S2. At a point with F=0, the vector x lies in TnS2 by [F5], and the tangent variation (n˙,x˙)=(x,0) satisfies dF(n˙,x˙)=x⋅x=1. Thus dF≠0; [F6] makes Z=F−1(0) an analytic hypersurface of dimension 3 in the 4-dimensional product.

2.1F1F5step 1.1algebrachoose

A tangent vector to Z satisfies x⋅n˙+n⋅x˙=0. Given any n˙∈TnS2, set x˙=−(x⋅n˙)n; then x˙∈TxS2 by [F5], and the tangent equation holds. Hence πG is a submersion. Given any x˙∈TxS2, set n˙=−(n⋅x˙)x; the same calculation proves πX is a submersion. Both maps are analytic restrictions of product projections. Since Z is compact, πG is proper on πX−1(K) for every compact K⊂X. The density induced by the positive-definite analytic product round metric has strictly positive analytic coefficient in every local incidence chart. The dimensions satisfy 4>3>2≥2, so [F1] applies.

2.2F2F3step 1.1algebra

The defining differential on tangent vectors to the product is dF=(x♭,n♭) by the metric identification [F4]. Therefore by [F3] the relation at (n,x) consists of (n,η;x,ξ)=(n,τx♭;x,−τn♭),τ≠0. At the antipodal point −x, choose multiplier −τ. Its data covector is (−τ)(−x)♭=τx♭, the same as at x, while the object point changes from x to −x. Both are nonzero relation points. Hence the left projection is not injective, and [F2] implies failure of global Bolker.

3.1F7F8F9F10F11givenalgebra∎

Let v=n×x, a unit tangent vector at x. For s near zero put ns=cos⁡(s)n+sin⁡(s)v, vs=ns×x, and γs(t)=cos⁡(t)x+sin⁡(t)vs,0≤t≤π. Each ns is a unit vector orthogonal to x, and vs is a unit vector orthogonal to x; hence γs parametrizes a great circle at unit speed and is a geodesic by [F10]. Thus [F7] makes this a geodesic variation. Its endpoints are fixed: γs(0)=x and γs(π)=−x. Differentiating in s at zero gives J(t)=∂sγs(t)∣s=0=sin⁡(t)(v×x)=−sin⁡(t)n, because (n×x)×x=−n when n⋅x=0 and ∣x∣=1. By [F11] this field is Jacobi and hence satisfies the Jacobi equation in [F8]. It is nonzero for 0<t<π and vanishes at both endpoints, so [F9] makes the endpoints conjugate. The proof uses no choice axiom; the sphere has no boundary, so this example lies outside the simple-surface hypotheses.

Geometric diagram

The great circle is drawn as a circle in its plane, with n normal to the plane and x,−x antipodal. At n, the data covector is represented by the same tangent vector x for both relation points; the object covectors at the antipodes are opposite. The calculation in step 2.1, rather than the schematic, establishes noninjectivity.

x¡xn´=x[»=¡n[»0=n[Cn
ExampleConstruction: Literature-sourcedVerification: AI-adaptedverified 2026-10-22 (gpt-6.1-sol)Open item page →

A visible Euclidean Radon conormal satisfies Bolker

Example

Let X=R2, G=S1×R, and

Z={(θ,s,x):s=x⋅θ}.

Write x=sθ+tθ⊥ on Z, use the analytic total density ∣dθ ds dt∣ and constant weight 1, so R is the full Euclidean Radon transform. This is a properly supported analytic double fibration. Its canonical relation satisfies the global Bolker condition. Using the Euclidean metric to identify covectors with vectors and its associated norm, for every x∈R2 and every ξ∈Tx∗R2∖0, set

θ=ξ/∣ξ∣,τ=∣ξ∣,s=x⋅θ,ηs=τ,ηθ=−τ x⋅θ⊥.

Then (θ,s;ηθ,ηs;x,ξ) is a canonical pair. Thus every nonzero Euclidean covector is visible. In particular, for a compactly supported distribution f, the analytic wavefront recovery theorem implies that if (x,ξ)∈WF⁡A(f), then this data covector belongs to WF⁡A(Rf).

Facts & Assumptions

Given: The full Euclidean Radon incidence, its constant analytic weight and line density, and a nonzero covector (x,ξ).

[F1]

The Euclidean Radon transform on R²: lines are parameterized by s=x⋅θ, with θ⊥=(−θ2,θ1).

[F2]

Incidence-coordinate conormal kernel and canonical relation: the defining function F(θ,s,x)=s−x⋅θ gives conormal components by differentiating F, with the relation sign (η,−ξ).

[F3]

Bolker condition for a double-fibration transform: the Bolker condition requires a singleton left fiber and injective differential at its point.

[F4]

Analytic wavefront recovery under Bolker: data-side analytic regularity at a Bolker pair implies object-side analytic regularity.

[F5]

Analytic double-fibration transform and canonical relation: both incidence projections must be analytic submersions and the support condition is properness over compact object supports.

[F6]

Real analytic inverse and implicit functions: a regular zero of a real-analytic scalar function is locally an analytic hypersurface.

Proof

Given: The incidence and the covector (x,ξ).

1.1F1F6givenalgebra

The defining function F(θ,s,x)=s−x⋅θ is analytic and has ∂sF=1. Thus [F6] makes Z=F−1(0) an analytic hypersurface of dimension 3.

1.2F1F2algebra

In a local angle coordinate α with ∂αθ=θ⊥, the defining function has differential dF=ds−(x⋅θ⊥)dα−θ⋅dx. By [F2], the canonical relation consists of ξ=τθ,ηs=τ,ηθ=−τx⋅θ⊥,τ≠0. These signs follow by matching (η,−ξ)=τdF.

2.1F5step 1.1algebrachoose

A tangent vector satisfies s˙−x˙⋅θ−x⋅θ˙=0. Given any data tangent pair (θ˙,s˙), choose x˙=(s˙−x⋅θ˙)θ; this lifts it to TZ, so πG is a submersion. Given any object tangent vector x˙, take θ˙=0 and s˙=x˙⋅θ, so πX is a submersion. Since πX−1(K) is parameterized by the compact set S1×K for every compact K⊂R2, the restriction to G is proper. The specified analytic total density is nowhere zero. The dimensions satisfy 4>3>2≥2, so [F5] applies.

2.2F1F2F3step 1.1algebra

For fixed (θ,s,ηs,ηθ) in the image, τ=ηs≠0 and x=sθ−ηθτθ⊥,ξ=τθ. Thus the left fiber is a singleton. To check the differential, let a tangent variation be in the kernel of the left projection. Then θ˙=0, s˙=0, τ˙=η˙s=0, and 0=s˙=x˙⋅θ,0=η˙θ=−τ x˙⋅θ⊥. Since τ≠0 and (θ,θ⊥) is a basis, x˙=0 and ξ˙=0. Hence the left projection is an injective immersion, so [F3] gives global Bolker.

2.3F2step 1.2given

For the specified object covector, θ=ξ/∣ξ∣ and τ=∣ξ∣ satisfy ξ=τθ. With s and η chosen in the statement, step 1.2 supplies the remaining relation equations. Therefore the data-side canonical pair exists for every nonzero ξ.

3.1F4step 2.1step 2.2step 2.3contrapositive∎

The contrapositive of [F4] shows that (x,ξ)∈WF⁡A(f) forces (θ,s;ηθ,ηs)∈WF⁡A(Rf). Its hypotheses hold by the analytic nonvanishing density, properness and full Bolker calculations in steps 1.1–2.2.

Geometric diagram

The line has normal θ, so its conormal covectors are scalar multiples of θ. The point x determines the offset s=x⋅θ; the tangential component of the data covector records −x⋅θ⊥. The drawing fixes these conventions only; the proof computes the full data covector and verifies Bolker directly.

µ?µx¢µ=sx»=¿µµ»visible

Sources