Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-6-sol)audited 2026-09-30
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.

Affine morphisms are relative spectra

Statement

Let f:X→S be a morphism of schemes. Then f is affine if and only if there is an affine-locally module-associated sheaf A of commutative unital OS-algebras and an isomorphism of S-schemes X≅Spec⁡SA. If f is affine, one may take A=f∗OX. Conversely, for any such A and S-isomorphism, f is affine and the induced isomorphism of OS-algebras is f∗OX≅A. No choice axiom is assumed.

Facts & Assumptions

Given: A scheme morphism f:X→S.

[F1]

A scheme morphism is affine exactly when the inverse image of every affine open of its target is affine; the empty scheme is affine (Affine morphisms).

[F2]

A sheaf of commutative unital OS-algebras is affine-locally module-associated when its restriction to every affine U=Spec⁡R is the module-associated sheaf of an R-algebra, with principal-open restrictions given by localization (Affine-local quasi-coherent algebras before general sheaf theory).

[F3]

The scheme morphism supplies the structure map of sheaves of rings f♯:OS→f∗OX, so f∗OX is an OS-algebra (Morphisms of ringed spaces).

[F4]

For an open V⊆S, (f∗OX)(V)=OX(f−1(V)), with restrictions induced by those of OX (Direct image of a sheaf along a continuous map).

[F5]

If f is affine, then on every affine U=Spec⁡R the restriction (f∗OX)∣U is the module-associated sheaf of the coordinate algebra of f−1(U), and principal-open sections and restriction maps are the corresponding localizations (Affine pushforward algebra localizes).

[F6]

An affine-locally module-associated algebra sheaf A has a relative spectrum π:Y=Spec⁡SA→S with π−1(U)≅Spec⁡Γ(U,A) for affine U; principal-open inverse images and their restriction maps are given by localization (Glue relative spectra of affine-local algebras).

[F7]

Global sections are quasi-inverse to the contravariant affine scheme--ring correspondence, which is natural for affine scheme morphisms (Affine schemes are contravariantly equivalent to commutative rings).

[F8]

Compatible isomorphisms between affine charts glue uniquely to an isomorphism of schemes respecting the chart maps (Gluing affine schemes along compatible open isomorphisms).

[F9]

Compatible local sheaf isomorphisms on an open cover glue uniquely to a sheaf isomorphism (Compatible local sheaves glue uniquely up to unique isomorphism).

[F10]

A morphism of schemes is a morphism of the underlying locally ringed spaces (Morphisms of schemes).

[F11]

An S-morphism is a scheme morphism commuting with the structure maps to S (Schemes and morphisms over a base).

Proof

technique · canonical affine-chart identifications and gluing
1.1F1F2F3F5

Suppose f is affine and put A=f∗OX, with its OS-algebra structure from [F3]. For each affine open U=Spec⁡R⊆S, the inverse image is affine by [F1]. The canonical identification in [F5] shows that A∣U is module-associated and that its principal-open restrictions are the required localizations. Thus A satisfies [F2]. This argument includes an empty inverse image, whose coordinate ring is 0.

1.2F3F4F6F7F11

Let π:Y=Spec⁡SA→S be the relative spectrum. For every affine U⊆S, [F6] gives π−1(U)=Spec⁡Γ(U,A). By [F4], Γ(U,A)=Γ(U,f∗OX)=Γ(f−1(U),OX). Both f−1(U) and π−1(U) are affine, so [F7] gives a canonical isomorphism between them. Its maps to U agree: the ring maps from Γ(U,OS) are the algebra structure maps of f∗OX.

1.3F4F6F7F8F11

These chart isomorphisms are compatible under restriction. Indeed, for an affine open W⊆U, the restriction Γ(f−1(U),OX)→Γ(f−1(W),OX) is exactly the restriction of f∗OX in [F4], while the relative-spectrum chart transition in [F6] is induced by the same restriction of A. Naturality in [F7] therefore identifies the restriction of the chart isomorphism over U with the one over W. Every overlap of two affine opens of S is covered by affine opens W contained in it, so these equalities give compatible chart isomorphisms on all overlaps. Applying [F8] yields an S-isomorphism X≅Y. The construction uses the full set of affine opens and canonical global-sections rings; it makes no simultaneous choice of affine presentations.

1.4F1F6F11

Conversely, let A be any algebra sheaf as in [F2], let Y=Spec⁡SA, and suppose e:X≅Y is an S-isomorphism. For every affine open U⊆S, [F6] makes π−1(U) affine, so f−1(U) is affine under e. By [F1], f is affine.

1.5F4F6F7F9F10F11

On an affine U=Spec⁡R, the isomorphism A∣U≅BU~ and [F6] identify π−1(D(r)) with Spec⁡(BU[φU(r)−1]) for every principal open D(r)⊆U. Global sections of this affine spectrum recover BU[φU(r)−1] by [F7]. By [F4] applied to π, these identifications are exactly the sections of (π∗OY)∣U on the principal-open basis, with the same restriction maps as A∣U. They give a canonical OU-algebra isomorphism (π∗OY)∣U≅A∣U. The identifications agree on overlaps because they are induced by the same restrictions in [F6]; [F9] glues them to π∗OY≅A on S. Finally, e and [F4] identify f∗OX with π∗OY, compatibly with the OS-algebra structures. Hence f∗OX≅A.

The zero ring is allowed throughout: when f−1(U)=∅, its ring of sections is 0 and its spectrum is empty. If S=∅, then X=∅ and the same construction gives the unique empty relative spectrum. For the identity morphism, A=OS and the local charts in [F6] are U=Spec⁡Γ(U,OS), so the identification is the identity. No AC is used. ∎

Depends on

Used by

Dependency tree · two levels

30 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources