Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-generatedprecheck passaudited 2026-09-05
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.

Every smooth function admits arbitrarily fine strong-topology perturbations whose differential is transverse to zero, supported away from a closed set where transversality already holds

Statement

Let M be a smooth manifold, let f:MR be smooth, and let AM be closed. Assume that df is transverse to the zero section on an open neighbourhood of A. Then every neighbourhood of f in the strong C topology contains a smooth function g such that:

  • gf is supported in MA;
  • g=f on some open neighbourhood of A; and
  • dg is transverse to the zero section of TM.

Equivalently, every strong neighbourhood of f contains a Morse function that agrees with f near A.

Facts & Assumptions

Given: A smooth manifold M, a smooth function f:MR, a closed set AM, an open neighbourhood V of A on which df is transverse to the zero section, and a chosen strong smooth neighbourhood U of f.

[F1]

A smooth function is Morse exactly when its differential section is transverse to the zero section (A smooth function is Morse if and only if its differential section is transverse to the zero section).

[L1]

The zero section is an embedded submanifold, parametric transversality makes the bad parameter set null for a transverse family, and a null subset of a positive-dimensional parameter manifold has dense complement (The zero section is a smooth embedding, Parametric transversality, A null set has dense complement in a positive-dimensional manifold).

[L2]

Smooth bump functions exist on prescribed compact subsets inside open sets, and locally finite sums of smooth functions are smooth (A manifold bump for a compact set inside an open set, A locally finite sum of smooth functions is smooth).

[L3]

Every smooth manifold admits a smooth proper exhaustion function (Every smooth manifold admits a smooth proper exhaustion function).

[A1]

A basic strong neighbourhood of f is determined by uniform bounds for finitely many derivatives on each member of a locally finite compact family. The derivative order and tolerance may vary from one member to another. Consequently one may choose a finite order rn and a sufficiently small derivative bound on each shell so that every locally finite perturbation satisfying all the shell bounds stays inside U.

Proof

technique · direct
1.1

Choose an open set W with AW and WV. By [L3], fix a smooth proper exhaustion h:M[0,), set K1=K0:=, and for n1 let Kn:=h1([0,n]). Using [A1], choose integers rnn+1 and positive tolerances εn so small that whenever smooth functions un satisfy supp(un)(intKn+1Kn1)W and unCrn(Kn+1)<εn for every n, the locally finite sum u:=nun gives f+uU. The varying orders rn dominate every derivative order imposed by the chosen strong neighbourhood on the corresponding shell.

givenL3A1choose
2.1

Construct inductively gn so that g0=f, the correction un:=gngn1 has the support and size prescribed in step 1.1, and dgn is transverse to the zero section on a neighbourhood of WKn. Suppose gn1 has been chosen. The set Cn where dgn1 is not transverse on Kn is compact. If Cn=, put un:=0 and gn:=gn1; openness of the transversality locus gives the required neighbourhood of Kn, so all the inductive conditions hold. Hence assume that Cn is nonempty. It is disjoint from both W and Kn1: on W every earlier correction vanishes and df is transverse, while transversality near Kn1 is the inductive hypothesis. Thus Cn has a finite coordinate cover whose chart closures lie in (intKn+1Kn1)W. Using [L2], choose bump-supported coordinate functions ϕi on these charts so that their differentials span every cotangent fibre on a neighbourhood of Cn. The resulting parameter space is positive-dimensional: a nontransverse point cannot occur when dimM=0, and a nonempty coordinate cover in positive dimension supplies at least one coordinate function.

L2step 1.1givenconstruct
3.1

In the nonempty case of step 2.1, consider the finite-dimensional family gn1,a:=gn1+iaiϕi. On a neighbourhood of Cn, its parameter derivatives span the cotangent fibres. On the compact remainder of Kn, the section dgn1 is already transverse, so after restricting to a sufficiently small parameter ball, transversality there persists for every parameter. Hence the total family is transverse to the zero section on a neighbourhood of Kn. By [L1], the bad slice parameters form a null set; because the parameter ball is positive-dimensional, its complement is dense. Choose an in that complement and close enough to 0 that un:=i(an)iϕi satisfies the bound in step 1.1. Put gn:=gn1+un. Transversality is open, so dgn is transverse on a neighbourhood of Kn, and it still agrees with df near W. Together with the empty case handled in step 2.1, this completes the induction.

L1step 2.1choose
4.1

The shell supports are locally finite, so [L2] makes u:=n1un smooth. Put g:=f+u. Step 1.1 gives gU. Every correction is supported outside W, so g=f on the neighbourhood W of A and gf is supported in MA. Fix pM and choose n with pKn. Step 3.1 makes dgn+1 transverse near p, while every um with mn+2 has support disjoint from Kn and therefore vanishes near p. Thus g agrees near p with gn+1, so dg is transverse there. Since p was arbitrary, dg is transverse everywhere.

L2step 1.1step 3.1
5.1

By [F1], the final function g is Morse. This proves the relative strong-topology density statement.

F1step 4.1

Depends on

Used by

Dependency tree · two levels

26 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