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.
Adapted excellent Morse functions exist on compact cobordisms
Statement
Assume (The Axiom of Countable Choice ()). For every compact collared triad there are an adapted excellent Morse function and an adapted complete downward gradient-like field , both agreeing with the product model on smaller neighborhoods in the fixed collars: near , near , and is its negative Riemannian gradient for a metric that is a product metric there. Here adapted complete has the collar-extension meaning of Morse function adapted to a cobordism: is the restriction of a complete smooth field on a boundaryless collar extension of ; a trajectory is followed in only until it exits a face.
Facts & Assumptions
Given: A compact collared triad with its fixed disjoint collars and .
Boundary product function on a collared cobordism supplies with the stated product formulas near the faces, , , and near the boundary, under .
A manifold bump for a compact set inside an open set supplies smooth bumps equal to one near a compact set with support in a prescribed open set.
Under , Parametric transversality says that if the evaluation of a finite-dimensional smooth family is transverse to an embedded submanifold, the parameters whose slices fail transversality form a null set.
A smooth function is Morse exactly when its differential section is transverse to the zero section; its critical Hessian is the vertical derivative of that section at a zero (A smooth function is Morse if and only if its differential section is transverse to the zero section, A smooth map transverse to an embedded submanifold).
A Morse function on a compact manifold has finitely many critical points gives finiteness, and Separating critical values far from the boundary separates their values by an arbitrarily small perturbation supported away from the boundary, preserving the critical points and their Hessians.
Under , Every smooth manifold admits a riemannian metric supplies a background metric; Morse lemma supplies the coordinates at each critical point; Riemannian gradient defines its metric gradient.
Under , Compactly supported smooth vector fields are complete makes a compactly supported smooth field on a boundaryless manifold complete.
Downward gradient-like vector fields for a Morse function requires off the critical set and in the preceding Morse coordinates. Morse function adapted to a cobordism adds the boundary directions and collar-extension completeness.
Proof
Start with from [F1]. Choose such that has its product formula on the two closed collar strips . Write for the union of the open strips and put ; this is a compact subset of . Consider all nested relatively compact interior chart neighborhoods and bumps supported in and equal to one near . They cover : each point has such nested chart neighborhoods and a bump by [F2]. Compactness selects finitely many triples covering , without choosing data simultaneously at every point. In coordinates on , extend by zero to . These are smooth and have compact support away from ; on their differentials are the coordinate basis .
For finitely many parameters , set . The finite union of the supports is a compact subset of the interior, so has a positive margin from both and there. On the compact collar annulus , is nonzero. Choose an open parameter ball about zero so small that every retains the value margin on the supports and has nonzero differential on this annulus; uniform bounds on the finitely many functions and first derivatives give this choice. On each perturbation is zero. Thus has no critical point outside , has the original product formula near the faces, and takes its values in with endpoint fibers exactly the faces.
The evaluation , , is transverse to the zero section. Indeed at a zero some contains , and the parameter derivatives span the vertical cotangent fiber. Projection to the normal quotient of the zero section is therefore surjective. There are no zeros outside by step 2.1. Apply [F3] to this finite-dimensional family; its bad parameters are null, so every sufficiently small open parameter ball contains a good parameter. Choose one and call its slice . By [F4] it is Morse, including at all interior points, and the collar contains no critical point. If is empty no perturbation is needed; in dimension zero every Hessian is the invertible map of the zero vector space, so is already Morse and this family step is omitted.
By [F5] the critical set of is finite. Choose a closed boundary collar and apply the value-separation result in [F5] by a perturbation supported in the interior and small enough to retain the positive endpoint margin on its compact support. The result has the same critical points and Hessians, distinct critical values, and equals near ; it still has endpoint fibers exactly . Thus is adapted excellent.
Fix a background metric by [F6]. On smaller face collars use , where is the restriction of the background metric to ; in disjoint Morse-coordinate neighborhoods of the finitely many critical points use the Euclidean metric. Take cutoffs equal to one on still smaller neighborhoods and supported in these mutually disjoint collars and charts, using explicit collar cutoffs and [F2]. If these cutoffs are and their local metrics are , the metric is smooth and positive definite, is product near the faces, and is Euclidean near each critical point. Set . Off the critical set . In a Morse chart it is , as required by [F8]. On the face collars it is and respectively, hence points outward at and inward at .
Construct the completeness carrier explicitly. Append to each face its negative collar , identifying with the face and using the fixed collar coordinates for . Signed-collar charts and the interior charts of give a boundaryless smooth manifold containing ; the product transition maps give compatibility, and the finite gluing along compact faces gives a Hausdorff second-countable carrier. Continue the collar fields and into the appended collars. Multiply there by a smooth scalar cutoff equal to one for and zero for , leaving unchanged on . The resulting smooth field on has support contained in the compact set , hence is complete by [F7]. Its restriction is ; integral curves in are its ambient curves restricted to the time intervals before a boundary exit, by local flow uniqueness. If the boundary is empty, take and .
The pair has all the required Morse, value-separation, descending-model, boundary and collar-extension properties by steps 4.1–6.1. Only is used: it is inherited from the boundary-product, parametric transversality, metric and complete-field suppliers. The finite chart and bump selection in step 1.1 and the single finite-dimensional parameter choice in step 3.1 use no full Axiom of Choice.
Depends on
Smooth cobordism triad for Morse theory
Morse function adapted to a cobordism
Boundary product function on a collared cobordism
Separating critical values far from the boundary
Parametric transversality
A manifold bump for a compact set inside an open set
A smooth map transverse to an embedded submanifold
A smooth function is Morse if and only if its differential section is transverse to the zero section
A Morse function on a compact manifold has finitely many critical points
Morse lemma
Every smooth manifold admits a riemannian metric
Compactly supported smooth vector fields are complete
Riemannian gradient
Downward gradient-like vector fields for a Morse function
The Axiom of Countable Choice ()