pith. sign in

arxiv: 2604.00939 · v2 · submitted 2026-04-01 · 🧮 math.GT

Cerf Diagrams and Hatcher-Wagoner Invariants for Barbell Maps

Xiayu Tan This is my paper

Pith reviewed 2026-05-13 22:00 UTC · model grok-4.3

classification 🧮 math.GT
keywords barbell mapspseudo-isotopyHatcher-Wagoner invariantsCerf diagramsdiffeomorphismsgeometric topologyimmersed barbellmanifold invariants
0
0 comments X

The pith

In dimensions six and higher every pseudo-isotopy whose first Hatcher-Wagoner invariant vanishes is isotopic to a composition of standard barbell pseudo-isotopies.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper constructs two explicit standard barbell pseudo-isotopies for each half-unknotted implanted (i,n-i)-barbell in a manifold. Each standard construction produces the barbell diffeomorphism and has a Cerf diagram consisting of only a single eye, so the Hatcher-Wagoner invariants become directly computable. Using these generators it proves that for n at least 6 any pseudo-isotopy with vanishing first invariant can be isotoped to a product of the i=2 and i=3 cases. In dimension 4 the same approach extends to immersed barbells and realizes prescribed values of the second induced invariant under explicit conditions on the manifold and its characteristic classes.

Core claim

For a half-unknotted implanted (i,n-i)-barbell beta in M^n we construct two standard barbell pseudo-isotopies, each yielding the barbell diffeomorphism and possessing a single-eye Cerf diagram. For n greater than or equal to 6 this implies every pseudo-isotopy with vanishing first Hatcher-Wagoner invariant is isotopic to a composition of such standards with i equal to 2 or 3. In dimension 4 we prove that for every s in Z_2, sigma in pi_2 M and gamma in pi_1 M satisfying s=0 or w_2^M(sigma) nonzero there exists a standard immersed barbell pseudo-isotopy realizing the second induced invariant Theta(f_beta) equal to (s,sigma) dot [gamma].

What carries the argument

The standard barbell pseudo-isotopy: one of two explicit pseudo-isotopies attached to a half-unknotted implanted (i,n-i)-barbell that realizes the barbell diffeomorphism while keeping the Cerf diagram a single eye and the Hatcher-Wagoner invariants computable by formula.

If this is right

  • Explicit formulas are supplied for the invariants of the standard barbell pseudo-isotopies when i equals 2 and for a special family when i equals 3.
  • In dimension 4 the second induced Hatcher-Wagoner invariant takes all values of the form (s, sigma) dot gamma whenever s is zero or the second Stiefel-Whitney class of sigma is nonzero.
  • The result reduces the classification of pseudo-isotopies with vanishing first invariant to the study of compositions of these two families of generators.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The explicit generators may permit inductive calculation of higher-order invariants by successively replacing complicated pseudo-isotopies with barbell products.
  • The same barbell technique could be tested on the 6-sphere or other simply connected manifolds where the space of pseudo-isotopies is already partially understood.
  • If the decomposition holds, the connected components of the space of pseudo-isotopies would be determined by the algebraic data carried by the i=2 and i=3 barbells alone.

Load-bearing premise

The constructions assume the manifold contains half-unknotted implanted (i,n-i)-barbells and that the chosen embeddings keep every Cerf diagram simple with exactly one eye.

What would settle it

A concrete counterexample would be a pseudo-isotopy on a manifold of dimension at least 6 whose first Hatcher-Wagner invariant is zero yet which cannot be isotoped to any finite composition of the i=2 and i=3 standard barbell pseudo-isotopies.

Figures

Figures reproduced from arXiv: 2604.00939 by Xiayu Tan.

Figure 1
Figure 1. Figure 1: A general barbell and data needed to compute Hatcher-Wagoner invariants [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: This is a Cerf diagram with 2 eyes, each eye begins with the birth and finishes with the [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The implanted barbell δ4 in M = S 1 × D3 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Finger-push R0 to get R such that R ∩ S = 2 points [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Do surgery along S to get an embedded torus TR in S 1 × D3#S 1 × S 3 with the standard S 1 (green one), this TR determines an element in π1(Emb(S 1 , S1 × D3#S 1 ×S 3 ), ∗) which is a loop of handle decompositions H = {Ht , t ∈ S 1} (a loop of (1,2)-handle pair) of S 1 × D3 × I which results in the barbell diffeomorphism δ4 = τW where W = (R, S). In this section we first describe the dotted version of triv… view at source ↗
Figure 6
Figure 6. Figure 6: Z ′ ∪ Z¯′′ = M × I Recall what we obtained in the last section: For any half-unknotted barbell β = βi,n−i = (R, S, γ) in Mn , we find a loop of framed S i−1 , i.e. an element in π1(Emb(S i−1×Dn−i+1, M#S i−1× S n−i+1), ∗) which represents (by the map psνS) that barbell diffeomorphism. And thus, it gives a Cerf diagram containing a single eye of (i − 1, i)-handle pair which results in the barbell diffeo￾morp… view at source ↗
Figure 7
Figure 7. Figure 7: Element in π1(Emb(S 1 × D3 , S1 × D3#S 1 × S 3 ), ∗) which represents barbell δk. Recall the loop of (1,2)-handle constructions for Cerf diagram resulting in δk (again we let k = 4 for simplicity), as shown in [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Dotted version of the loop of (1,2)-handle pair. The dark blue is the dotted [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Movement of γ • t : D2 ,→ S 1 × D3 . The figures illustrate γ • 0 , γ• t , γ• 1 . description of the mid-ball after performing a barbell. In [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Pulling γ1 back to γ0(therefore pulling γ • 1 back to γ • 0 ) is equivalent to doing the barbell diffeomorphism β = (R0, S, γ). 17 [PITH_FULL_IMAGE:figures/full_fig_p017_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: The mid-ball β • 1 becomes β • 2 = ∂(β • 1 ∪ h2,b ∪ h2,r) where h2,b is the 2-handle with attaching S 1 b the brown one and core D2 b = {S tube along the brown arc}, h2,r is the 2-handle with attaching S 1 r the red one and core D2 r = {R tube along the red arc}. As a result, β • 2 = (β • 1 \ (S 1 b × D2 ∪S 1 r × D2 ))∪(D2 b ×S 1 ∪ D2 r ×S 1 ). Therefore, β • 2 ∩γ • 2 = the green S 1 and the blue I in the… view at source ↗
Figure 12
Figure 12. Figure 12: Here we draw a local region of TR near the self-intersection point in an R 3 -slice. p = γt0 ∩ γt1 , q ∈ γt , and the blue region is the local part of γ • t in an R 3 -slice. When t > t1, the blue disk ⊂ γ • t will be finger-pushed along γt to ensure γ • t is embedded. and let ∗0 = γ ∩ S be the base point. For each i, find a path δ B i ⊂ β = (R0, S, γ) which is a path from ∗0 to pi ∈ S 1 i . Also, for eac… view at source ↗
Figure 13
Figure 13. Figure 13: Here we illustrate immersed torus TR and the standard γ • 0 on the left and the desired γ • 1 on the right. If TR is embedded, γ • 1 ∩ β • 0 = I ∩ S 1 where I is in blue and S 1 is in red. But since TR has a self-intersection point, by finger-pushing a small disk near the intersection point along γt as we said in [PITH_FULL_IMAGE:figures/full_fig_p023_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: An illustration of β = (Rσ, S, γ = γ1 ∗ γ2) its Cerf diagram being a single eye of (n − 2, n − 1)-handle pair resulting in the immersed barbell diffeomorphism with respect to β such that Θ(fβ) = (0, σ) · [α]. In particular, when n ≥ 5, β is embedded. Proof. For σ ∈ π2M with wM 2 (σ) = 0, the result just follows from Corollary 6.3. For wM 2 (σ) = 1. We find an immersed 2-sphere Rσ representing σ, and an im… view at source ↗
Figure 15
Figure 15. Figure 15: An illustration for the proof of Corollary 6.4 [PITH_FULL_IMAGE:figures/full_fig_p024_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: The description for the implanted (3, n−3)-barbell, with R0 = ∂D(E) and β • 0 ∩R0 = S 1 which is a fiber of the Hopf fibration R0 → S 2 . Proposition 7.2. For the constructed β = (R0, S, γ), the fβ ∈ ker Σ we constructed in section 4, whose Cerf diagram containing a single eye of (n − 3, n − 2)-handle pair, satisfies Θ(fβ) = (1, 0) · [γ]. Proof. We review the procedure in section 4 in the case i = 3, firs… view at source ↗
Figure 17
Figure 17. Figure 17: Here we draw TR = S 2 × S 1 tubeγR0, γ• 0 and γ • 1 = γ • 0 tubeγR0 in the dotted version. Then we apply Lemma 4.5 and use the strategy of one-parameter version of dotted and 0-framed replacement to get fβ. To calculate Θ(fβ), like what we did in section 5, we consider β • 2∩γ • 2 = I⊔S 1 and two natural framings on ν(β • 2 , M×I)|S1 = ν(S 1 , γ• 2×I). Note that ν(S 1 , γ• 2×I) = ν(S 1 , γ• 2 )⊕E′ and ν(β… view at source ↗
Figure 18
Figure 18. Figure 18: On the left, we draw how R0 (thus γ • 1 ) intersects β • 1 locally, and thus eA is induced by (π ∗T S2 )|π−1(p) where π stands for the Hopf fibration π : R0 = S 3 → S 2 . On the right we draw a section of (π ∗T S2 )|π−1(p) induced by v ∈ TpS 2 , which is the 1-framing on the (1,1) torus knot, which is the usual unknot. 27 [PITH_FULL_IMAGE:figures/full_fig_p027_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: D1 is obtained from D0 by two embedded surgeries: The first one is by attaching a trivial i-handle hi on D0, this gives the resulting ball D′ 0 = D0#S i × S n−1−i , then we attach a (i+ 1)-handle hi+1 with the attaching Di+1 being the standard one tubed along the loop to another S i+1, which is a perturbed Hopf fibration over S i where S i is obtained from the core of hi . See [5] for detailed description… view at source ↗
read the original abstract

For a half-unknotted implanted $(i,n-i)$-barbell $\beta=\beta_{i,n-i}$ in $M^n$, we construct two specific pseudo-isotopies, which we denote by standard barbell pseudo-isotopies, both resulting in that barbell diffeomorphism, each having a Cerf diagram only containing a single eye and with easily computable Hatcher-Wagoner invariants. We give an explicit formula for $\beta_{2,n-2}$ and a special class of $\beta_{3,n-3}$. Using this we show that for $n\geq 6$, every pseudo-isotopy with vanishing first Hatcher-Wagoner invariant can be isotoped to a composition of standard barbell pseudo-isotopies with $i=2$ or $3$. In dimension $n=4$, we further generalize the constructions and computations to half-unknotted immersed barbell diffeomorphisms and prove that for every $s\in \mathbb{Z}_2, \sigma\in \pi_2 M,\gamma\in \pi_1 M$ with $s=0 \text{ or }w_2^M(\sigma)\neq0$, there exists a standard immersed barbell pseudo-isotopy $f_\beta$ with the second induced Hatcher-Wagoner invariant $\Theta(f_\beta)=(s,\sigma)\cdot [\gamma]$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper constructs two standard barbell pseudo-isotopies for each half-unknotted implanted (i,n-i)-barbell β in M^n; both have single-eye Cerf diagrams and explicitly computable Hatcher-Wagoner invariants. Explicit formulas are supplied for β_{2,n-2} and a special class of β_{3,n-3}. The central generation theorem asserts that for n≥6 every pseudo-isotopy with vanishing first Hatcher-Wagoner invariant is isotopic to a composition of these standard barbell maps with i=2 or 3. In dimension 4 the constructions are extended to immersed barbell diffeomorphisms and the second induced invariant Θ(f_β) is shown to realize every pair (s,σ)·[γ] under the stated parity condition on w_2^M(σ).

Significance. If the generation statement is correct, the work supplies an explicit set of generators for the kernel of the first Hatcher-Wagoner invariant in high dimensions, together with concrete Cerf-diagram representatives whose invariants can be read off by inspection. The n=4 immersed case further gives a computable family of examples realizing prescribed values of the second invariant, which may be useful for low-dimensional diffeomorphism problems.

major comments (2)
  1. [§5] §5 (Generation theorem): the reduction of an arbitrary pseudo-isotopy with vanishing first HW invariant to a composition of standard barbell maps is asserted to preserve single-eye Cerf diagrams, yet the argument is given only for half-unknotted implanted barbells; no verification is supplied that the isotopy process itself cannot introduce additional eyes or critical points when the ambient manifold is arbitrary.
  2. [§6] §6 (n=4 immersed case): the existence statement for f_β realizing Θ(f_β)=(s,σ)·[γ] when s=0 or w_2^M(σ)≠0 relies on the immersed barbell remaining half-unknotted after immersion; the manuscript does not exhibit an explicit isotopy or Cerf-diagram computation confirming that the second invariant is unaffected by the immersion.
minor comments (2)
  1. [Abstract] The abstract refers to “a special class of β_{3,n-3}” without defining the class; the definition should appear in the statement of the main theorem or in §3.
  2. [§3] Notation for the two standard barbell pseudo-isotopies associated to a single β is introduced without a uniform symbol; a single subscripted notation would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address the major comments point by point below.

read point-by-point responses

  1. Referee: [§5] §5 (Generation theorem): the reduction of an arbitrary pseudo-isotopy with vanishing first HW invariant to a composition of standard barbell maps is asserted to preserve single-eye Cerf diagrams, yet the argument is given only for half-unknotted implanted barbells; no verification is supplied that the isotopy process itself cannot introduce additional eyes or critical points when the ambient manifold is arbitrary.

    Authors: The generation theorem reduces an arbitrary pseudo-isotopy with vanishing first Hatcher-Wagoner invariant to a composition of standard barbell pseudo-isotopies supported on half-unknotted implanted (2,n-2)- or (3,n-3)-barbells via high-dimensional Cerf theory. The standard constructions are given explicitly with single-eye diagrams. In the revised manuscript we will add a paragraph verifying that the connecting isotopy may be chosen generic so that any extra critical points appear in canceling pairs (possible precisely because the first invariant vanishes) and can therefore be eliminated without introducing new eyes, uniformly for arbitrary ambient manifolds. revision: yes

  2. Referee: [§6] §6 (n=4 immersed case): the existence statement for f_β realizing Θ(f_β)=(s,σ)·[γ] when s=0 or w_2^M(σ)≠0 relies on the immersed barbell remaining half-unknotted after immersion; the manuscript does not exhibit an explicit isotopy or Cerf-diagram computation confirming that the second invariant is unaffected by the immersion.

    Authors: The immersed barbell diffeomorphisms are obtained by immersing a half-unknotted barbell while preserving the local half-unknotted condition in a tubular neighborhood. We will add an explicit isotopy from the immersed model to its embedded counterpart together with the corresponding Cerf-diagram computation; under the stated parity condition the additional double points do not alter the value of the second invariant Θ, confirming that every prescribed pair (s,σ)·[γ] is realized. revision: yes

Circularity Check

0 steps flagged

Derivation chain is self-contained via explicit constructions

full rationale

The paper supplies direct, explicit formulas for the standard barbell pseudo-isotopies (including for β_{2,n-2} and a class of β_{3,n-3}) together with their single-eye Cerf diagrams and the resulting Hatcher-Wagoner invariants computed from those diagrams. The central generation statement for n≥6 is obtained by composing these explicitly constructed objects; no step reduces a claimed prediction or uniqueness result to a fitted parameter, a self-citation chain, or a definitional tautology. The constructions are presented as independent of the target statement, and the isotopy claim follows from the given embeddings and diagram simplifications rather than from any circular renaming or imported ansatz.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work relies on standard background in Cerf theory, pseudo-isotopy spaces, and Hatcher-Wagoner invariants without introducing new free parameters or invented entities.

axioms (2)
  • domain assumption Existence and standard properties of Cerf diagrams for pseudo-isotopies of diffeomorphisms
    Invoked throughout the construction of the standard barbell pseudo-isotopies.
  • domain assumption Properties of Hatcher-Wagoner invariants for pseudo-isotopies
    Used to define the vanishing condition and to compute the invariants from the diagrams.

pith-pipeline@v0.9.0 · 5541 in / 1418 out tokens · 45345 ms · 2026-05-13T22:00:22.314932+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

10 extracted references · 10 canonical work pages

  1. [1]

    Knotted 3-balls ins 4, 2021

    Ryan Budney and David Gabai. Knotted 3-balls ins 4, 2021

    work page 2021

  2. [2]

    On the automorphism groups of hyperbolic manifolds.Inter- national Mathematics Research Notices, 2025(7), April 2025

    Ryan Budney and David Gabai. On the automorphism groups of hyperbolic manifolds.Inter- national Mathematics Research Notices, 2025(7), April 2025

    work page 2025

  3. [3]

    Freedman and Frank Quinn.Topology of 4-Manifolds (PMS-39)

    Michael H. Freedman and Frank Quinn.Topology of 4-Manifolds (PMS-39). Princeton Uni- versity Press, 1990

    work page 1990

  4. [4]

    Diffeomorphisms of the 4-sphere, cerf theory and montesinos twins.Algebraic & Geometric Topology, 25(5):2817–2849, September 2025

    David T Gay. Diffeomorphisms of the 4-sphere, cerf theory and montesinos twins.Algebraic & Geometric Topology, 25(5):2817–2849, September 2025

    work page 2025

  5. [5]

    Concordance spaces, higher simple-homotopy theory, and applications

    Allen E Hatcher. Concordance spaces, higher simple-homotopy theory, and applications. In Algebraic and geometric topology (Proc. Sympos. Pure Math., Stanford Univ., Stanford, Calif., 1976), Part, volume 1, pages 3–21, 1978

    work page 1976

  6. [6]

    Hatcher and John B

    Allen E. Hatcher and John B. Wagoner.Pseudo-isotopies of compact manifolds. Number 6 in Ast´ erisque. Soci´ et´ e math´ ematique de France, 1973

    work page 1973

  7. [7]

    K. Igusa. What happens to hatcher and wagoner’s formula forπ 0c(m) when the first postnikov invariant ofmis nontrivial?Bak, A. (ed.) Algebraic K-Theory, Number Theory, Geometry and Analysis, Proceedings of Bielefeld. Series Lecture Notes in Mathematics, 1046, 1982

    work page 1982

  8. [8]

    Diffeomorphisms of 4-manifolds from graspers, 2025

    Danica Kosanovi´ c. Diffeomorphisms of 4-manifolds from graspers, 2025. 29

    work page 2025

  9. [9]

    Mapping class groups of 4-manifolds pisa lectures

    MARK POWELL. Mapping class groups of 4-manifolds pisa lectures

    work page

  10. [10]

    Pseudo-isotopies and diffeomorphisms of 4-manifolds, 2022

    Oliver Singh. Pseudo-isotopies and diffeomorphisms of 4-manifolds, 2022. QIUZHENCOLLEGE, TSINGHUAUNIVERSITY, BEIJING, CHINA Email address, Xiayu Tan:tan-xy22@mails.tsinghua.edu.cn 30

    work page 2022