Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Cabling in terms of immersed curves

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Cabling a knot is one explicit shear of its immersed curves

desk verdict Clean geometric formula for cabling in immersed-curve Floer theory; the main theorem is credible and useful, but the local-system extension is asserted rather than proved. read the letter →

arxiv 1908.04397 v1 pith:TKYNRNSF submitted 2019-08-12 math.GT

Jonathan Hanselman , Liam Watson This is my paper · ORCID
classification math.GT MSC 57K1857K10
keywords knotFloerhomologyimmersedcurvescablingborderedHeegaardconcordanceinvariantsL-spacesurgeriesmergeoperationplaneshear
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This note establishes a formula for how the Heegaard Floer invariant of a knot complement—encoded as a collection of immersed curves in a punctured torus—changes when the knot is replaced by its (p,q)-cable. The formula says that the cable's immersed multicurve is obtained from the original by drawing p copies, shearing the plane along lines of slope q/p so that lattice points slide onto a common vertical line, and rescaling. This makes cabling a purely pictorial operation rather than a fresh three-manifold gluing computation. The authors use the formula to reprove known results on the tau and epsilon concordance invariants under cabling and to construct a new infinite family of linearly independent topologically slice knots.

What carries the argument

The load-bearing object is the merge operation from the authors' earlier loop calculus, reinterpreted as a fractional plane shear. When one input to the merge is a curve made only of c_k segments, merging acts by shearing each vertical line of lattice points so that the horizontal axis is carried to the piecewise-linear curve that tracks the second input just below its lattice points. Cabling is the special case where that second input is a straight line of rational slope q/p; the shear then slides points along lines of slope q/p, and the bookkeeping of the cable framing turns this into the explicit plane map f_{p,q} appearing in the theorem.

What would settle it

Compute, from the bordered Floer complexes, the immersed multicurve for the (2,1)-cable of a knot whose complement's invariant includes a component with a nontrivial local system, and compare the result with the image of the original multicurve under f_{2,1}; a mismatch at the decorated component would refute the theorem as stated for arbitrary immersed multicurves.

Watch

Extended reading notes

Core claim

The paper proves that cabling a knot acts on its knot Floer data by an explicit homeomorphism of the punctured plane. Knot complements carry an invariant called an immersed multicurve: a finite set of immersed curves, possibly decorated with vector spaces, drawn in a punctured torus. The central theorem states that the multicurve for the (p,q)-cable is obtained from the multicurve for K by sliding each lattice point leftward along a line of slope q/p until it lands on a vertical line x = np, then compressing horizontally and stretching vertically by p, with a fixed vertical shift recorded. In drawing terms: lay p staggered copies of the original curve side by side, connect the loose ends, then slide the lattice pegs horizontally so they all line up vertically. The same map can be read as a periodic tiling of the plane by new tiles; the image of the original lattice under the map is the lattice of that tiling, and the cable's invariant is the image of the original curve under the tiling transformation.

Load-bearing premise

The whole result rests on an earlier gluing construction being correct for curves with extra decorations; the paper asserts rather than proves that extension, so if that gluing construction fails for decorated curves, the cabling formula fails.

Editorial extensions

If this is right

  • The (p,q)-cable invariant can be computed by drawing p copies of the original curve, staggering them vertically by q, joining ends, and compressing horizontally; no new bordered Floer computation is needed.
  • The formulas of Hom for tau and epsilon under cabling are recovered directly from the first intersection of the first copy of the distinguished curve with the vertical axis.
  • For (2,1)-cables, the refined Phi_i counts obey parity rules: Phi_{2n} = Phi_n^{++} + Phi_n^{--} and Phi_{2n+1} = Phi_n^{+-} + Phi_{n+1}^{-+} for n at least 1.
  • Iterated (2,1)-cabling doubles the length of a unique maximal ++ left arc, yielding an infinite family of topologically slice knots that are linearly independent in the smooth concordance group.
  • If a knot's curve set contains a closed component enclosing two adjacent lattice points of different height modulo p, that knot cannot be a (p,q)-cable of another knot.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same plane-shear picture should describe how satellite operations act on immersed curves in general: a satellite operation is likely realized by a multivalued Lagrangian correspondence, and the cabling theorem is the first instance where that action is visibly a fractional plane shear.
  • The immersed-curve encoding loses only diagonal arrows from the full chain complex, so a refined curve-with-decorations invariant would be needed to lift the cabling formula to complete CFK^- complexes rather than just their curve shadows.
  • The asserted extension of the merge operation to nontrivial local systems could be tested explicitly on a small example with a nontrivial local system; this is the one step of the proof that is stated as a straightforward computation rather than written out.
  • The tiling reformulation may make the cabling action available to other invariants that admit curve or graph models in the punctured torus, provided those models satisfy the same merge rule.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes a concise geometric formula for the behavior of the immersed-curve invariant of knot Floer homology under cabling. Concretely, for a knot K in S^3 with associated immersed multicurve γ in the punctured torus, the paper claims that the immersed curve of the (p,q)-cable complement γ_{p,q} is obtained, after lifting to the universal cover, by applying an explicit plane map f_{p,q} to γ. The argument is based on a merge operation imported from the authors' earlier loop-calculus work with J. Rasmussen, translated into the immersed-curve language via a graphical toroidal-grid calculus. The paper also gives quick re-derivations of Hom's theorems on τ and ε under cabling, a re-proof of the L-space surgery criterion for cables, and a concordance application producing a Z^∞ summand from iterated (2,1)-cables of a knot with trivial Alexander polynomial and prescribed γ_0.

Significance. If the main theorem is fully established, the paper's formula is valuable: it turns cabling into a transparent geometric operation on immersed curves and immediately explains several known numerical cabling formulas. Its strengths are that the geometric picture is clear, the re-proofs of Hom's theorems are genuine consequences rather than assumed inputs, and the consistency checks against known results are convincing. The concordance application is also interesting, since it gives an independent family of topologically slice, smoothly independent knots from simple curve data. However, the paper currently depends in an essential place on an unproved local-system extension of the merge operation, so the full generality of the main theorem is not yet established.

major comments (3)
  1. [§1.2, Proposition 12] Proposition 12 is load-bearing for Theorem 1, but its proof is not supplied: the text says only 'This is a straightforward computation' after referring to [6, Figures 10–12]. Since [6] does not handle nontrivial local systems, Proposition 12 is a new claim rather than an imported theorem. The concern is not cosmetic: in the local-system expansion, the extra arrows representing Φ interact with the c_k columns in the toroidal grid, and a basepoint, ordering, or contraction issue could change the holonomy on the image component. Without a written verification, Theorem 1 is unproved for any K whose HF-hat(M) has a nontrivial local system on a closed component. Please include the actual computation, or explicitly restrict the statement of Theorem 1 (and of Corollary 13) to trivial local systems.
  2. [§1.2, Corollary 13] The proof of Corollary 13 is descriptive rather than formal: it says 'This is the main thrust of Figure 13' and then asserts that the toroidal grid produces p copies of the word for ϑ with shifted c-indices, distinguishing only between components homologous to λ and nullhomologous components. Since Corollary 13 is used directly in the proof of Theorem 1 in §2.2, the argument should be written out as a verification that the grid assembly respects endpoints, periodicity, and the stated period pq. In particular, the claim that relative primality of p and q forces the new curve to make p vertical passes before closing deserves a precise justification.
  3. [§2.2, proof of Theorem 1] The final step of the proof of Theorem 1 is a sequence of geometric shears described in words and pictures, but the map f_{p,q} is not given by an explicit formula or by a precise composition of piecewise-linear maps on the plane. In particular, the vertical shift (p-1)(q-1)/2 and the claim that f_{p,q} sends Z^2 to Z^2 are stated without calculation. This is not fatal, but a precise definition of f_{p,q} would make the theorem checkable and would also clarify the meaning of 'homotopic' for the lifted curves.
minor comments (4)
  1. [Abstract and §2] There are several typos: 'corresonding' in Theorem 1, 'the the' in the caption of Figure 7, and 'effects' in §2.1 should read 'affects'.
  2. [§1.1] The definitions of letters a_k, b_k, c_k, d_k, and e are given informally through Figure 10 and the surrounding text; a short formal definition or an explicit reference to the corresponding notation in [6, Figure 1] would improve precision.
  3. [§1.1, local systems] The phrase 'a letter decorated by the trivial local system of dimension n corresponds to n parallel copies of the relevant curve segment' is clear, but the notation for the local system (V, Φ) on a cyclic word would benefit from an explicit convention about the order of composition of the endomorphisms.
  4. [Figure 16] The captions in Figures 15 and 16 are long and partly repeat the text; labeling the key curves in the figures themselves would make the two-step shear argument easier to follow.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem 1 is a genuine geometric translation of the authors' earlier merge theorem, with the main caveat being an unproved but non-circular local-system extension in Proposition 12.

full rationale

The claimed derivation is not circular. Theorem 1 is not assumed in the construction of f_{p,q}: the map is defined from the framings in the gluing of M, P×S1, and D2×S1 (Section 2.1), and the curve formula is derived by first invoking the authors' earlier merge theorem for bordered Floer homology (Proposition 11, from [6]), then translating the letter-level merge rules into a plane shear (Corollary 13 and Section 2.2). The result is checked against external theorems of Hom and Hedden (Theorems 3, 4, 9), so the formula has independent content rather than being a fitted restatement. The main weakness is Proposition 12: the extension of the merge to non-trivial local systems is asserted with 'This is a straightforward computation' and a reference to [6, Figures 10–12], even though the paper itself states that [6] does not handle local systems. That is a load-bearing omitted proof for the full generality of Theorem 1, and Corollary 13's proof is likewise descriptive; but neither step assumes the target formula or renames a fitted input, so these are correctness/completeness gaps, not circularity.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper is pure mathematics with no fitted parameters. Its assumptions are imported theorems from the authors' earlier program and from standard low-dimensional topology.

assumptions (5)
  • domain assumption Bordered Floer invariants for manifolds with torus boundary are equivalent to immersed multicurves in the punctured torus.
    This is the foundational theorem of [4] and [5]; Theorem 1 is stated entirely in this language.
  • domain assumption The merge operation m(gamma,theta) computes the bordered invariant of the manifold glued along essential annuli, with the letter-by-letter rules of Proposition 11.
    Proposition 11 and Corollary 13 are imported from [6]; the proof of the cabling theorem depends on them.
  • domain assumption Type D structures of knot complements can be represented by cyclic words in the alphabet {a_k, b_k, c_k, d_k, e}, with nontrivial local systems concentrated on an a_k segment.
    Used in Section 1 to translate between train tracks and immersed curves; from [4,5,6].
  • standard math Knots with trivial Alexander polynomial are topologically slice (Freedman).
    Used in Corollary 7 to ensure the knots K_n are topologically slice.
  • domain assumption The invariants phi_i are concordance homomorphisms.
    Imported from [1]; used in Corollary 7 to conclude linear independence in the smooth concordance group.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cabling in terms of immersed curves." pith.science (2026). https://pith.science/paper/TKYNRNSF

@misc{pith2026190804397,
  author       = {Pith},
  title        = {Pith review of: Cabling in terms of immersed curves},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TKYNRNSF}},
  note         = {Machine review of arXiv:1908.04397}
}
read the original abstract

In joint work with J. Rasmussen, we gave an interpretation of Heegaard Floer homology for manifolds with torus boundary in terms of immersed curves in a punctured torus. In particular, knot Floer homology is captured by this invariant. Appealing to earlier work of the authors on bordered Floer homology, we give a formula for the behaviour of these immersed curves under cabling.

Figures

Figures reproduced from arXiv: 1908.04397 by the authors.

Figure 1
Figure 1. The Heegaard Floer homology for the (2, −1) and (2, 1) cables of the right-hand trefoil; the invariant for the tre￾foil complement is shown in grey. Bordered Floer homology provides an essential tool for study￾ing decompositions of three-manifolds along essential tori [15]. A framework of bimodules, of relevance to satellite opera￾tions, is laid out in [14]. The work of Levine [13], Hom [11], and Petkova [19], for e… view at source ↗
Figure 2
Figure 2. Computation of the immersed curve associated with the (3, 2)-cable of the right handed trefoil, starting from the trefoil curve pictured on the left. The two middle diagrams are two ways of thinking about the construction starting from three copies of the trefoil curve: we either slide lattice points along lines of slope 2 3 or we stagger the heights of the three copies of the trefoil curve and then slide lattice po… view at source ↗
Figure 3
Figure 3. Calculating τ(Kp,q). A quick re-proof of Theorem 4. The first intersection of γ 0 0 with the vertical axis clearly comes from the first inter￾section of the first copy of γ0 with the vertical axis. This intersection occurs between the lattice points at height τ (K) and τ (K) + 1; after applying f p,q and the appro￾priate vertical shift, these lattice points map to heights h1 = pτ (K)+ (p−1)(q−1) 2 and h2 = pτ (K)+p+… view at source ↗
Figures from the paper (13 more)
Figure 4
Figure 4. Figure 4: Four cases complete the proof. The key observation is that every component of γR lies to the right of every even height lattice point, and therefore any left arc on γ 0 0 which lies in γR must have length one. It is also clear that any left arc of γ 0 0 intersects at m…
Figure 5
Figure 5. Figure 5: Immersed curves for the first few iterated (2, 1)-cables of the right handed trefoil. These are also the distinguished curve γ0 for the knots K0, K1, K2, and K3 from Corollary 7. The longest left arc (highlighted) is stretched by a factor of two with each cabling itera…
Figure 7
Figure 7. Figure 7: The the (3,4)-cable (left) and the (3,2)-cable (right) of the right hand trefoil. Remark 8. We leave the behavior of φ1 under (2, 1) cabling, which was not needed in the above application, as an exercise to the motivated reader, who will find that φ1(K2,1) = − X j≥1 φj…
Figure 6
Figure 6. Figure 6: Because the first left arc has length 2, if this curve comes from a ( [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: The (3, 2)-cabling operation interpreted as a plane tiling: 3 copies of the standard square tile (above) are carried to a new regular tile in T3,2 (below) under the operation fp,q appearing in Theorem 1. To illustrate, the image of the longitude has been included (grad…
Figure 9
Figure 9. Figure 9: The torus algebra A as the path algebra of a quiver with relations. The torus algebra A is obtained as the path algebra of the quiver described in [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Segments of immersed curve in the cover of the marked torus, labelled to be consistent with the puzzle pieces given in [6]. The integer subscript k > 0 indicates the number of ◦ generators in the segment. These letters can appear forwards or backwards in a cyclic word…
Figure 11
Figure 11. Figure 11: Three different views of the invariant associated with the exterior of the right-hand trefoil. In all three cases, we have fixed the preferred (µ, λ) framing in order to present the torus boundary. On the lower left, the decorated graph describing the type D structure…
Figure 12
Figure 12. Figure 12: A 3-dimensional local system, expanded at an a2 to give a train track. It is relatively straightforward to extend this language to cases admitting a non-trivial local system. Recall that each letter in A corresponds to a (portion of a) type D structure that is a linea…
Figure 13
Figure 13. Figure 13: Merging a pair of curves, as in Proposition 11, described graphically: On the left hand side of the diagram, the output curve is interpreted on a toroidal grid, where the ck from γ (written on the horizontal) act on the letters in ϑ (written on the vertical). On the r…
Figure 14
Figure 14. Figure 14: The PL key curves approximating γ. Note that as the resulting loop is traversed, horizontal motion in the grid corresponds exactly to horizontal motion of the corresponding curve in the plane. In particular, each vertical line of lattice points in the plane correspond…
Figure 15
Figure 15. Figure 15: The fractional plane shear in the vertical direction associated with computing a (p, q)-cable, viewed with respect to the (−f, b) framing. Other relevant curves are shown, with respect to this framing, in the top left. The bottom left shows a copy of m through the ori…
Figure 16
Figure 16. Figure 16: Starting with the curve HFd(M) drawn in the plane with respect to the standard (µ, λ) framing, the fractional plane shear in the f direction which takes dµe to b produces the curve HFd(Mp,q), though not in terms of a convenient parametrization. shearing back partially…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Fractional Dehn twist coefficients and rank bounds for categorified link invariants

    math.GT 2026-08 conditional novelty 7.0 of 10

    Fibered links whose monodromy twists many times around a boundary component have large next-to-top link Floer homology, and braid closures with large fractional Dehn twist coefficient have large annular Khovanov homology.

Reference graph

Works this paper leans on

21 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [6]

    A calculus for bordered Floer homology

    Jonathan Hanselman and Liam Watson. A calculus for bordered Floer homology. Preprint, arXiv.1508.05445

    work page Pith review arXiv

  2. [1]

    More concordance homomorphisms from knot Floer homology

    Irving Dai, Jennifer Hom, Matthew Stoffregen, and Linh Truong. More concordance homomorphisms from knot Floer homology. Preprint, arXiv:1902.03333

    arXiv 1902

  3. [2]

    The topology of four-dimensional manifolds

    Michael Hartley Freedman. The topology of four-dimensional manifolds. J. Differential Geom. , 17(3):357–453, 1982

    work page 1982

  4. [3]

    Bordered Heegaard Floer homology and graph manifolds

    Jonathan Hanselman. Bordered Heegaard Floer homology and graph manifolds. Algebr. Geom. Topol. , 16(6):3103–3166, 2016

    work page 2016

  5. [4]

    Bordered Floer homology for manifolds with torus boundary via immersed curves

    Jonathan Hanselman, Jacob Rasmussen, and Liam Watson. Bordered Floer homology for manifolds with torus boundary via immersed curves. Preprint, arXiv:1604.03466

    arXiv

  6. [5]

    Heegaard Floer homology for manifolds with torus boundary: properties and examples

    Jonathan Hanselman, Jacob Rasmussen, and Liam Watson. Heegaard Floer homology for manifolds with torus boundary: properties and examples. Preprint, arXiv.1810.10355

    arXiv

  7. [7]

    On knot Floer homology and cabling

    Matthew Hedden. On knot Floer homology and cabling . ProQuest LLC, Ann Arbor, MI, 2005. Thesis (Ph.D.)– Columbia University. 18 JONATHAN HANSELMAN AND LIAM W ATSON

    work page 2005

  8. [8]

    On knot Floer homology and cabling

    Matthew Hedden. On knot Floer homology and cabling. Algebr. Geom. Topol., 5:1197–1222, 2005

    work page 2005

Show all 21 references
  1. [9]

    On knot Floer homology and cabling

    Matthew Hedden. On knot Floer homology and cabling. II. Int. Math. Res. Not. IMRN , (12):2248–2274, 2009

    2009

  2. [10]

    A note on cabling and L-space surgeries

    Jennifer Hom. A note on cabling and L-space surgeries. Algebr. Geom. Topol., 11(1):219–223, 2011

    2011

  3. [11]

    Bordered Heegaard Floer homology and the tau-invariant of cable knots

    Jennifer Hom. Bordered Heegaard Floer homology and the tau-invariant of cable knots. J. Topol., 7(2):287–326, 2014

    2014

  4. [12]

    A survey on Heegaard Floer homology and concordance

    Jennifer Hom. A survey on Heegaard Floer homology and concordance. J. Knot Theory Ramifications , 26(2):1740015, 24, 2017

    2017

  5. [13]

    Knot doubling operators and bordered Heegaard Floer homology

    Adam Simon Levine. Knot doubling operators and bordered Heegaard Floer homology. J. Topol., 5(3):651–712, 2012

    2012

  6. [14]

    Ozsv´ ath, and Dylan P

    Robert Lipshitz, Peter S. Ozsv´ ath, and Dylan P. Thurston. Bimodules in bordered Heegaard Floer homology. Geom. Topol., 19(2):525–724, 2015

    2015

  7. [15]

    Ozsvath, and Dylan P

    Robert Lipshitz, Peter S. Ozsvath, and Dylan P. Thurston. Bordered Heegaard Floer homology. Mem. Amer. Math. Soc., 254(1216):viii+279, 2018

    2018

  8. [16]

    What is a train track? Notices of the AMS , 50(3):354–355, 2003

    Lee Mosher. What is a train track? Notices of the AMS , 50(3):354–355, 2003

    2003

  9. [17]

    Holomorphic disks and knot invariants.Adv

    Peter Ozsv´ ath and Zolt´ an Szab´ o. Holomorphic disks and knot invariants.Adv. Math., 186(1):58–116, 2004

    2004

  10. [18]

    Ozsv´ ath, Andr´ as I

    Peter S. Ozsv´ ath, Andr´ as I. Stipsicz, and Zolt´ an Szab´ o. Concordance homomorphisms from knot Floer homology. Adv. Math., 315:366–426, 2017

    2017

  11. [19]

    Cables of thin knots and bordered Heegaard Floer homology

    Ina Petkova. Cables of thin knots and bordered Heegaard Floer homology. Quantum Topol., 4(4):377–409, 2013

    2013

  12. [20]

    Floer homology and knot complements

    Jacob Andrew Rasmussen. Floer homology and knot complements . ProQuest LLC, Ann Arbor, MI, 2003. Thesis (Ph.D.)–Harvard University

    2003

  13. [21]

    Van Cott

    Cornelia A. Van Cott. Ozsv´ ath-Szab´ o and Rasmussen invariants of cable knots.Algebr. Geom. Topol., 10(2):825– 836, 2010. Department of Mathematics, Princeton University. E-mail address: jh66@math.princeton.edu Department of Mathematics, University of British Columbia. E-mai...

    2010

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.