Pith. sign in

REVIEW 3 major objections 5 minor 18 references

Higher path groupoids and the holonomy of formal power series connections

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read A formal power series connection on a smooth manifold determines higher holonomy functors from the path 2-groupoid and the path 3-groupoid into categories built from the completed cobar complex, and the Komendarczyk-Koytcheff-Volic…

desk verdict Genuinely useful extension of Kohno's higher holonomy program, but an off-by-one grading error in the connection condition makes Theorems 2 and 3 not type-check; fixable, and worth refereeing. read the letter →

arxiv 2506.23880 v1 pith:O3ESZMIC submitted 2025-06-30 math.AT math.CTmath.QA

classification math.ATmath.CTmath.QA MSC 18N1055P3555R8058A10
keywords higherholonomyfunctorsformalpowerseriesconnectionsiteratedintegralspath2-groupoid3-groupoidGray3-categorycobarcomplexconfigurationspaces
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 paper tries to show that the same formal power series transport that produces ordinary parallel transport along curves can be read one and two dimensions higher. For any formal power series connection $\omega$ on a smooth manifold, it constructs a holonomy 2-functor from the path 2-groupoid and a holonomy 3-functor from the path 3-groupoid into algebraic target categories built from the completed cobar complex of a differential graded coalgebra. The point of doing this is that Chen's iterated integrals, long used to compute loop-space homology, now become a source of higher categorical invariants of the manifold. As an application, the known Komendarczyk-Koytcheff-Volic connection on the configuration space of $m$ points in $\mathbb{R}^n$, for $n\ge 4$, is shown to produce a Gray functor out of the path 3-groupoid of that configuration space. If the proof is right, higher holonomy is not a special feature of particular connections but a general property of formal power series connections.

What carries the argument

The central object is the transport $T_\omega$ of a formal power series connection, which lives in the completed tensor product of Chen's bar complex of iterated integrals with the completed cobar complex $\widehat{\Omega}(C)$ and is defined as the sum of all iterated integrals of the connection's form components times the corresponding products of cobar components. The identity doing the work is the twisted cochain condition $d_\Omega\omega+d\omega=\varepsilon(\omega)\wedge\omega$, which, combined with Stokes' theorem, converts the boundary of a homotopy into the cobar differential of the integral over the homotopy. That conversion is what makes the holonomy values independent of representatives and compatible with the various compositions. The domain groupoids $P_2(M)$ and $P_3(M)$ are the path 2-groupoid and the fundamental Gray 3-groupoid, whose rank conditions on homotopies are chosen so that the integrated forms land in the degrees where the target categories are defined.

What would settle it

Find one laminated rank-2 homotopy $H$ in a manifold and forms $\omega_1,\dots,\omega_k$ of total degree $k+2$ such that the integral over $\Delta_k\times[0,1]^2$ of $\widetilde H^*(\omega_1\times\cdots\times\omega_k)$ is nonzero; Lemma 4.4 says this integral is always zero. A nonzero value would immediately show that $\langle T_\omega,h\rangle$ depends on the representative 2-path and that the holonomy 3-functor is not well defined.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem A: for a formal power series connection $\omega$ on a smooth manifold $M$ with values in the completed cobar complex $\widehat{\Omega}(C)$ of a coaugmented differential graded coalgebra $C$, the transport $T_\omega$ integrates over paths, 2-paths, and good 3-paths to define functors $\mathrm{Hol}^2_\omega\colon P_2(M)\to C_2(C)$ and $\mathrm{Hol}^3_\omega\colon P_3(M)\to C_3(C)$. The 2-functor lands in a quotient by $d_\Omega(\widehat{\Omega}^1(C)\otimes\widehat{\Omega}^1(C))$, while the 3-functor uses laminated rank-2 homotopies for its 2-morphism assignment and lands in a quotient by $d_\Omega\widehat{\Omega}^{1,2}(C)$ for its 3-morphisms. In the configuration-space application, the target simplifies because $\bar D(m)^*$ is 1-connected, so the 2-morphisms of $C_3(\bar D(m)^*)$ are linear combinations of degree-2 diagrams with additive compositions.

Load-bearing premise

The paper asserts without proof that a laminated homotopy is so thin that evaluating it at $k$ points never fills more than a $(k+1)$-dimensional region, which is what forces all relevant iterated integrals to vanish; if that assertion fails for even one homotopy, the holonomy 3-functor would depend on choices of representatives and would not exist as stated.

Editorial extensions

If this is right

  • The holonomy 2-functor assigns to every rank-2 homotopy class of 2-paths an element of $\widehat{\Omega}^1(C)/d_\Omega(\widehat{\Omega}^1(C)\otimes\widehat{\Omega}^1(C))$, with vertical composition given by addition and horizontal composition by mixed product formulas.
  • The holonomy 3-functor assigns to every rank-3 homotopy class of good 3-paths an element of $\widehat{\Omega}^2(C)/d_\Omega\widehat{\Omega}^{1,2}(C)$, and it is a Gray functor compatible with upward, vertical, and horizontal whiskered compositions.
  • For $n\ge 4$, the transport of the Komendarczyk-Koytcheff-Volic connection defines a Gray functor $\mathrm{Hol}^3_\omega\colon P_3(\mathrm{Conf}(m,\mathbb{R}^n))\to C_3(\bar D(m)^*)$, giving a diagrammatic higher holonomy for configuration spaces.
  • The ordinary holonomy functor on $P_1(M)$ is recovered as the 1-dimensional case, so the higher functors extend rather than replace classical parallel transport.
  • The paper conjectures that holonomy $k$-functors exist for all $k\ge 4$, conditional on a precise definition of the path $k$-groupoid.

Reading between the lines

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

  • If Lemma 4.4 is sound, the laminated rank-2 homotopy relation rather than the weaker rank-2 relation is the correct thin-homotopy notion for this transport, and one could test this by searching for natural rank-2 homotopic 2-paths that are not laminated and whose transports differ.
  • The same construction pattern suggests that any flat formal connection valued in a complete differential graded algebra, not only a cobar complex, would induce analogous higher holonomy functors with the target category read off from degrees 0, 1, and 2 of that algebra.
  • The configuration-space Gray functor gives an explicit diagrammatic invariant of 3-dimensional homotopies of point configurations, which is a natural place to look for invariants of braided surfaces in $\mathbb{R}^4$; the author's introduction gestures at this motivation without developing it.
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 / 5 minor

Summary. The paper develops a Chen–Kohno style framework in which the transport of a formal power series connection with values in the cobar complex of a coaugmented dg coalgebra C produces higher holonomy functors. The main results are Theorem 2, asserting a 2-functor Hol^2_ω : P2(M) → C2(C), and Theorem 3, asserting a Gray 3-functor Hol^3_ω : P3(M) → C3(C), where C2(C) and C3(C) are defined from the completed cobar complex. These are applied in Corollary 4 to the Komendarczyk–Koytcheff–Volić connection on Conf(m,R^n), n ≥ 4, yielding a Gray functor from the path 3-groupoid of that configuration space to C3(\bar D(m)^*). The paper is clearly organized and follows constructions of Chen, Kohno, and Faria Martins–Picken, but the main theorems as stated are affected by a systematic off-by-one degree issue, and several load-bearing verifications are only sketched.

Significance. If the construction is made correct, the paper would provide a uniform, explicit way to promote parallel transport to higher holonomy functors, and it would give a concrete Gray functor for configuration spaces of points in R^n for n ≥ 4, connecting diagram coalgebras to higher path groupoids. This is a natural and potentially useful step beyond Kohno's higher holonomy maps. The paper is honest about its debts to Chen and Kohno, and the intended application to the KKV connection is concrete. However, the central theorems currently rely on an incorrect degree convention and on unproved rank/dimension assertions, so the significance cannot be fully assessed until those gaps are repaired.

major comments (3)
  1. [§3.2, §4.4, §4.8] There is an off-by-one grading error in the definition of a formal power series connection. In §3.2, the condition is deg(ω_i)=deg(c_i) with c_i ∈ Ω(C). In the transport formula of §3.3, the coefficient c_{i1}⋯c_{ik} is a product in Ω(C), so its degree is Σ_j deg(c_{i_j}). A 2-path is a 1-dimensional plot of the pointed path space, so §2.6 selects terms with Σ_j deg(ω_{i_j}) − k = 1. Under the stated equality, the surviving coefficient has degree Σ_j deg(c_{i_j}) = Σ_j deg(ω_{i_j}) = k+1, not 1. Therefore Hol2([g]) does not lie in Ω^1(C)/dΩ(Ω^1⊗Ω^1), the prescribed 2-morphism set of C2(C), and Theorem 2 fails as stated. The analogous computation for a 3-path (a 2-dimensional plot) gives degree k+2 rather than 2, so Hol3 does not land in the 3-morphism set of C3(C), making Theorem 3 and Corollary 4 incorrect as written. The manuscript is internally inconsistent: Lemma 4.4 uses the condition deg(ω_1)+⋯+deg(ω_k)=k+2, which is the correct condition for integration over a 3-path only under the shifted convention deg(ω_i)=deg(c_i)+1; and §5.2's description of 2-morphisms as 'diagrams of degree 2' matches [Γ^*] of degree 1 under that shift. The definition in §3.2 should be corrected to deg(ω_i)=deg(c_i)+1 throughout, and the twisted cochain condition and all subsequent lemmas should be rechecked with this convention.
  2. [§4.5, Lemma 4.4; §4.6, Lemma 4.7] Lemma 4.4 is load-bearing for the well-definedness of Hol3 on laminated 2-tracks, but its key assertion is not proved. The lemma claims that for a laminated rank-2 homotopy H, the induced map \tilde H: Δ_k × [0,1]^2 → M^k has rank at most k+1, so that the pullback of any form of total degree k+2 vanishes. This is stated in a single sentence with no derivation or citation. The rank bound is not immediate from the rank-2 condition: the tangent space of \tilde H at a point is spanned by the r-, s-, and t_i-derivatives, and one must show that the lamination conditions impose enough relations to reduce the dimension from k+2 to k+1. Without a proof, the independence of the integral from the choice of laminated representative is unverified. The same issue appears in Lemma 4.7, where the assertion that ⟨Tω,W⟩ ∈ Ω^{1,2}(C) for a rank-3 homotopy W is made without a dimension-count argument. Please supply complete proofs or precise references.
  3. [§4.3, §4.7] The definitions of the target categories C2(C) and C3(C) are not fully verified, although the main theorems assert functors into them. For C3(C), the displayed structure does not include all operations of a Gray 3-category: the operation called 'vertical composition of 3-morphisms' in §4.7 is actually a horizontal composition along 2-morphisms, with source M·P and target N·Q, while the usual vertical composition of 3-morphisms with identical source and target 2-morphisms is only the 'upward composition'. No interchange laws or coherence data are stated, and both occurrences of 'straightforward verification' are left to the reader. Since Theorems 2 and 3 assert functors into these structures, the target categories must be fully defined and the coherence verified, or a precise reference that establishes this exact construction must be supplied.
minor comments (5)
  1. [§4.2, Lemma 4.2 proof] The last equality of the proof writes '⟨∫ω1...ωk, g⟩ + ⟨∫ω1...ωk, g⟩'; the second summand should be h.
  2. [§4.5 and §4.6, Lemmas 4.6 and 4.10] Several displayed formulas have missing commas or arguments, e.g. '⟨Tωh1(1,−)⟩' should be '⟨Tω,h1(1,−)⟩' and '⟨TωJ(0,1,−)⟩' should be '⟨Tω,J(0,1,−)⟩'.
  3. [§4.8, paragraph before Theorem 3] The text says 'the association Hol2ω is compatible' when it should say 'Hol3ω'.
  4. [§4.2, remark after Lemma 4.1] The remark would be more informative if it included an explicit example where rank-2 homotopy changes the iterated integral, since it is used to motivate the laminated condition.
  5. [§5.2] The phrase 'since \bar D(m)^* is 1-connected' should be defined or accompanied by a reference, as 1-connectedness for a dg coalgebra is not standard terminology in this paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: higher holonomy functors are derived forward from transport, target categories are fixed independently, and cited input is external.

full rationale

The derivation is not circular under any of the enumerated patterns. The transport Tω is defined in Section 3.3 directly from the formal power series connection as a completed iterated-integral series, and the functor values Hol2_ω([g]) = [⟨Tω,g⟩] and Hol3_ω([J]) = [⟨Tω,J⟩] are defined in Sections 4.4 and 4.8 as evaluations of this single forward construction. The target 2- and 3-categories C2(C) and C3(C) are built from the cobar complex of the fixed coalgebra C in Sections 4.3 and 4.7, before the functors are assembled, and their morphism conditions dΩM = m − n and dΩα = M − N are checked via Stokes' theorem and the twisted cochain condition, not imposed after the fact. The invariance lemmas 4.1, 4.4, and 4.7 use genuine geometric hypotheses (rank-2, laminated rank-2, and rank-3 homotopies) together with the cobar differential; they do not rename an input as a conclusion. There are no fitted parameters, no data subsets, and no predictions forced by a fit. All citations to Chen and Kohno are to external prior work on iterated integrals and rank-1 homotopy invariance, not to the present author, so the self-citation patterns do not apply. The application to configuration spaces (Corollary 4) feeds an existing KKV connection into the same machine and verifies functoriality; it does not reduce to the definition of the connection. The unsupported rank bound in Lemma 4.4 is an evidentiary gap rather than a circular step, since no equation identifies the conclusion with the hypothesis. A separate grading mismatch noted in Sections 3.2 and 5.2 appears to place transport values in the wrong degree for the claimed target categories; that is a correctness concern that would break the statements as written, but it is a degree-bookkeeping error, not a circular derivation. Therefore the circularity score is 0.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No free parameters are fitted to data; the formal power series connection is an input and the KKV connection is a fixed construction. The paper introduces new target categories C2(C) and C3(C), but these are explicit definitions rather than postulated entities with an independent empirical handle. The axioms listed are the unproved or externally cited results on which the central functor theorems rest, with the rank bound in Lemma 4.4 being the most fragile.

assumptions (7)
  • domain assumption The path 3-groupoid P3(M), as defined in Section 1.15, is a Gray 3-groupoid; proof deferred to [FMP11].
    The holonomy 3-functor in Theorem 3 has P3(M) as its domain; the paper does not prove the Gray coherence and relies on the cited reference.
  • standard math Lemma 2.1: differential formula for iterated integrals on pointed path spaces, attributed to [Koh16].
    Used throughout Section 4 to relate dT_omega to the twisted cochain condition; no proof is given in this paper.
  • standard math Rank-1 homotopy invariance of iterated integrals, attributed to Kohno in Section 4.1.
    Needed for the well-definedness of Hol1, Hol2 and Hol3 on 1-morphisms.
  • standard math Chen's product and composition formulas [Che73, Proposition 2.1.1 and 2.1.2], used in Lemmas 4.3, 4.5, 4.10.
    These supply the multiplicative behavior of transport under concatenation and horizontal composition.
  • ad hoc to paper For a laminated rank-2 homotopy H, the map tilde H from Delta_k times [0,1]^2 to M^k has rank at most k+1 for forms of total degree k+2.
    Asserted in the proof of Lemma 4.4 without derivation or citation; it is exactly what makes integration over laminated 2-tracks well defined.
  • ad hoc to paper C2(C) is a strict 2-category and C3(C) is a Gray 3-category; the paper says this is a straightforward verification in Sections 4.3 and 4.7.
    The target categories' coherence is asserted, not demonstrated, and the functor theorems depend on these structures.
  • domain assumption For n at least 4, each grading of Kontsevich's diagram algebra bar D(m) is finite dimensional and the dual is a 1-connected dg coalgebra, taken from [LV14, KKV24].
    Used in Section 5.1 to ensure the KKV formal connection is well defined and the target Gray 3-category simplifies.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Higher path groupoids and the holonomy of formal power series connections." pith.science (2026). https://pith.science/paper/O3ESZMIC

@misc{pith2026250623880,
  author       = {Pith},
  title        = {Pith review of: Higher path groupoids and the holonomy of formal power series connections},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O3ESZMIC}},
  note         = {Machine review of arXiv:2506.23880}
}
abstract

Building on ideas of Kohno, we develop a framework for the construction of higher holonomy functors via the transport of formal power series connections. Using these techniques, we obtain functors from the path groupoid, the path 2-groupoid, and the path 3-groupoid of a manifold. As an application, we construct a Gray functor from the path 3-groupoid of the configuration space of $m$ points in $\mathbb{R}^n$ for $n\geqslant 4$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 17 canonical work pages

  1. [1]

    Higher gauge theory: 2-connections on 2-bundles

    John Baez and Urs Schreiber. Higher gauge theory: 2-connections on 2-bundles. arXiv preprint hep-th/0412325 , 2004

  2. [2]

    Cohen and Samuel Gitler

    Frederick R. Cohen and Samuel Gitler. On loop spaces of configuration spaces. Transactions of the American Mathematical Society , 354(5):1705--1748, 2002

  3. [3]

    Iterated integrals of differential forms and loop space homology

    Kuo-Tsai Chen. Iterated integrals of differential forms and loop space homology. Annals of Mathematics , 97(2):217--246, 1973

  4. [4]

    Iterated path integrals

    Kuo-Tsai Chen. Iterated path integrals. Bulletin of the American Mathematical Society , 83(5):831--879, 1977

  5. [5]

    Knotted surfaces and their diagrams , volume 55

    J Scott Carter and Masahico Saito. Knotted surfaces and their diagrams , volume 55. 1998

  6. [6]

    A journey through loop braid groups

    Celeste Damiani. A journey through loop braid groups. Expositiones Mathematicae , 35(3):252--285, September 2017

  7. [7]

    On two-dimensional holonomy

    Jo \ a o Faria Martins and Roger Picken. On two-dimensional holonomy. Transactions of the American Mathematical Society , 362(11):5657--5695, 2010

  8. [8]

    The fundamental Gray 3-groupoid of a smooth manifold and local 3-dimensional holonomy based on a 2-crossed module

    João Faria Martins and Roger Picken. The fundamental Gray 3-groupoid of a smooth manifold and local 3-dimensional holonomy based on a 2-crossed module. Differential Geometry and its Applications , 29(2):179--206, March 2011

Show all 18 references
  1. [9]

    Braid and Knot Theory in Dimension Four

    Seiichi Kamada. Braid and Knot Theory in Dimension Four . volume 95 of Mathematical Surveys and Monographs , Providence, Rhode Island, May 2002. American Mathematical Society

  2. [10]

    Diagrams for primitive cycles in spaces of pure braids and string links

    Rafal Komendarczyk, Robin Koytcheff, and Ismar Volić. Diagrams for primitive cycles in spaces of pure braids and string links. Annales de l'Institut Fourier , 74(4):1745--1807, 2024

  3. [11]

    Higher holonomy of formal homology connections and braid cobordisms

    Toshitake Kohno. Higher holonomy of formal homology connections and braid cobordisms. Journal of Knot Theory and Its Ramifications , 25(12):1642007, 2016

  4. [12]

    Higher holonomy maps for hyperplane arrangements

    Toshitake Kohno. Higher holonomy maps for hyperplane arrangements. European Journal of Mathematics , 6(3):905--927, 2020

  5. [13]

    Higher holonomy and iterated integrals

    Toshitake Kohno. Higher holonomy and iterated integrals. In A. Papadopoulos, editor, Topology and Geometry, A collection of papers dedicated to Vladimir G. Turaev , pages 307--325. European Mathematical Society Press, 2021

  6. [14]

    Formal connections, higher holonomy functors and iterated integrals

    Toshitake Kohno. Formal connections, higher holonomy functors and iterated integrals. Topology and its Applications , 313:107985, 2022

  7. [15]

    Formality of the little n -disks operad

    Pascal Lambrechts and Ismar Voli \'c . Formality of the little n -disks operad. Memoirs of the American Mathematical Society , 230(1079), 2014

  8. [16]

    Quantum invariants: A study of knots, 3-manifolds, and their sets , volume 29

    Tomotada Ohtsuki. Quantum invariants: A study of knots, 3-manifolds, and their sets , volume 29. World Scientific, 2002

  9. [17]

    Parallel transport and functors

    Urs Schreiber and Konrad Waldorf. Parallel transport and functors. J. Homotopy Relat. Struct. , 4, 05 2007

  10. [18]

    Smooth functors vs

    Urs Schreiber and Konrad Waldorf. Smooth functors vs. differential forms. Homology, Homotopy Appl. , 13, 02 2008

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.