Pith. sign in

REVIEW 3 major objections 4 minor 5 references

On injectivity of Cohen-Wu homomorphism

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read This paper proves that the Cohen-Wu homomorphism from Milnor's $F[S^1]$ to the pure braid simplicial group $\mathrm{AP}$ is injective, using only group theory and HNN extensions.

desk verdict A clever, group-theoretic new proof of a known theorem; the induction has a missing base case and a suspicious Hopfian step, but the core reduction is sound and worth refereeing. read the letter →

arxiv 2507.09636 v1 pith:USJ7R2HB submitted 2025-07-13 math.GR

classification math.GR MSC 20F3620E06
keywords Cohen-WuhomomorphismpurebraidgroupsMilnorfreegroupconstructionsubgroupsHNNextensionshomotopyofthe2-sphereinjectivity
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 proves that the Cohen-Wu homomorphism — the natural map from Milnor's free group construction on the circle, $F[S^1]$, to the simplicial group $\mathrm{AP}$ of pure braids — is injective. The original proof of Cohen and Wu descended to graded Lie algebras and used the infinitesimal braid relations; this proof stays inside group theory. The whole injectivity is reduced to showing that the braids $\tilde A_{0,2},\dots,\tilde A_{0,n+1}$ in $P_{n+1}$ freely generate a subgroup of rank $n$. That freeness is proved by rewriting the ambient subgroup as an HNN extension and applying the normal-form theorem for such extensions inductively. The theorem matters because it embeds homotopy information of $S^2$ into braid groups, so elements of $\pi_n(S^2)$ can be represented by braids.

What carries the argument

The central machinery is the HNN-extension ladder that computes the freeness of $H_n=\langle y_1x_1y_1,\dots,y_nx_ny_n\rangle$ inside $G_n$. The subgroup $G_n=\langle x_1,\dots,x_n,y_1,\dots,y_{n-1}\rangle\subset P_{n+1}$ is presented as a semidirect product $F_n\rtimes\mathbb{Z}^{n-1}$ with conjugation rules $x_i^{y_j}=x_i^{x_j^{-1}}$ for $j<i$ and $x_i^{y_i}=x_i^{p_i^{-1}}$; then $G_n$ is obtained from $K_n$ by an HNN extension with associated subgroup $A_n=\langle y_1x_1,\dots,y_{n-2}x_{n-2},y_{n-1}p_{n-1}\rangle$. A corollary of the normal-form theorem for HNN extensions (Lemma 6) says that a subgroup $L$ of the base with $L\cap A=L\cap B=1$ freely generates together with the stable letter; Theorem 2 feeds this criterion through an induction that proves $H_n$ free of rank $n$.

What would settle it

For $n=3$ (or $n=4$), express $x_i=A_{i,n+1}$, $y_j=A_{j,j+1}\cdots A_{j,n}$, and $p_i=x_i\cdots x_n$ as explicit words in the Artin generators $A_{a,b}$ of $P_{n+1}$, and verify the displayed identities in Lemma 3: $x_i^{y_j}=x_i^{x_j^{-1}}$ for $j<i$ and $x_i^{y_i}=x_i^{p_i^{-1}}$. One mismatch would break the semidirect-product presentation and the subsequent HNN argument; a complete verification would close the only pictorial gap.

Watch

Extended reading notes

Core claim

The paper's central claim is that Theorem 1 holds and has a purely group-theoretic proof. Concretely, for every $n$, the image of the rank-$n$ free group $F[S^1]_n$ in $P_{n+1}$ is generated by the braids $\tilde A_{0,2},\dots,\tilde A_{0,n+1}$; after applying the automorphism $\mu_{n+1}$, these become $y_kx_ky_k$ for $1\le k\le n$, where $x_i=A_{i,n+1}$ and $y_i=A_{i,i+1}\cdots A_{i,n}$. The proof shows directly that the subgroup these elements generate is free of rank $n$: it presents the ambient group $G_n$ as a semidirect product $F_n\rtimes\mathbb{Z}^{n-1}$, builds $G_n$ as an HNN extension, and uses the normal-form theorem to prove freeness by induction. This establishes the injectivity of $\Theta$ without Lie algebras.

Load-bearing premise

The proof rests on Lemma 3's claim, supported only by pictures, that the conjugation action in $G_n$ satisfies $x_i^{y_j}=x_i^{x_j^{-1}}$ for $j<i$ and $x_i^{y_i}=x_i^{p_i^{-1}}$; if those braid equalities are wrong, the HNN presentation and the induction collapse.

Editorial extensions

If this is right

  • The quotient simplicial set $\mathrm{AP}/F[S^1]$ has the homotopy type of $S^2$, since $\Theta$ is injective.
  • Homotopy groups $\pi_n(S^2)$ appear as sub-quotients of pure braid groups, with braids representing homotopy classes.
  • The free basis $y_1x_1y_1,\dots,y_nx_ny_n$ (after the automorphism) gives an explicit description of the image subgroup $\mathrm{Im}(\Theta_n)$.
  • The proof avoids the graded-Lie-algebra and infinitesimal-braid-relation machinery, so the embedding can be established by group-theoretic rewriting.

Reading between the lines

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

  • A fully algebraic verification of the conjugation equalities in Lemma 3, written in the standard Artin generators of pure braid groups, would make the entire proof checkable by a proof assistant and would remove the only pictorial step.
  • The same HNN-extension criterion for freeness could be tried on other candidate braid subgroups, such as spherically trivial braids or kernels of other string-deletion maps, where free subgroups often admit explicit bases.
  • Because the proof only uses the semidirect-product structure of pure braid groups, a similar argument might establish injectivity for Milnor's $F[S^k]$ into the corresponding spherical braid simplicial groups, if an analogous untwisting automorphism exists.
Share X Bluesky LinkedIn Reddit HN

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. This paper gives a new proof of the Cohen–Wu theorem that the simplicial group homomorphism Θ : F[S^1] → AP from Milnor's free group construction to the pure braid simplicial group is injective. The proof is purely group-theoretic: after identifying Im(Θ_n) with a subgroup generated by transformed braids in P_{n+1}, the author defines a subgroup G_n and gives a presentation for it as an HNN extension; the desired freeness of H_n is then proved by induction using Britton normal form and deletion maps. The theorem itself was previously proved by Cohen and Wu via Lie algebra methods, so the contribution is a new elementary proof rather than a new result.

Significance. If completed, the paper would provide a considerably more accessible proof of an important embedding theorem, replacing Lie algebra computations with braid manipulation, semidirect products, and HNN extensions. The reduction from the simplicial statement to Problem 2 is explicit, and the HNN-extension strategy is well chosen; the paper also makes good use of standard facts such as Hopfianity of free groups and Britton normal form. However, several load-bearing steps are only sketched or left to pictures, and the induction is not fully specified; these issues are patchable but must be addressed before the proof can be accepted as rigorous.

major comments (3)
  1. [Section 3, Lemma 3] The semidirect product decomposition G_n = <x_1,...,x_n> ⋊ <y_1,...,y_{n-1}> and the three conjugation formulas x_i^{y_j} = x_i^{x_j^{-1}} (j<i), x_i^{y_i} = x_i^{p_i^{-1}}, x_i^{y_j}=x_i (j>i) are asserted to be 'easily seen pictorially' with reference to Figures 4 and 5. These formulas are load-bearing: they determine the presentation of G_n used in Problem 2, and hence the whole HNN-extension argument. The figures alone are not a substitute for a verifiable algebraic proof. Please provide a derivation from standard Artin relations, or an explicit labelled braid diagram with a written conjugation calculation for a representative case and a clear statement that the remaining cases are analogous.
  2. [Section 4, Theorem 2] The induction proving Theorem 2 has no stated base case. The text says the scheme is '(1) for k = n, (2) for k = n, (3) for k = n, (1) for k = n+1', but an induction needs an explicit starting point. In particular, the definitions of L_1, A_1, and p_1, and the verification of the base case (or cases) of the chain, are missing. Please spell out the base case and the simultaneous inductive hypothesis used in the cyclic implication.
  3. [Section 4, Theorem 2, Step III] The statement that the restriction of d_{n+1} to L_{n+1} 'must be injective, since its image, H_n, is free of rank n by (2)' is not valid as written: a surjection from a finitely generated group onto a free group need not be injective unless the domain is also free of the same rank and Hopfianity is invoked. The missing argument is that L_{n+1} is generated by n elements, so the composite F_n → L_{n+1} → H_n ≅ F_n is an epimorphism; Hopfianity of F_n then forces the first map to be injective. The same Hopfianity reasoning is needed in the earlier sentence 'Since d_n(L_n)=H_{n-1}, the group L_n would be free of rank n−1'. Please write these steps out explicitly.
minor comments (4)
  1. [Section 2, Proposition 1] The proof of the second equality ‹A_{0,1}...‹A_{0,n}=1 is only sketched ('one shows that this braid is in the center... and notices that its image... is zero'). Since this equality is used in the reduction to Problem 1, please give a complete proof or a precise reference for each of the two assertions.
  2. [Section 3, Lemma 4] In the displayed computation for p_i^{y_i}, the expression 'x_i · x_{i+1} . . . x_i' appears to contain a typo; it should presumably read 'x_i · x_{i+1} . . . x_n'. Please correct this and check the surrounding indices.
  3. [Section 4, after K_n] The notation 'x_i := p_i p_{i+1}^{-1} for 1 ≤ i < n−1' is confusing because p_n is not yet introduced in K_n and the later definitions of x_{n-1} and x_n are given in G_n. Please clarify the indexing and state explicitly which group each x_i belongs to.
  4. [Section 4, Theorem 2, Step III] In the chain of inclusions in Step III, the phrase 'by (1)' appears to be attached to the wrong equality: the inclusion L_{n+1} ∩ A_{n+1} ⊂ d_{n+1}^{-1}(H_n ∩ <A_n,p_n>) ∩ L_{n+1} follows from the definitions, while the subsequent equality uses the inductive hypothesis (3). Please rephrase for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the proof reduces injectivity to an independent freeness statement and solves it group-theoretically.

full rationale

The derivation is self-contained. Theorem 1 is not used as an input; after an explicit computation of Theta_n(z_i) in Proposition 2, injectivity is reduced to the assertion that certain braids generate a free subgroup of rank n (Problem 1), an equivalent but independent group-theoretic statement. The subsequent solution is internal: an automorphism sends the generators to y_i x_i y_i, Lemma 3 gives a semidirect-product description of the ambient subgroup, and Problem 2 is solved by an HNN-extension argument relying on Britton's normal form (Lemma 6 and Theorem 2). The only self-citation, [AIM25] for A0,1...A0,n = z_n^{-2}, is peripheral: the needed redundancy of the generating set follows from the second identity of Proposition 1, for which the paper gives its own argument, and the cited equation is not the injectivity theorem. The pictorial verification in Lemma 3 is a rigor gap and a potential correctness risk, but it is not a circular step: the asserted conjugacy relations are checked against the braid group, not derived from the target result. No fitted parameter, imported uniqueness theorem, or ansatz-via-citation appears in the load-bearing chain.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The proof relies only on standard facts about pure braid groups, free groups, and HNN extensions. No new objects are introduced and no free parameters are fitted to data.

assumptions (4)
  • standard math Pure braid group facts: Ker(d_{n+1}) is free with basis x_i=A_{i,n+1}, and P_{n+1} splits as F_n ⋊ P_n.
    Used in Lemma 3 to identify G_n as a semidirect product; classical Fadell-Neuwirth result.
  • standard math Finitely generated free groups are Hopfian.
    Used in Section 3 to pass from a free rank-n image to injectivity of Θ_n.
  • standard math Britton's normal form for HNN extensions.
    Used in Lemma 6 to conclude that L_n * <p_n> injects into G_n.
  • standard math The center of P_3 is generated by A_{1,2}A_{1,3}A_{2,3}.
    Used in the final paragraph of Theorem 2 to separate two cyclic subgroups in P_3.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On injectivity of Cohen-Wu homomorphism." pith.science (2026). https://pith.science/paper/USJ7R2HB

@misc{pith2026250709636,
  author       = {Pith},
  title        = {Pith review of: On injectivity of Cohen-Wu homomorphism},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/USJ7R2HB}},
  note         = {Machine review of arXiv:2507.09636}
}
abstract

We give a new elementary proof of the theorem that a natural map from Milnor's construction $F[S^1]$ to the simplicial group $\mathrm{AP}$ of pure braids is injective. Our approach is group-theoretic and does not rely on Lie algebras.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

  1. [1]

    Action of automorphisms of pure braid groups on homotopy groups of two-sphere

    I. Alekseev, V. Ionin, and M. Mikhailov, ``Action of automorphisms of pure braid groups on homotopy groups of two-sphere'', to appear in: Homology Homotopy Appl. (2025). arXiv: 2112.05805 https://arxiv.org/abs/2112.05805

  2. [2]

    F. R. Cohen, J. Wu, On braid groups, free groups, and the loop space of the 2-sphere, from: ``Categorical decomposition techniques in algebraic topology'', Progr. Math. 215 (2004), 93–105

  3. [3]

    F. R. Cohen, J. Wu Artin's braids groups, free groups, and the loop space of the 2-sphere, Q. J. Math. 62.4 (2010), 891–921

  4. [4]

    R. C. Lyndon, P. E. Schupp, Combinatorial Group Theory, Springer-Verlag, New York, (2001)

  5. [5]

    Murasugi, B

    K. Murasugi, B. Kurpita, A Study of Braids, Springer, 1999

Pith tools

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