REVIEW 3 major objections 3 minor 4 references
Monodromy groups of polynomials of composition length 2
T0 review · 3 major / 3 minor · reviewed 2026-08-03 · deepseek-v4-flash
Pith's one-line read Two-factor polynomial compositions have almost-maximal monodromy unless they fall into an explicit short list — a dichotomy that feeds a solution of a long-standing reducibility problem.
desk verdict A real classification result, but its exception list depends on Magma searches not reproducible from the text; send to review with a demand for the code. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the block kernel Γ = ker(Mon(g∘h) → Mon(g)), viewed as a subgroup of the wreath product Mon(h)^{deg g} ⋊ Mon(g). The proof establishes largeness by showing Γ contains the full diagonal power of the socle, the augmentation subgroup, or a subgroup of codimension one, using four ingredients: (1) the classification of monodromy groups of indecomposable polynomials; (2) permutation-module submodule bounds (only the diagonal, augmentation, and full modules are invariant under A_n/S_n); (3) braid-group and branch-point-coalescing arguments that reduce configurations to normal forms; and (4) the Ritt-move dichotomy, which forces exceptional embeddings into direct products. The
What would settle it
Run an independent exhaustive search over all transitive groups of degree at most 62 that embed in the relevant wreath products C_p ≀ G or S_4 ≀ G, looking for a group with a genus-zero generating tuple of polynomial ramification type (containing a full cycle and product equal to identity) that is not listed in Table 1 and whose kernel is not large in the paper's sense. Finding one would refute the completeness of Theorem 1.2.
Extended reading notes
Core claim
The authors' main discovery is the dichotomy of Theorem 1.2: for any two indecomposable polynomials g,h of degree > 1 over a characteristic-zero field k, the geometric monodromy group Mon(g∘h) either has a 'large kernel' — meaning the ordinary (or, in one solvable family, cyclic) powers of the socle of Mon(h) lie inside the kernel Γ of the block projection to Mon(g) — or g∘h is linearly equivalent to X^{p^2} or T_{p^2}, admits a Ritt move, or Mon(g∘h) appears in an explicit table of exceptional groups (all with h equivalent to X^2, X^3, or T_3). The proof reduces to algebraically closed fields, draws on the known classification of primitive monodromy groups of indecomposable polynomials, and
Load-bearing premise
The theorem's explicit exception list is only as complete as the finite computer searches over transitive groups that enumerated it; if one of those searches missed a genus-zero generating tuple, an exceptional group is missing from the list.
Editorial extensions
If this is right
- If Theorem 1.2 is correct, every length-2 polynomial over a number field has either a large kernel or one of the listed exceptional monodromy groups; this yields the first complete classification of this kind.
- The solvable restriction of the theorem is a key step in a companion paper's solution of the long-standing reducibility problem for f(X) − g(Y).
- For prime-degree polynomials linearly related to Chebyshev polynomials, the large-kernel theorem implies the infinite dynamical Galois group contains the infinite iterated wreath product of cyclic groups of order p, with relative Hausdorff dimension at least 1 − log_p(2)/(p(1+log_p 2)).
- The methods extend inductively: Theorem 6.3 gives large-kernel bounds for compositions of arbitrarily many prime-degree polynomials.
- The list of exceptions is finite and of bounded degree (at most 62), so the theorem reduces the study of two-step monodromy to a concrete, checkable family.
Reading between the lines
- The explicit exception list is small and of bounded degree, suggesting that as degrees grow, the only failures of largeness are the monomial and Chebyshev degenerations; this could be tested by random polynomial generation over finite fields.
- The same ramification-coalescing technique may prove a version of the theorem for rational functions (not just polynomials), where the branch at infinity is no longer a full cycle.
- The Hausdorff-dimension lower bound for Chebyshev-like maps suggests that the exact relative density of the dynamical Galois group may be computable by a finer analysis of the growth of the 2-part.
- A natural testable extension is to classify length-3 compositions: Theorem 1.2 plus induction on Ritt moves should yield a similar dichotomy, though the exception list may grow with the number of factors.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies the monodromy group of a composition g∘h of two indecomposable polynomials over a field of characteristic 0. It introduces a 'large kernel' condition, a weakening of the maximal kernel Mon(h)^{deg g}, and proves Theorem 1.2: unless the composition is linearly equivalent to T_{p^2} or X^{p^2}, admits a Ritt move, or Mon(g∘h) is one of the groups in Table 1, the kernel is large. The proof splits into solvable and nonsolvable cases, using the classification of indecomposable polynomial monodromy groups [Mül95], group-theoretic lemmas, braid-group coalescing arguments, and Magma computations. Applications to higher compositions and arithmetic dynamics are sketched.
Significance. If correct, this is a substantial result: it gives an essentially complete classification of length-2 monodromy groups up to a finite explicit list, with immediate applications to the Davenport–Lewis–Schinzel problem and to dynamical Galois groups. The paper also contains several generally useful tools, such as Lemma 3.9 on kernels in wreath products and Lemma 3.12 on subgroups of U≀V, and it states explicit exceptional groups and ramification data. However, the completeness of the classification currently rests on computer searches that are not fully documented in the text and on an imported result from [KNR24]; these dependencies are load-bearing for the main theorem.
major comments (3)
- [§4.4, §5.3] The proof of Theorem 1.2 depends for its exhaustiveness on a series of Magma searches. In §4.4, after reducing to the case that every branch point of g∘h is a branch point of g, the paper states that 'the assertion is most conveniently checked using a direct Magma search for genus zero 3- and 4-tuples inside S4≀S4≤S16' without specifying the exact tuple set, the equivalence relation, or the completeness criterion. Similarly, §5.3 lists ten transitive-group candidates and states that a 'direct search for generating genus zero tuples leaves only 39T206, 39T248, 33T60', again without details. Since Theorem 1.2 is an 'unless' classification, a missed tuple would enlarge the exception list. The ancillary code is not part of the manuscript and was not supplied for review. This is a load-bearing completeness issue; please provide a detailed, reproducible description of the searches, ideally wit
- [§5.1, Theorem 2.5] The nonsolvable-h case, which is a central ingredient of Theorem 1.2, is not proved in the paper. The proof is essentially one sentence: 'This follows from [KNR24, Corollary 4.4]' except for the Ritt-move subcase. The reader cannot verify the main theorem without consulting [KNR24], and if that corollary has any unlisted exceptional case, Theorem 1.2 is incomplete. Please state Corollary 4.4 explicitly (or include its proof) and confirm that no exceptional cases beyond those in Theorem 2.5 can occur.
- [Proposition 3.14, §5.2.2, §5.3] Proposition 3.14 is used in the nonsolvable case to rule out embeddings Mon(g∘h)≤Mon(h)×Mon(g) and thus to force large kernel or a Ritt move. Its proof again delegates a decisive exclusion to Magma: 'a direct check in Magma's transitive group database yields that no embedding ...' and 'can be verified with Magma to indeed occur'. The completeness of this check is part of the classification, since a missed embedding would enlarge the exception list. The same applies to the Magma verifications in §5.2.2 (splitness of Γ~.Mon(g), construction and decomposition of induced modules) and §5.3. Please provide the code or a detailed enough specification for independent verification.
minor comments (3)
- [Lemma 3.9] The proof of Lemma 3.9 refers twice to 'Assumption iii)', but the statement lists only assumptions i) and ii). The intended reference is presumably Assumption ii).
- [§4.1, proof of Theorem 2.1] The notation appears to interchange p and q in places. Since deg(g)=q and deg(h)=p, the kernel Γ should be a subgroup of C_p^q (or D_p^q), not C_q^p. Please check all occurrences of these symbols in §4.1 and in the statement of Lemma 3.2/3.3 as used there.
- [§5.2.2] The sentence 'A Magma computation, using the list of exceptional ramification types from [Mül95], now yields the following intermediate result' is a black box even by the standards of the paper. At minimum, state precisely which ramification types were used as input so that the computation is reproducible.
Circularity Check
One load-bearing branch of the nonsolvable case is imported from the authors' own [KNR24]; the rest of Theorem 1.2 is an independent case analysis with no definitional or fitted-input circularity.
-
self citation load bearing
[Section 5.1, proof of Theorem 2.5]
"This follows from [KNR24, Corollary 4.4], except in the case when there is a Ritt move for g∘h, in which case necessarily h = X^p up to linear equivalence, by Ritt's theorems."
Theorem 2.5 supplies the branch of Theorem 1.2 where g is an arbitrary indecomposable polynomial and h is nonsolvable. Instead of a proof, the paper cites [KNR24, Corollary 4.4], a preprint by two of the same authors (König and Neftin) plus Rosenberg. Thus a substantial part of the classification is obtained from the authors' own prior work rather than derived in the present paper. This is a load-bearing self-citation, though not an equation-level reduction: the cited corollary is prior work, and the rest of the theorem is proved independently.
full rationale
Apart from the imported Theorem 2.5, I found no construction-level circularity. The 'large kernel' predicate is a stipulated weakening, not the theorem's conclusion; the exceptions in Table 1 are obtained by explicit ramification and module arguments, Ritt-move classification, and finite Magma searches rather than by fitting the answer into the statement. The solvable case (Theorems 2.1–2.4) is proved from branch configurations and lemmas on cyclic codes and induced modules; Theorems 2.6–2.7 are likewise proved by combining [Mül95] with kernel estimates and computational checks. The Magma searches (§4.4, §5.2.2, §5.3) are a completeness premise, not a circular reduction: if a search missed a genus-zero tuple, Table 1 would be incomplete, but that is a verification risk. The score reflects the one load-bearing same-author citation in Theorem 2.5; the central claim still has substantial independent content.
Assumptions & free parameters
assumptions (6)
- standard math Riemann–Hurwitz formula and branch-cycle description: inertia generators satisfy product=1 and the genus formula (Eq. 1).
- domain assumption Ritt's polynomial decomposition theorems over characteristic 0: non-unique decomposition iff a Ritt move, with Ritt moves of types (T_n,T_m) and (X^n, X^s h(X^n)).
- domain assumption Müller's classification of primitive monodromy groups of indecomposable polynomials over C and Q (Proposition 3.1).
- domain assumption Mortimer's classification of submodules of F_p permutation modules of known doubly transitive groups (used in Lemma 3.11 and §5.2.2).
- domain assumption Base field has characteristic 0; geometric monodromy is taken over an algebraically closed field.
- domain assumption KNR24, Corollary 4.4: a large-kernel conclusion for certain compositions with a nonsolvable factor and no Ritt move.
Cite this review
Pith. "Pith review of Monodromy groups of polynomials of composition length 2." pith.science (2026). https://pith.science/paper/CIVC4MKB
@misc{pith2026260327609,
author = {Pith},
title = {Pith review of: Monodromy groups of polynomials of composition length 2},
year = {2026},
howpublished = {\url{https://pith.science/paper/CIVC4MKB}},
note = {Machine review of arXiv:2603.27609}
}
read the original abstract
We study the monodromy groups of compositions of two indecomposable polynomials. In particular, we show that such monodromy groups either fulfill a certain ``largeness" property, or belong to an explicit list of exceptions. Such largeness results are crucial for dealing with compositions of more than two polynomials, and consequently are expected to have a wide range of applications to problems concerning the arithmetic of polynomials and arithmetic dynamics. In particular, our main result is a key ingredient in the solution of a long-standing open problem due to Davenport, Lewis and Schinzel, achieved in a companion paper.
Reference graph
Works this paper leans on
-
[1]
[AH25] Ophelia Adams and Trevor Hyde,Profinite iterated monodromy groups of unicritical polynomials, 2025, https://arxiv.org/abs/2504.13028. [AMT20] Jacqueline Anderson, Michelle Manes, and Bella Tobin,Cubic post-critically finite polynomials de- fined overQ, LMS Journal of Computation and Mathematics23(2020), 20–34. [BCP97] Wieb Bosma, John Cannon, and C...
arXiv 2025
-
[1966]
[DLS61] Harold Davenport, D. J. Lewis, and Andrzej Schinzel,Equations of the formfpxq “gpyq, Q. J. Math., Oxf. II. Ser.12(1961), 304–312 (English). [Elk13] Noam Elkies,The complex polynomialsPpxqwithGalpPpxq´tq–M 23, ANTS X: Proceedings of the Tenth Algorithmic Number Theory Symposium, The Open Book Series1(2013), 359–367. [KNR24] Joachim K¨ onig, Danny N...
arXiv 1961
-
[1999]
[Mor80] Brian Mortimer,The modular permutation representations of the known doubly transitive groups, Proc. Lond. Math. Soc. (3)41(1980), 1–20 (English). [M¨ ul95] Peter M¨ uller,Primitive monodromy groups of polynomials, Recent developments in the inverse Galois problem. A joint summer research conference, July 17-23, 1993, University of Washington, Seat...
arXiv 1980
-
[2026]
4, 291–317 (English)
[CNC99] Pierrette Cassou-Nogu` es and Jean-Marc Couveignes,Explicit factorizations ofgpyq´hpzq, Acta Arith.87(1999), no. 4, 291–317 (English). [Cou96] Jean-Marc Couveignes,Tools for the computation of families of coverings, London Math. Soc. Lecture Note Ser. (Gainesville, United States) (Cambridge University Press, ed.), London Math. Soc. Lecture Note Se...
1999
Reviewed August 3, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.