{"id":"a2c9c90d-0176-4fbc-83cd-c67d691ad168","arxiv_id":"1908.03644","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A historical and mathematical study showing that Petrovic's and Fine's 1890s polygon methods generalize Newton-Puiseux theory and anticipate modern power geometry.","lead":"This paper resurrects a forgotten 1890s geometric method, developed by Mihailo Petrovic and Henry Fine, that uses polygons to predict how solutions of algebraic differential equations behave. It argues that this early work preceded and anticipated modern power geometry, and it restores the historical record.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's novelty and priority claims hinge on an unverified Briot–Bouquet precedence question: the retrospective itself contains a missing 'see [?]' at that exact point.","rationale":"The reader's weakest assumption was the faithfulness of the historical reading and the independence and priority of Fine and Petrovic. I agree and sharpen it: the load-bearing weak point is not the Fine–Petrovic independence per se, but the explicitly unresolved question of whether Briot–Bouquet already used a Newton polygon for ODEs. The paper's self-identified 'see [?]' is direct evidence of missing support, which the review rules require flagging. I verified that the mathematical core (Theorem 9) does not obviously fail: for a first-order algebraic ODE at a nonsingular point, each polygon vertex corresponds to a single combined monomial, so the leading coefficient at a vertex cannot vanish; hence a nonconstant leading term must come from a slanted edge, making the edge criterion plausible. Thus the concern is historical rather than mathematical. It is load-bearing because the paper's abstract and introduction make neglect and priority the central claims. The appropriate verdict remains CONDITIONAL pending completion of the references and verification of Briot–Bouquet; since the reader already reached this verdict, I recommend no change.","tokens_in":30851,"tokens_out":9115,"duration_ms":99938,"concrete_test":"Retrieve Briot and Bouquet's 1856 memoir 'Propriétés des fonctions définies par des équations différentielles' (J. École Polytechnique, cahier 36, pp. 133–198) and check whether it introduces a Newton-polygon construction for differential equations—specifically, associating monomials of an ODE with points and using convex-hull edges to determine leading exponents of solutions. Also check whether Fine (1889) or Petrovic (1894) cite this memoir. If the polygon construction appears there, the paper's priority and neglect claims must be revised; if it does not, the missing '[?]' should be replaced with a precise citation and the claim retained.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central historical assertion—that Petrovic's polygon method was an original, neglected generalization of the Newton–Puiseux method for ODEs—depends on the claim that no earlier work (in particular Briot–Bouquet) already contained the polygon construction. The paper explicitly flags this as unresolved in the final 'Retrospective' section: 'The first ideas to use the Newton – Puiseux methods in the theory of differential equations probably goes back to Broit and Bouquet, see [?].' This incomplete citation is an internal admission that the priority question has not been checked. If Briot–Bouquet's 1856 paper contains an ODE polygon construction, then (i) the abstract's statement that Petrovic's geometric ideas were 'left completely unnoticed' and (ii) the characterization of Fine 1889 and Petrovic 1894 as the originators of the ODE polygon method would both need substantial qualification. The same section also has a second unresolved placeholder for Cano ('see [?]'), further indicating the historical chain is not fully documented. The mathematical content of Theorem 9 is not itself put in doubt by this concern—for first-order equations vertices carry only a single monomial, so nonzero leading coefficients cannot cancel and only slanted edges can produce nonconstant leading terms—but the paper's primary contribution is historical, and that contribution rests on citations that the authors themselves have not completed.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript is a historical survey of geometric methods in the analytic theory of algebraic ordinary differential equations, centered on Mihailo Petrovic's 1894 polygon method and Henry Fine's 1889 independent construction. It presents the Petrovic polygon as a generalization of the Newton-Puiseux method from algebraic equations to algebraic ODEs, reconstructs both constructions with explicit coordinates, and proves a relation between them: Theorem 7 states that, for a nonsingular point, the Fine polygon of the translated equation coincides with the Petrovic polygon rotated by a quarter turn. The later sections survey applications: Petrovic's criteria for absence of movable poles and zeros (Theorem 9), his theorem on rational first-order ODEs with at most three essentially distinct transcendental single-valued solutions (Theorem 12), results on binomial ODEs, and the relation to modern power geometry, Painleve equations, and works of Cano, Bruno, and others. The final 'Retrospective' asserts that the first Newton-Puiseux-type ideas for differential equations may go back to Briot-Bouquet, but the relevant citation is left incomplete.","tokens_in":31144,"tokens_out":5778,"duration_ms":63172,"significance":"If the historical narrative is correct, the paper recovers a genuinely forgotten anticipation of modern power geometry: Petrovic's polygon test for movable poles and zeros is simpler to apply than Fuchs's discriminant-based criterion and predates the modern 'power geometry' literature by nearly a century. The paper's mathematical core is transparent and checkable: the coordinate definitions in Sections 3 and 4 are explicit, Theorem 7 follows directly from them, and Examples 1-3 contain enough detail for the reader to verify the leading asymptotics. The authors are also commendably candid about the limitations of planar polygons for higher-order equations (Section 3.1) and about Petrovic's absence from the Painleve program (Section 10.1). The main weakness is historical, not mathematical: the priority and neglect claims currently rest on incomplete citations, so the contribution is valuable but not yet in final form.","major_comments":[{"comment":"The retrospective contains two unresolved citation placeholders: 'Broit and Bouquet, see [?]' and 'Cano ... see [ ?]'. The first is load-bearing: the abstract's claim that Petrovic's geometric ideas were 'left completely unnoticed by the experts' and the attribution of the ODE polygon method to Fine and Petrovic presuppose that no earlier work, in particular Briot-Bouquet, already contained a Newton-Puiseux polygon construction for differential equations. The authors must locate the relevant Briot-Bouquet passage, state explicitly whether it contains such a construction, and adjust the priority and neglect claims accordingly. The incomplete Cano citation should also be completed.","section":"Retrospective (unnumbered, before Acknowledgements)"},{"comment":"The independence of Fine and Petrovic, stated as 'independently' in the abstract and elaborated in Section 1.2, is not supported by documentary evidence; the text only notes that neither cited the other. Because independence is part of the paper's novelty narrative, the authors should either provide evidence such as correspondence, library records, or contemporary reviews, or explicitly qualify the claim as an inference from the absence of mutual citation.","section":"Sections 1.1-1.2"},{"comment":"Theorem 9, the paper's central mathematical exhibit, is presented with only a two-sentence proof sketch: sufficiency is deferred to Proposition 1 and necessity to 'the methods of analytic theory,' with a reference to Fuchs's techniques, without spelling out the argument. Since the paper states it as a theorem in its own text and uses it to claim that Petrovic anticipated modern criteria, please provide a complete proof or an exact page reference to [67] where Petrovic's proof is given, and explain how Painleve's Theorem 6 enters. This is needed for the reader to verify the 'necessary' direction.","section":"Section 6, Theorem 9"}],"minor_comments":[{"comment":"The name 'Broit and Bouquet' is a typo for 'Briot and Bouquet'; elsewhere the spelling is correct.","section":"Retrospective"},{"comment":"Reference [25] is described as 'to appear, Bulletin of the AMS, 2020'; if the paper has appeared by now, the reference should be updated with full publication data.","section":"References, [25]"},{"comment":"The term 'angular coefficient' is used repeatedly without a definition; define it explicitly as the slope of an edge in the (M,N)-plane and state its relation to the exponent lambda in (15).","section":"Section 3 and Theorem 10"},{"comment":"The sentence 'each its vertex corresponds to the exactly one monomial' is grammatically garbled and should be rewritten; a one-sentence justification for first-order equations would also help, since for different terms with the same (M,N) the correspondence is to a unique monomial only after summing coefficients.","section":"Section 6, paragraph before Theorem 9"}],"recommendation":"major_revision","confidential_remarks":"For the editor: the stress-test concern is real. The two unresolved '[?]' citations sit at exactly the point where the paper's priority claim is vulnerable. I would support acceptance after the authors verify the Briot-Bouquet passage and the Cano reference, and after the independence claim is appropriately qualified. The mathematical content of Section 3 and Theorem 7 does not appear to need restructuring."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague, here's my honest read of 1908.03644. The one genuinely new mathematical item is Theorem 7: at a nonsingular point, the Fine polygon of the translated equation is the Petrovic polygon rotated by π/2. The proof is a coordinate check, and it is correct. The rest of the paper is a historical survey with modern restatements of Petrovic's 1894 thesis, and the worked examples (1–3) are clear and check out. If the historical claims hold up, this is a useful recovery of a neglected geometric method for first-order algebraic ODEs and a serviceable bridge to power geometry.\n\nNow the soft spots, and the stress-test note lands where it should. The retrospective has two unresolved placeholders, 'see [?]' after the Briot–Bouquet remark and another after Cano. The Briot–Bouquet one is load-bearing: the paper's priority claim—that Petrovic and Fine originated the ODE polygon method and that Petrovic was 'left completely unnoticed'—would need real qualification if Briot–Bouquet's 1856 work already contains a polygon construction for differential equations. The authors flag the open question themselves, which is honest, but a published survey cannot leave the key link of its chronology as a blank. Similarly, the claim that Fine and Petrovic worked independently is supported mostly by absence of evidence; plausible as a conjecture, not established as a fact.\n\nThe mathematics, separate from the history, is in good shape. Theorem 9 gives a simple geometric criterion for movable zeros and poles, and the paper is explicit that it does not cover movable critical points in general, which Fuchs's theorem does. No fitted parameters, no circularity. The examples demonstrate the method well.\n\nMinor: the biographical sections are longer than a math paper usually needs, but they do not damage the argument. The citation pattern is acceptable; self-citations appear as examples of power geometry rather than as padding.\n\nBottom line: this deserves a serious referee, not a desk reject. A referee should verify the primary sources, complete the missing citations, and soften or support the priority claims. After that, it becomes a citable historical synthesis with a small but real mathematical observation. I would cite it for Theorem 7 and for the historical context.","headline":"A useful historical synthesis with one correct new observation, but the priority claim rests on a Briot–Bouquet citation left as a placeholder.","tokens_in":31621,"tokens_out":2704,"would_cite":true,"duration_ms":29605,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-14T14:06:33.271836+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}