{"id":"9a9c09cf-7c92-42ad-a34a-26e869f9a13e","arxiv_id":"1908.06465","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The golden-mean semi-Siegel Hénon family is not weakly J*-stable at any parameter, yielding Newhouse phenomenon and disconnected Julia sets on dense parameter sets.","lead":"Sufficiently dissipative semi-Siegel complex Hénon maps with the golden-mean rotation number are structurally unstable at every parameter, and this forces infinitely many sinks and disconnected Julia sets for dense sets of parameters. The result shows that a two-dimensional family of holomorphic maps that is a perturbation of the quadratic family bifurcates in a strong, unavoidable way.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof of the Main Theorem rests on the unproved assertion, after (6.2), that the tangency parameters form a dense set; 'discrete dense' is contradictory and no argument is supplied, so instability at every parameter is not established.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing gap: the unsupported density assertion after (6.2). My reading confirms that this is the step on which the Main Theorem's 'every parameter' conclusion depends. I also checked the surrounding argument: given dense non-persistent heteroclinic tangencies, Theorem 1.3(v) of Dujardin–Lyubich indeed forces the family to bifurcate at every parameter, and the cited renormalization estimates would make the density claim plausible. The difficulty is that the paper supplies no proof of density, and the phrase 'discrete dense' suggests a genuine confusion about the topological meaning of 'discrete'. I do not see a counterexample or an internal contradiction beyond the missing argument, so the appropriate disposition remains the reader's CONDITIONAL verdict: the gap is real but likely fillable. Hence no change to the reader's verdict is needed.","tokens_in":18017,"tokens_out":12013,"duration_ms":131732,"concrete_test":"Prove the missing density claim directly. Set Q_n = q_{2(n-1)} and write (6.2) as F_{n,k}(a) = a^{Q_n} - c\\,\\lambda_*^{2k}(1+\\varepsilon_{n,k}(a)) = 0, with c = u_*(x_*-1)/\\bar\\Delta_v. Fix any a_0 = re^{i\\theta} in D_{\\bar\\epsilon}\\{0\\}. Choose k_n so that |c|\\,|\\lambda_*|^{2k_n} approximates r^{Q_n} within a factor 1+O(1/Q_n), and choose m_n so that Q_n\\theta + 2\\pi m_n approximates \\arg c + 2k_n\\arg\\lambda_* within O(1/Q_n). Then verify, on a disk of radius about 1/Q_n around a_0, that the leading term |a^{Q_n} - r^{Q_n}e^{iQ_n\\theta}| dominates the accumulated O(\\rho^n)+O(\\rho^k) error terms uniformly in a_{; if it does, Rouché's theorem gives an actual solution of (6.2) in every 1/Q_n-neighborhood of a_0, proving X is dense. If this estimate fails for some a_0, the Main Theorem should be weakened to instability on the set X only.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The decisive step is the sentence after (6.2): the set X of parameters solving the heteroclinic tangency condition is said to be a 'discrete dense subset' of D_{\\bar\\epsilon}\\{0\\}. This is the only place where the Main Theorem is upgraded from 'there are some unstable parameters' to 'every parameter is unstable': if X is not dense, an open set avoiding X could be a weakly J*-stable neighborhood for some parameter. No proof is given; the word 'Observe' is the entire argument. The wording is also self-contradictory: a topologically discrete subset of an open disc has no accumulation points and cannot be dense. If 'discrete' is only meant as 'each solution is isolated for fixed n,k', density still requires a separate argument. The leading equation is of the form a^{q_{2(n-1)}} = \\lambda_*^{2k} u_*(x_*-1)/\\bar\\Delta_v (1+O(\\rho^n)+O(\\rho^k)); for fixed n,k it has q_{2(n-1)} roots whose angular grid has spacing 2\\pi/q_{2(n-1)}, so density is plausible as n grows. But it is not automatic: the error terms in (6.2) depend on a, and the paper neither states the required uniformity nor bounds how far the true roots move from the leading-order roots. Without a proof that every open U\\subset D_{\\bar\\epsilon}\\{0\\} meets X, the conclusion that the family is not weakly J*-stable at every parameter, and consequently Corollaries 1.13 and 1.14, is unsupported. The gap appears fillable by a Rouché/approximation argument, so the right status is conditional rather than rejected.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the family F_a of sufficiently dissipative golden-mean semi-Siegel Hénon maps. The main theorem asserts that there exists \\bar{\\epsilon}>0 such that F_a is not weakly J*-stable at every parameter a in D_{\\bar{\\epsilon}}\\{0\\}. The proof uses the renormalization theory from the authors' prior work: for each n,k, the stable manifold of a saddle point near the nth fold is nearly vertical, while a pulled-back unstable manifold is a nearly quadratic graph. The horizontal distance between these two graphs is computed asymptotically, and a tangency occurs exactly when equation (6.2) holds. The paper then claims, with the single word \"Observe,\" that the set of solutions to (6.2) forms a \"discrete dense\" subset of D_{\\bar{\\epsilon}}\\{0\\}, and uses this to conclude that every parameter admits arbitrarily close non-persistent heteroclinic tangencies. Invoking Dujardin--Lyubich's Theorem 1.3(v), the authors derive the Main Theorem and the corollaries on the Newhouse phenomenon and disconnected Julia sets.","tokens_in":18322,"tokens_out":8944,"duration_ms":94961,"significance":"If the Main Theorem is correct, it is a significant result: it gives the first example of a dissipative Hénon family that is weakly J*-unstable at every parameter in a punctured disk, and it yields, via [DuLy], a dense G_delta set of Newhouse parameters and a dense set of parameters with disconnected Julia set in the semi-Siegel family. The geometric strategy, comparing nearly vertical stable manifolds with quadratic unstable graphs via renormalization microscope maps, is illuminating and well-suited to the problem. The paper builds on published theorems [Yan1], [Yan2], [GaRaYam] with independent derivations, and I do not see circularity. However, the final step contains an unproved and internally contradictory density assertion that is load-bearing for the \"every parameter\" conclusion; the gap appears fillable by a Rouché-type argument, so the result should be treated as conditional pending that proof.","major_comments":[{"comment":"The assertion that the set X of parameters solving (6.2) is a \"discrete dense subset\" of D_{\\bar{\\epsilon}}\\{0\\} is unproved and self-contradictory: a topologically discrete subset of a connected open set cannot be dense. If \"discrete\" is only intended to mean that solutions are isolated for each fixed pair (n,k), density still requires a separate argument. Equation (6.2) is of the form a^{q_{2(n-1)}} = \\lambda_*^{2k} u_*(x_*-1) / \\bar{\\Delta}_v (1+O(\\rho^n)+O(\\rho^k)), whose leading-order solutions lie on a grid with angular spacing 2\\pi/q_{2(n-1)}; density is plausible as n grows, but the error terms depend on a, and the paper neither states the required uniformity nor bounds how far the true roots move from the leading-order roots. Without a proof that every open U \\subset D_{\\bar{\\epsilon}}\\{0\\} meets X, the Main Theorem, and consequently Corollaries 1.13 and 1.14, is unsupported.","section":"Section 6, after Eq. (6.2)"},{"comment":"The proof asserts that the tangency q_{n,k} \"does not persist in any neighborhood of a1\" without demonstrating that it is a simple tangency whose separating distance has nonzero derivative with respect to a. Weak J*-stability requires that the intersection point moves under an equivariant unbranched holomorphic motion, so a tangency at a1 only contradicts stability if no holomorphic branch of the intersection point exists near a1. The paper should show, for instance, that the horizontal separation between M^s_loc(s_n^{n+k}) and C_n near the tangency changes at first order in (a-a1), so that the double intersection splits into two or disappears and no single branch persists. This is a standard unfolding argument, but it is not supplied.","section":"Proof of Main Theorem, last paragraph"},{"comment":"The density argument is the only place where the conclusion is upgraded from \"there exist unstable parameters\" to \"the family is not weakly J*-stable at every parameter.\" If the solution set X were merely infinite with an accumulation point at 0, the argument would only show instability at 0, not at every a. The paper must prove that X accumulates at every point of D_{\\bar{\\epsilon}}\\{0\\}, including points away from 0 where the error terms in (6.2) have not been controlled. This is a load-bearing step that cannot be replaced by the heuristic observation about the leading-order roots.","section":"Section 6, equation (6.2)"}],"minor_comments":[{"comment":"The error term in (6.2) is printed as (1+O(\\rho^n)+O(\\rho^n)); the second error should almost certainly be O(\\rho^k), matching Proposition 4.5.","section":"Section 6, Eq. (6.2)"},{"comment":"The abstract says the maps are \"not J-stable in a very strong sense,\" while the Main Theorem concerns weak J*-stability; the terminology should be aligned to avoid confusing readers who distinguish structural J-stability from weak J*-stability.","section":"Abstract"},{"comment":"The phrase \"discrete dense subset\" is contradictory and should be replaced by a precise statement such as \"the union over n,k of the solution sets is dense.\"","section":"Section 6, paragraph before proof of Main Theorem"},{"comment":"The paper would benefit from a short discussion of why the tangencies produced by (6.2) are heteroclinic tangencies for the original Hénon map F_a and not merely for the renormalized pair; this is implicit in the identifications of Sections 4--5, but an explicit sentence would improve readability.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The density gap after (6.2) is the main obstacle; it looks fixable by a Rouché/approximation argument using the explicit leading-order form of the equation, and the rest of the renormalization technology appears sound. I would encourage the authors to supply the missing uniformity estimates and to replace the 'discrete dense' phrasing. I do not see grounds for rejection, provided the density and non-persistence steps are completed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nYou should know two things. The result is important if true: it gives the first global structural instability for golden-mean semi-Siegel Hénon maps, with Newhouse phenomenon on a dense G_delta and disconnected Julia sets on a dense parameter set as corollaries via Dujardin-Lyubich. The renormalization strategy continues prior work by the same authors, but the target is genuinely new. The second thing is that the main theorem as written is not established. After (6.2), the proof says the solution set forms a 'discrete dense subset' of D_{bar-epsilon}\\{0} and gives no argument. A subset of a disc cannot be both discrete and dense; if 'discrete' only means isolated for fixed n,k, that does not imply density. This is the step that upgrades 'there are tangencies somewhere' to 'every parameter is unstable', and it is load-bearing: both corollaries and the main theorem collapse if the solution set only accumulates at 0.\n\nWhat the paper does well: the geometric estimates in Sections 4 and 5 are careful and detailed. The point is to show that, near the fold, stable manifolds stay almost vertical while unstable manifolds are quadratic and move in a controlled way, and the paper tracks the corrections in a and in the Fibonacci denominators. Corollary 5.6 is a neat telescoping argument, and Proposition 5.5 gives concrete nonzero constants. The reliance on prior work is legitimate: [Yan1], [Yan2], and [GaYam] are published results with independent derivations, not assumptions of the conclusion.\n\nThe soft spots are in proportion. The density gap is major, but it looks fixable: (6.2) is a leading-order equation a^{q_{2(n-1)}} = lambda_*^{2k} const, and for fixed n,k it has many roots whose angular separation shrinks as n grows. A Rouche or uniform-continuity argument could prove density if the error terms are uniform in a. The paper does not state such uniformity. There is also a minor gap in Corollary 5.6: the infinite telescoping sum needs a summability estimate (|lambda_*|^k q_{2(n+k)} is likely summable, but the bound should be written). Neither gap looks fatal to the strategy, but the density one is exactly what a referee must see.\n\nBottom line: this deserves a serious referee. I would send it out, but I would tell the referee to insist on a proof of the density claim or, failing that, a weakening of the theorem to 'a dense set of unstable parameters'. If the density can be proved, this is a substantial paper for the Hénon and Siegel renormalization community.","headline":"A genuinely new instability route for semi-Siegel Hénon maps, but the main theorem needs a real proof of the density step before the 'every parameter' conclusion is trustworthy.","tokens_in":18895,"tokens_out":3962,"would_cite":false,"duration_ms":39771,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37F10","37F50","37F25"],"pacs":[],"model":"deepseek-v4-flash","headline":"The golden-mean semi-Siegel Hénon family is structurally unstable at every sufficiently small nonzero parameter.","keywords":["golden-mean semi-Siegel Hénon maps","weak J*-stability","holomorphic motions","renormalization","heteroclinic tangencies","Newhouse phenomenon","Julia set","structural instability"],"falsifier":"A reader could test the density claim by computing the universal constants $\\lambda_*$, $u_*(x_*-1)$ and $\\bar{\\Delta}_v$ from the renormalization fixed point and then solving $a^{q_{2(n-1)}} = \\lambda_*^{2k}u_*(x_*-1)/\\bar{\\Delta}_v$ for all $n,k$; if the resulting solutions leave an open gap anywhere in $\\mathbb{D}_{\\bar{\\epsilon}}\\setminus\\{0\\}$, the Main Theorem collapses. A direct numerical search for non-persistent heteroclinic tangencies in small parameter windows would also settle the question.","tokens_in":17763,"feed_emoji":"🌀","tokens_out":15382,"duration_ms":129056,"temperature":0.7,"pith_summary":"This paper proves that every sufficiently dissipative golden-mean semi-Siegel Hénon map is structurally unstable in a strong, higher-dimensional sense. Concretely, at every nonzero parameter $a$ with $|a|$ small, the family is not weakly $J^*$-stable: even allowing branched holomorphic motions, there is no neighborhood of $a$ over which the saddle periodic points move holomorphically. The reason is that renormalization reveals stable and unstable manifolds of saddles creating heteroclinic tangencies at parameters arbitrarily close to every $a$, and such tangencies cannot persist. Two known consequences then follow: a dense $G_\\delta$ set of these maps has infinitely many attracting periodic orbits (the Newhouse phenomenon), and a dense set of parameters has disconnected Julia sets. This matters because, unlike the attracting and semi-parabolic Hénon families, the semi-Siegel family is non-rigid at every parameter, with bifurcations occurring everywhere.","feed_headline":"Golden-mean Hénon maps are unstable at every small parameter","feed_subtitle":"Every parameter gains non-persistent tangencies, so Newhouse sinks and disconnected Julia sets are dense.","key_machinery":"The central mechanism is the renormalization microscope of the golden-mean semi-Siegel family. The $n$-th renormalization $\\Sigma_n=(A_n,B_n)$ is a rescaled first-return map whose one-dimensional projections $\\eta_n,\\xi_n$ converge to the universal renormalization fixed point $(\\eta_*,\\xi_*)$, with universal scaling factor $\\lambda_*$. A dynamically defined point $(\\kappa_n,0)$, the $n$-th fold, arises as the limit of the microscope maps $\\Phi_n^k$ and moves holomorphically in $a$. Near this fold, the local stable manifolds of selected saddle periodic points are graphs $y\\mapsto\\psi_n^k(y)$ with derivative $O(a^{q_{2n}})$, while a related unstable manifold $\\mathcal{C}_n$ is the graph of a map $y\\mapsto\\tau_n(y)$ that is $O(a^{q_{2(n-1)}})$-close to $\\xi_n$ and has a unique vertical tangency. Equating the horizontal positions of these manifolds via Proposition 4.5 and Corollary 5.7 gives the tangency equation (6.2), whose solutions over $n,k\\in\\mathbb{N}$ are asserted to be dense; that density is the step that turns local geometry into global instability.","core_discovery":"The central claim is the Main Theorem: there exists $\\bar{\\epsilon}>0$ such that the golden-mean semi-Siegel Hénon family $(F_a)$ is not weakly $J^*$-stable at every parameter $a\\in\\mathbb{D}_{\\bar{\\epsilon}}\\setminus\\{0\\}$. For each such $a$, no neighborhood admits an equivariant unbranched holomorphic motion of the saddle periodic points, and no equivariant branched motion of $J^*(F_a)$ exists over any neighborhood. The proof shows, via renormalization, that near a dynamically defined point called the fold, the stable manifolds of certain saddle orbits are nearly vertical while unstable manifolds are quadratic graphs, and their relative position changes with $a$ at the tiny scale $a^{q_{2n}}$. This forces, in every neighborhood of every parameter, a heteroclinic tangency between a stable and an unstable manifold that cannot persist; by the stability criterion of Theorem 1.3(v), such a tangency would have to persist if the family were weakly $J^*$-stable. Consequently, the family is unstable at every small nonzero parameter, and the corollaries of that criterion give a dense $G_\\delta$ set of Newhouse parameters and a dense set of parameters with disconnected Julia sets.","pith_inferences":["Inference: if the density assertion in equation (6.2) can be proved, the same argument is likely to extend to any bounded-type rotation number whose renormalization converges, so this instability may be a general feature of semi-Siegel Hénon maps rather than a golden-mean special case.","Inference: the tangencies occur at exponentially small scales $a^{q_{2n}}$, so the unstable set, while dense, could still be small in measure; the paper does not estimate the measure of stable parameters, and a measure estimate would be a natural next step.","Inference: numerically computing the universal constants $\\lambda_*$, $u_*(x_*-1)$ and $\\bar{\\Delta}_v$ would give explicit predicted locations of non-persistent tangencies, offering a direct quantitative check of the renormalization geometry.","Inference: if the open question $J^*(F_a)=J(F_a)$ were answered positively for this family, the theorem would upgrade automatically to ordinary $J$-instability at every small parameter; the paper leaves this upgrade conditional."],"forward_implications":["At every $a\\in\\mathbb{D}_{\\bar{\\epsilon}}\\setminus\\{0\\}$, the family bifurcates immediately: no neighborhood is weakly $J^*$-stable, so saddle periodic points cannot be followed by any equivariant holomorphic motion, branched or unbranched.","A dense $G_\\delta$ set of these Hénon maps exhibits the Newhouse phenomenon: infinitely many coexisting attracting periodic orbits.","A dense set of parameters has disconnected Julia set $J(F_a)$.","Heteroclinic tangencies occur at parameters accumulating at every point of the punctured disk, making the bifurcation set dense rather than isolated."],"supporting_citations":[{"why":"It equates weak J*-stability with persistence of homoclinic and heteroclinic tangencies (Theorem 1.3(v)), which turns the constructed tangencies into non-stability and yields the Newhouse and disconnected-Julia corollaries.","marker":"[DuLy]"},{"why":"It provides the quantitative renormalization estimates (Theorems 2.10-2.12) describing the universal shape of $A_n$ and $B_n$ and the microscope derivative.","marker":"[Yan1]"},{"why":"It supplies Theorem 3.3, the expansion of $A_n\\circ\\Phi_{n+1}$ used to locate the second vertical tangency.","marker":"[Yan2]"},{"why":"It supplies the a priori bounds and exponential convergence of the one-dimensional maps $\\xi_n$ in Proposition 3.1, underpinning the fixed-point and inverse-branch estimates.","marker":"[Yam]"},{"why":"It defines the one-dimensional renormalization fixed point $(\\eta_*,\\xi_*)$ and the operator whose universality gives $\\lambda_*$ and the limiting maps used throughout.","marker":"[GaYam]"}],"fun_headline_variants":["Golden-mean Hénon maps unstable at every parameter","No stability for golden-mean Hénon maps at any small parameter","Every small parameter yields golden-mean Hénon instability","Golden-mean Hénon: structural instability everywhere"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the assertion after equation (6.2) that the set of parameters $a$ solving the tangency equation is dense in $\\mathbb{D}_{\\bar{\\epsilon}}\\setminus\\{0\\}$; the word 'Observe' is the only support given, and if that solution set merely accumulates at $0$ rather than being dense, the conclusion that every parameter is unstable does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Golden-mean Hénon maps unstable at every parameter","No stability for golden-mean Hénon maps at any small parameter","Every small parameter yields golden-mean Hénon instability","Golden-mean Hénon: structural instability everywhere"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000468,"raw_usage":{"total_tokens":2294,"prompt_tokens":866,"completion_tokens":1428,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":482,"completion_tokens_details":{"reasoning_tokens":1359}},"tokens_in":482,"tokens_out":1428,"duration_ms":11221,"temperature":1.0,"reasoning_tokens":1359,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:44:49.506717+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A reader could test the density claim by computing the universal constants $\\lambda_*$, $u_*(x_*-1)$ and $\\bar{\\Delta}_v$ from the renormalization fixed point and then solving $a^{q_{2(n-1)}} = \\lambda_*^{2k}u_*(x_*-1)/\\bar{\\Delta}_v$ for all $n,k$; if the resulting solutions leave an open gap anywhere in $\\mathbb{D}_{\\bar{\\epsilon}}\\setminus\\{0\\}$, the Main Theorem collapses. A direct numerical search for non-persistent heteroclinic tangencies in small parameter windows would also settle the question.","supporting_citations":[],"review_version":1}