{"id":"0fd9dd00-5313-4683-8030-121b46a36203","arxiv_id":"2603.11238","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Harmonic 1D self-gravitating systems relax on a time ~N²; partially degenerate systems cross from N² to N scaling as particle number grows.","lead":"This paper uses exact 1D N-body simulations to show that self-gravitating particles in a harmonic potential relax on a timescale that grows with the square of the particle number, not the usual linear growth. It maps how partial degeneracy shifts relaxation between these two regimes, which matters for dense stellar cores such as those in dwarf galaxies.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The N² scaling for Harmonic is inferred from a narrow N range at a fixed KS threshold; partially degenerate runs show this branch can be pre-asymptotic, so the asymptotic large-N exponent for fully degenerate systems is not yet established.","rationale":"The reader's weakest assumption correctly identifies the fragility of the N² scaling: it is read from a narrow N window at a single threshold, and small-N points are excluded without a quantitative criterion. My stress test agrees with this and sharpens it. The Anharmonic family is the key evidence: it explicitly shows that the N² branch is a low-N regime for all finite ε, and the transition to N-scaling moves to larger N as ε decreases. For ε=0, the paper assumes the transition has moved to infinity, but no simulation at large N and no analytic argument establishes this. The data are therefore compatible with an alternative interpretation in which fully degenerate systems eventually relax linearly, just at N beyond the simulated range. This is not an internal inconsistency or a fatal error; it is an underdetermination of the central claim. The appropriate verdict remains conditional on additional large-N simulations or a quantitative theory of the transition. I do not recommend rejection: the numerical setup is exact, the threshold robustness check to D0=0.022 is present, and the paper is suitably cautious about its scope. The concern is sufficient to require confirmation, not sufficient to overturn the empirical result as a tentative finding.","tokens_in":17779,"tokens_out":5375,"duration_ms":54473,"concrete_test":"Run additional Harmonic simulations at N=200, 300, and if possible N=500, using the exact integrator or the O(N log N) approximate integrator from Roule et al. (2022), and measure t_rel at both D0=0.015 and D0=0.022. Compute the local log-log slope over disjoint N bins [21–50], [50–100], [100–300]. If the slope in the largest bin is consistent with 2, the N² claim is supported; if it falls toward 1, the N² branch is a finite-N transient. As a second check, measure N_trans(ε) for ε=0.05, 0.10, 0.25 and extrapolate to ε→0; a finite intercept would falsify the asymptotic N² claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim (Section 3.3, Figures 2–3) is that fully degenerate Harmonic clusters relax as t_rel ∝ N² t_dyn, in contrast to the N-scaling of the Balescu–Lenard equation. The evidence for this is t_rel measured at D0=0.015 (Eq. 8) over only 21≤N≤141, with small-N points excluded as 'polluted by small-N effects' without a stated criterion. The deeper problem is the Anharmonic series: for every finite ε>0, the same N² branch appears at low N and then bends to N at larger N, with the transition point N_trans(ε) increasing as ε decreases. Since ε=0 is not simulated at large N, the statement that N_trans diverges—so that N² is the true asymptotic law for Harmonic—is an extrapolation, not a measurement. The data are equally consistent with N_trans(ε) remaining finite at ε=0; in that case Harmonic would eventually relax linearly at some N≫141 and the headline result would be a pre-asymptotic transient. No analytic argument resolves this: the cited thermodynamic blocking (Deme & Fouvry 2025) is qualitative, and the 1/N² kinetic theory used for kinetically blocked systems does not apply because Harmonic has a flat frequency profile. Thus the load-bearing assumption—that the N² branch seen at small N for ε=0 is the asymptotic regime—is underdetermined by the present data.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the very long-term collisional relaxation of one-dimensional self-gravitating systems using an exact, collision-driven N-body integrator. It measures the Kolmogorov–Smirnov distance between the ensemble-averaged cumulative mass profile and the finite-N thermodynamic equilibrium of Rybicki (1971), and defines a relaxation time through an ad hoc threshold D0. The main numerical results are: Plummer and Compact (non-degenerate) systems relax on a timescale t_rel ∝ N t_dyn; Harmonic (fully dynamically degenerate) systems relax as t_rel ∝ N^2 t_dyn; and Anharmonic (partially degenerate) systems show a quadratic branch at low N that transitions to linear scaling at larger N, with the transition pushed to larger N as the degeneracy fraction increases. The paper interprets the slow harmonic relaxation as a manifestation of thermodynamic blocking and discusses astrophysical implications for density cores.","tokens_in":18176,"tokens_out":6230,"duration_ms":59910,"significance":"If the central claim holds, the paper is of genuine interest: it identifies a regime in which the standard inhomogeneous Balescu–Lenard scaling t_rel ∝ N t_dyn fails, and it provides a clean numerical benchmark for future kinetic theories of dynamically degenerate self-gravitating systems. The numerical methodology is a major strength: the collision-driven integrator is exact up to round-off, global invariants are conserved to ~1e-23, and the authors use ensemble averaging and a two-threshold consistency check. The public code and detailed appendices also make the measurements reproducible. However, the asymptotic N^2 scaling for the fully degenerate case is inferred from a narrow N window and from the extrapolation of a trend seen in the Anharmonic series; the current evidence is suggestive but not conclusive.","major_comments":[{"comment":"The central claim that Harmonic relaxes as t_rel ∝ N^2 is inferred from a narrow window, 21≤N≤141, after discarding small-N points as ‘polluted by small-N effects’ without a stated criterion. The Anharmonic series in Fig. 3 shows that the same N^2 branch is, for every finite ε>0, a pre-asymptotic regime that bends to N at larger N, with the bend moving to larger N as ε decreases. Since no ε=0 run extends beyond N=141, the data do not distinguish between an N_trans(ε) that diverges as ε→0 and a finite N_trans(0)>141; the headline distinction between Harmonic and non-degenerate systems is therefore an extrapolation. A quantitative fit (log t_rel vs log N with slope uncertainty), an explicit estimate of N_trans(ε), or an analytic argument that N_trans diverges is needed to support the asymptotic claim.","section":"§3.3, Figs. 2–3"},{"comment":"The relaxation time is defined by an ad hoc threshold D0. The check with D0=0.022 (Appendix F) is reassuring, but both thresholds are crossed in the same finite-N window, so it does not test whether the measured exponent is the asymptotic one. The paper also does not report a regression or slope uncertainties; the lines in Figs. 2–3 are visual guides. Since the inferred exponent is the main quantitative result, the crossing-time definition, the treatment of multiple crossings, and the systematic uncertainty from D0 and the N range should be quantified.","section":"§3.2, Eq. (8)"},{"comment":"The statement that small-N measurements are ‘polluted by small-N effects’ is not demonstrated. If the finite-N Rybicki equilibrium (Appendix D) is used exactly, one must specify what the pollution is—for example, the initial KS distance, finite-N corrections to the equilibrium, or remnants of violent relaxation. This matters because excluding those points is part of what determines the fitted exponent.","section":"§3.3, Fig. 2 caption"}],"minor_comments":[{"comment":"Typo: ‘particule’ should be ‘particle’.","section":"§2.1"},{"comment":"Typo: ‘mesure’ should be ‘measure’.","section":"§3"},{"comment":"The x-axis begins at 20, while the text states 21≤N≤141. Please make the plotting range and the stated range consistent.","section":"Figs. 2–3"},{"comment":"The error-bar definition is unclear: after ensemble averaging there is one DKS(t) curve per N, yet the text says t_rel is the mean of all crossing times with min/max error bars. Clarify whether the scatter is over realizations, over sampling intervals, or over threshold crossings of the averaged curve.","section":"§3.2"},{"comment":"The appeal to Deme & Fouvry’s thermodynamic blocking is qualitative and is not used to predict the N^2 scaling. Please mark this as an interpretative hypothesis or provide a quantitative connection.","section":"§3.3 and §4.2"},{"comment":"Define ε precisely in the main text and figure captions: ε is described as the fraction of non-degenerate orbits, while the constant-density core contains a fraction 1−ε. A reader could confuse the two.","section":"Appendix C.5"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed numerical study with a strong exact-integrator methodology. My main concern is that the headline N^2 scaling for Harmonic is overclaimed relative to the evidence: the Anharmonic control series explicitly shows that the quadratic branch can be pre-asymptotic, and no ε=0 run reaches the N values where a bend would be visible. I would not reject the paper, but I would ask the authors to add an explicit slope fit with uncertainties, characterize N_trans(ε) or argue why it diverges, and qualify the abstract/conclusions accordingly. The paper would then be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a serious, clean numerical study of a regime where standard Balescu–Lenard theory is ill-defined, and it reports a genuinely new scaling law. The claim is plausible but the asymptotic statement outruns the data. The paper deserves refereeing.\n\nWhat is actually new: for fully degenerate harmonic 1D systems, all orbits share the same frequency, so the BL resonance condition degenerates and the standard 1/N relaxation theory does not apply. The paper measures relaxation time as a function of N with an exact collision-driven integrator (round-off controlled, invariants conserved to ~1e-23) and finds t_rel ∝ N² t_dyn, versus linear for Plummer and Compact. It also shows that partially degenerate Anharmonic systems show N² at low N and N at high N, with the transition moving to larger N as degeneracy increases. That crossover has not been reported before.\n\nWhat I like: the numerical setup is honest. Ensemble sizes N×N_r ~ 1e5, the threshold choice is tested with D0=0.022 and the same trends appear, and the relaxation is measured against the externally defined finite-N thermodynamic equilibrium from Rybicki, so no constants are fitted to force the scaling. The code is on GitHub. The paper is also explicit that quasilinear expansions fail for flat frequency profiles, and it does not oversell the theoretical interpretation.\n\nSoft spots: the N² result is read off a log-log plot over N=21–141. Small-N points are dropped as “polluted by small-N effects” without a stated criterion, and there is no fit with slope uncertainties. More importantly, the Anharmonic family shows that for every ε>0 the N² branch is a low-N transient that bends to N at larger N. The transition N_trans(ε) grows as ε shrinks. The claim that at ε=0 the N² branch is the true asymptotic law is an extrapolation: it is not measured at large N, and nothing in the paper rules out N_trans remaining finite at ε=0. The cited “thermodynamic blocking” is qualitative and invoked interpretively, not used to predict the exponent. So the headline asymptotic statement is underdetermined; what is solid is the numerical trend over the simulated range and the crossover pattern.\n\nThat said, this is not a fatal flaw. The paper frames itself as a first numerical exploration and explicitly lists larger-N inexact integration as future work. For people working on long-range interacting systems or kinetic theory, this is important enough to take seriously and to probe further. I would send it to peer review, and in revision I would ask for a fitting procedure, a stated criterion for dropping small-N points, and ideally larger-N runs for at least one small ε to see whether the crossover persists at ε=0. I would not yet cite the N² law as established beyond the simulated range.","headline":"A clean numerical study that plausibly identifies N² relaxation for degenerate harmonic 1D systems, but the asymptotic claim outruns the data; deserves refereeing with requests for larger-N checks.","tokens_in":18623,"tokens_out":2241,"would_cite":true,"duration_ms":20849,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"For a 1D self-gravitating system in a harmonic potential, the relaxation time scales as the square of the particle number, t_rel ∝ N², not linearly as Balescu–Lenard theory would predict.","keywords":["one-dimensional self-gravitating systems","relaxation time","dynamical degeneracy","harmonic potential","Balescu-Lenard equation","thermodynamic equilibrium"],"falsifier":"Run the same relaxation measurement for the harmonic system at substantially larger N (e.g., N = 300 to 1000) using a faster, approximate integrator and check whether log(t_rel) versus log(N) remains a clean power law with slope 2. If the local slope flattens toward 1 as N grows, the quadratic law is a finite-N artifact. A second check is to repeat the measurement at a much lower threshold (e.g., D0 = 0.005) with much longer integration times and see whether the same N² slope emerges from the late-time relaxation.","tokens_in":17706,"feed_emoji":"🌌","tokens_out":5587,"duration_ms":45892,"temperature":0.7,"pith_summary":"The paper asks how long a one-dimensional self-gravitating system takes to relax when all its orbits share exactly the same orbital frequency, a regime in which standard kinetic theory is ill-defined. Using an exact collision-driven N-body integrator, it reports that a fully degenerate harmonic system relaxes on a timescale proportional to N² (the square of the particle number), whereas non-degenerate systems relax proportionally to N. It also shows that partially degenerate systems start with the N² behavior at low N and cross over to the linear N scaling once N is large enough, with the crossover happening at larger N when the degenerate fraction is larger. If correct, this means harmonic density cores relax much more slowly than conventional estimates, a result directly relevant to the survival of shallow cores in dwarf galaxies.","feed_headline":"Harmonic self-gravitating systems relax quadratically in particle number","feed_subtitle":"Standard kinetic theory says linear N; harmonic degeneracy slows relaxation by another factor of N.","key_machinery":"The load-bearing object is the dynamically degenerate harmonic potential, a constant-density slab whose quadratic mean-field potential gives every orbit the same frequency Ω, making the Balescu–Lenard resonance condition degenerate. The measurements rely on an exact collision-driven 1D integrator (forces are constant between crossings, collisions are processed with a heap-based event scheduler) and on a relaxation metric: the ensemble-averaged Kolmogorov–Smirnov distance between the evolving cumulative mass profile and the Rybicki finite-N thermodynamic equilibrium. The relaxation time is defined implicitly as the time when this distance crosses an ad hoc threshold D0 = 0.015.","core_discovery":"For a 1D self-gravitating system with a harmonic mean-field potential, every particle has the same orbital frequency Ω, so the resonance condition δ(kΩ − k'Ω) in the Balescu–Lenard equation no longer selects specific resonant pairs and the equation becomes ill-posed. The paper's central numerical discovery is that such fully degenerate systems still relax toward the finite-N thermodynamic equilibrium, but on a timescale t_rel ∝ N² t_dyn, instead of the usual t_rel ∝ N t_dyn found for non-degenerate systems such as Plummer or compact-support clusters. The quadratic scaling is read off from the crossing time of a Kolmogorov–Smirnov distance threshold between the instantaneous and equilibrium c","pith_inferences":["The same quadratic law may apply, with a different prefactor, to 3D harmonic spheres, which would extend the paper's 1D conclusion to the globular-cluster core-stalling problem; this is a conjecture beyond the paper's own scope.","The N² law is measured only up to N = 141; a faster approximate integrator could test whether the slope 2 persists at N ~ 1000, or whether the finite-N window is showing a transient dressed by the threshold crossing.","A time-averaged Hamiltonian or renormalized kinetic description (e.g., a point-vortex analogy) might predict the prefactor of the N² law analytically, turning the empirical scaling into a computed one."],"forward_implications":["Fully degenerate harmonic 1D clusters relax on N² timescales, an order of magnitude slower in N than the linear law predicted by Balescu–Lenard theory.","Partially degenerate clusters exhibit two scaling regimes, with the N²-to-N crossover shifting to higher N as the fraction of degenerate orbits increases.","Compact but non-degenerate systems relax linearly with N, with a prefactor about ten times larger than that of the infinite-support Plummer system.","The long N² relaxation of harmonic cores implies that density cores in galaxy centers can persist much longer than standard relaxation estimates, affecting how long shallow cores survive and how efficiently substructures sink via dynamical friction."],"fun_headline_variants":["Harmonic self-gravity relaxes in N², not N","When all orbits match, relaxation slows to N²","Degenerate orbit frequencies: relaxation switches from N to N²","Self-gravity in 1D harmonic potential: N² relaxation time"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The quadratic scaling is inferred from the crossing time of an adopted threshold D0 = 0.015 in a Kolmogorov–Smirnov distance, measured only for N between 21 and 141 with small-N points excluded as 'polluted by small-N effects'; if these crossing times blend a transient phase with the asymptotic relaxation, the fitted exponent may not be the true large-N law.","fun_headline_variants_meta":{"raw":{"variants":["Harmonic self-gravity relaxes in N², not N","When all orbits match, relaxation slows to N²","Degenerate orbit frequencies: relaxation switches from N to N²","Self-gravity in 1D harmonic potential: N² relaxation time"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000227,"raw_usage":{"total_tokens":1323,"prompt_tokens":777,"completion_tokens":546,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":521,"completion_tokens_details":{"reasoning_tokens":474}},"tokens_in":521,"tokens_out":546,"duration_ms":5069,"temperature":1.0,"reasoning_tokens":474,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T18:22:18.051107+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same relaxation measurement for the harmonic system at substantially larger N (e.g., N = 300 to 1000) using a faster, approximate integrator and check whether log(t_rel) versus log(N) remains a clean power law with slope 2. If the local slope flattens toward 1 as N grows, the quadratic law is a finite-N artifact. A second check is to repeat the measurement at a much lower threshold (e.g., D0 = 0.005) with much longer integration times and see whether the same N² slope emerges from the late-time relaxation.","supporting_citations":[],"review_version":1}