{"id":"33198463-7f9d-4d85-b673-8389a7e86410","arxiv_id":"2412.03658","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":15,"one_line_summary":"For NGC1052-DF44, MOND and RGGR modified gravity fits with free orbital anisotropy match the observed velocity dispersion as well as an NFW dark matter halo.","lead":"This paper fits the observed motions of star clusters in the ultra-diffuse galaxy NGC1052-DF44 using three modified gravity models and compares them with a standard dark matter model. It finds that MOND and a running gravitational constant model fit the data about as well as a dark matter halo, once the shapes of the stellar orbits are allowed to vary.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim that MOND and RGGR are competitive with NFW rests on an isolated-galaxy approximation that omits the Coma external field; prior EFE-inclusive MOND modeling of the same galaxy fails, so the fit may lose physical basis once EFE is included.","rationale":"I agree with the reader's weakest-assumption identification: the isolated-galaxy treatment of the external field is the single most load-bearing premise. The paper is transparent about this limitation, but transparency does not make the premise secure. Reference [36] directly reports that EFE-inclusive MOND fails for DF44, and the paper's own conclusion acknowledges the omission. Since the abstract does not carry the 'isolated' qualifier, the headline claim is broader than what the analysis supports. The Jeans and MOND/gravity-model math appears internally consistent, but there is no machine-checked proof, code, or data release, so independent support is absent. Secondary reproducibility gaps strengthen the conditional reading: the NFW mass differs between the Fig.1 caption (0.70e11 M_sun) and the text (3.98e10 M_sun), the number of observed data points N is not reported despite its role in the BIC formula, and the MCMC details are incomplete. These do not by themselves overturn the fits, but they reinforce that the verdict should remain CONDITIONAL. The recommended action is to keep the current conditional verdict and require an explicit qualifier in the abstract plus an EFE sensitivity test for MOND and RGGR.","tokens_in":17154,"tokens_out":7708,"duration_ms":79729,"concrete_test":"Re-fit DF44 with the EFE-inclusive MOND model of Freundlich et al. (2021) using the same observed bins, Sersic mass model, and priors on gamma* and xi; determine whether MOND remains within ΔBIC<2 of NFW. For RGGR, repeat with the Coma potential added to phi_N in Eq.19 and the same anisotropy scenarios. If the EFE-inclusive MOND/RGGR fits degrade to chi^2_red>2 or ΔBIC>6 relative to NFW, the central claim fails outside the isolated approximation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing premise is that DF44 can be treated as isolated from the Coma cluster. MOND's Eq.13 and RGGR's Eq.19 use only the galaxy's own potential, and the paper states in the conclusion: \"we avoid the complications of EFE.\" But DF44 sits in Coma, and reference [36] reports that including the Coma external field in MOND fails to reproduce DF44's observed velocity dispersion; the only proposed rescue is a non-equilibrium infall scenario [41], which is not modeled here. If the external field is not effectively suppressed, the MOND fit is not a physically valid MOND prediction for DF44, and the abstract's statement that MOND and RGGR are \"competitive\" with NFW overstates the evidence. The same SEP-type external-potential issue applies to RGGR, as the paper itself notes, yet no EFE-inclusive RGGR model is given. This is not an internal inconsistency, but it makes the headline claim conditional on an external-field suppression that is untested in this work.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper performs a kinematic analysis of the ultra-diffuse galaxy NGC1052-DF44 under four dynamical models: a Navarro-Frenk-White (NFW) dark matter halo, MOND, a generic Yukawa-like f(R) gravity, and RGGR. For each model, the line-of-sight velocity dispersion is computed with the Jeans equation under three anisotropy prescriptions (isotropic, constant, and Osipkov-Merritt), with parameters constrained by MCMC and models compared by BIC. The central findings are that all three gravity models can fit the observed velocity dispersion in at least one anisotropy scenario, that MOND and RGGR are statistically competitive with NFW while f(R) is less favored, and that constant tangential anisotropy is competitive with isotropy while the Osipkov-Merritt profile is strongly disfavored.","tokens_in":17439,"tokens_out":5594,"duration_ms":55513,"significance":"If the central claim holds, the paper would provide a useful systematic comparison of modified-gravity models against dark-matter halos in a single well-studied UDG, extending earlier isotropic analyses to anisotropic velocity dispersion. The paper's strengths are the use of a common anisotropic Jeans framework, the simultaneous treatment of three gravity models and an NFW halo, and the use of BIC rather than only chi-squared for model comparison. However, the headline 'competitive' claim is conditional on an isolated-galaxy approximation that is explicitly acknowledged but not tested, and the statistical reporting is incomplete (no parameter uncertainties, no stated number of data points, and a non-standard BIC penalty). These issues currently limit the strength of the conclusions.","major_comments":[{"comment":"The central claim that MOND and RGGR are competitive with NFW rests on the isolated-galaxy approximation. Equation (13) for MOND and Eq. (19) for RGGR use only the internal Newtonian potential of the galaxy, and Sec. VI states 'we avoid the complications of EFE.' However, DF44 is embedded in the Coma cluster, and reference [36] (cited in the introduction) found that including the Coma external field in MOND fails to reproduce the observed velocity dispersion, with only a non-equilibrium infall scenario [41] as a possible rescue that is not modeled here. The abstract's statement that 'only MOND and RGGR remain competitive with NFW DM' is therefore not a physical MOND/RGGR prediction for DF44 unless the external field is effectively suppressed. The paper should either include an EFE-inclusive analysis (at least for MOND, where the framework is available) or explicitly and prominently qualify the headline conclusion to the isolated approximation.","section":"Sec. III A, III C, VI"},{"comment":"The text states that the authors 'probe the kinematics for two RGGR frameworks, i.e., an isolated scenario and under the influence of external effects, as discussed below,' but only the isolated scenario is presented. No external-field RGGR model or associated results appear in the paper. This promised analysis is directly relevant to the EFE concern raised in the previous comment; the paper should either provide the external-field RGGR results or remove the claim that two frameworks are studied.","section":"Sec. V C"},{"comment":"The BIC formula as written is non-standard: BIC = -2 log L + 2k log(n) differs from the Schwarz criterion, BIC = -2 log L + k log N. The number of data points N (called n in Eq. (21)) is never given anywhere in the paper. Since the BIC differences and the interpretation thresholds in Sec. IV (e.g., ΔBIC < 2, 2-6, >6) depend on the penalty term, the model-comparison results in Tables I-IV cannot be verified as presented. Please correct the formula, define and report N, and recompute the BIC values and thresholds accordingly.","section":"Eq. (21), Tables I-IV"},{"comment":"The best-fit parameters are reported without any uncertainties, despite the use of an MCMC sampler. Without posterior intervals (e.g., 16th-84th percentiles), it is impossible to assess whether the parameters are well constrained, which is especially important for the f(R) parameters that lie close to the prior boundary (e.g., δ = -0.90 in Table II) and for the 'inconclusive' ΔBIC differences of about 2 between isotropic and constant anisotropy. Please report parameter uncertainties and state the priors and their boundaries explicitly.","section":"Tables I-III"}],"minor_comments":[{"comment":"The text refers to a 'green dashed line' for the NFW case, but Fig. 1 and its caption show a green solid line; please correct the inconsistency.","section":"Sec. II, Fig. 1"},{"comment":"The anisotropy profile is repeatedly called 'Osikpov-Merritt'; the correct spelling is 'Osipkov-Merritt' (also in the figure captions and tables).","section":"Throughout"},{"comment":"Several equations in Sec. II appear to have lost the radial coordinate symbol in the typeset version (e.g., the density argument in Eq. (1) and the mass integral in Eq. (3)). Please ensure all radial variables are shown consistently.","section":"Eqs. (1), (3), (4)"},{"comment":"The BIC interpretation thresholds (ΔBIC < 2, 2-6, >6) are stated without a specific citation; please add a reference, and ensure the thresholds are consistent with the corrected BIC definition.","section":"Sec. IV"}],"recommendation":"major_revision","confidential_remarks":"The paper is a useful fitting exercise, but the main scientific claim is currently conditional on the isolated-galaxy approximation, and the statistical reporting has gaps (BIC penalty, missing N and uncertainties). These are addressable within the scope of a revision, so I recommend major revision rather than rejection. If the authors can provide an EFE-inclusive treatment or substantially weaken the abstract/conclusions, the paper could become acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You can skip the exotic-gravity packaging; the real result is a model comparison with anisotropy added. The paper fits DF44's velocity dispersion with MOND, f(R), and RGGR under three Jeans anisotropy assumptions, and finds that constant tangential anisotropy is as good as isotropy while the Osipkov-Merritt radial profile is strongly disfavored. That is a legitimate new axis for a single UDG, even though the machinery is standard and no new formalism is proposed.\n\nWhat the paper does well: the fits are internally consistent, the BIC comparison is transparent, and it does not oversell f(R) — that model lands worse than NFW, which is the honest outcome. The anisotropy conclusion is robust across all three gravity models, which is a genuinely useful finding for future multi-galaxy work.\n\nThe soft spot is exactly where the reader and stress-test put it. The MOND and RGGR fits assume DF44 is isolated from Coma. The paper is explicit about this in the conclusion: \"we avoid the complications of EFE.\" But reference [36] found that including the Coma external field in MOND fails to reproduce DF44's observed velocity dispersion, and the only proposed rescue is an out-of-equilibrium infall scenario that is not modeled here. So the abstract's claim that MOND and RGGR remain competitive with NFW is physically conditional on an EFE suppression mechanism that is untested in this work. The RGGR case has the same SEP-type exterior-potential issue, which the paper itself notes. This is not an internal inconsistency, but it does mean the headline result is weaker than it looks.\n\nMinor but real issues: best-fit tables report no parameter uncertainties, the number of data points entering the BIC is not given, and no code or data are released. The BIC comparison between models with different parameter counts is only meaningful if N is known; with a small dataset, a ΔBIC of ~6 is not a strong preference. These are fixable in revision.\n\nThis is a useful fitting study for the modified-gravity-versus-DM debate on UDGs. It does not resolve the question, but it shows that the exclusion of MOND for DF44 is not as clean as isotropic fits suggested — provided EFE is ignorable. That is a conditional result, and the paper frames it honestly.\n\nRecommendation: send it to peer review. It deserves a serious referee, and the referee should push on the EFE issue, require error bars, and ask for the data and number of data points. With those changes it could be a solid contribution to the UDG kinematics literature.","headline":"A solid, honest fitting study that adds anisotropy to the DF44 modified-gravity comparison; the MOND/RGGR competitiveness claim is real but conditional on ignoring the Coma external field, which prior work found fatal for MOND.","tokens_in":17980,"tokens_out":1472,"would_cite":false,"duration_ms":16112,"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":"The paper claims that MOND and RGGR can match the line-of-sight velocity dispersion of the dark-matter-dominated ultra-diffuse galaxy DF44 as well as an NFW dark matter halo does, once orbital anisotropy is allowed; a generic f(R) model…","keywords":["ultra-diffuse galaxies","NGC1052-DF44","line-of-sight velocity dispersion","orbital anisotropy","modified Newtonian dynamics (MOND)","f(R) gravity","RGGR running gravitational constant","NFW dark matter halo"],"falsifier":"Recompute the best-fit MOND and RGGR models with the Coma cluster external field included; if the predicted line-of-sight velocity dispersion at 5.1 kpc falls below the observed about 41 km/s, as the earlier external-field MOND calculation quoted in the paper found, the competitive-fit claim fails. A complementary check is DF44's three-dimensional position and velocity relative to Coma: if the galaxy is not on a first infall, the proposed suppression of the external field cannot rescue the isolated fits.","tokens_in":16936,"feed_emoji":"🌌","tokens_out":15287,"duration_ms":139948,"temperature":0.7,"pith_summary":"NGC1052-DF44 is an ultra-diffuse galaxy, a large but very faint galaxy, whose stellar motions are so fast that ordinary stars alone cannot bind it; this is usually read as a dark matter signal. The paper asks whether three gravity models that avoid dark matter can produce the same observed line-of-sight velocity dispersion. It concludes that MOND and RGGR fit the data as well as the standard cuspy NFW dark matter halo profile does, while a generic f(R) model fits but is statistically less competitive. It also finds that allowing the stellar orbits to have a constant tangential anisotropy fits as well as the common isotropic assumption in all models. If this result holds, DF44 does not cleanly discriminate dark matter from modified gravity, and orbital anisotropy is a necessary part of such tests.","feed_headline":"Modified gravity rivals dark matter in matching DF44's motion","feed_subtitle":"MOND and RGGR reproduce the galaxy's observed velocity dispersion as well as a dark matter halo does, once orbital anisotropy is allowed.","key_machinery":"The load-bearing machinery is the spherical Jeans equation, converted into a fast analytic integral for the projected line-of-sight velocity dispersion $\\sigma_{\\rm LOS}^2(r)$ with a kernel $K$ that encodes the anisotropy parameter $\\xi$: zero for isotropic motion, a fitted constant for the tangential/radial case, and the Osipkov-Merritt scale radius for the radial profile. The modified gravity models enter through an effective mass function in that integral: MOND through its interpolation function and the fixed acceleration scale $a_0$, $f(R)$ through a Yukawa correction to the Newtonian potential with coupling $\\delta$ and scale $\\lambda$, and RGGR through the running of $G$ represented by the parameter $\\bar\\nu$. The stellar mass is a de-projected Sersic profile scaled by a fitted mass-to-light ratio $\\gamma_*$, and the parameters are scanned with a Markov chain Monte Carlo sampler and ranked with the Bayesian information criterion.","core_discovery":"The paper's central claim is that, in an isolated, spherically symmetric Jeans treatment of NGC1052-DF44, dark matter is not uniquely required: MOND with its fixed acceleration scale ($a_0 = 1.14 \\times 10^{-8}\\,\\text{cm/s}^2$) and RGGR with a slowly running gravitational coupling ($\\bar\\nu \\approx 2.5 \\times 10^{-8}$) both reproduce the observed line-of-sight velocity dispersion as well as the standard cuspy NFW dark matter halo does ($\\chi^2_{\\rm red} = 0.39$ and $0.45$ versus $0.75$). A generic $f(R)$ model with a Yukawa correction also fits the data, but its Bayesian information criterion is worse than that of NFW, so the authors count it as not competitive. In every model, a constant tangential orbital anisotropy fits as well as, or slightly better than, the usual isotropic assumption, while the radially dependent Osipkov-Merritt profile is strongly disfavored. The conclusion is that, with anisotropy allowed, DF44 does not separate dark matter from modified gravity.","pith_inferences":["If the same anisotropy-gravity degeneracy holds for other ultra-diffuse galaxies, re-fitting existing isotropic velocity-dispersion datasets with a constant anisotropy parameter is a direct test of how much of the reported dark matter signal is actually orbital structure.","The external-field issue that already undermines the Coma MOND fit applies in principle to the RGGR and f(R) fits here; recomputing both models with the Coma cluster's gravitational field included would show whether their competitiveness survives.","Because MOND and RGGR scale differently with acceleration and gravitational potential energy, a sample of ultra-diffuse galaxies at different masses and cluster-centric distances could separate the two models, something one galaxy cannot do.","The fits prefer tangentially biased globular-cluster orbits, a preference that is in principle checkable with tangential proper motions or higher-order velocity moments."],"forward_implications":["MOND and RGGR each match the observed line-of-sight velocity dispersion of DF44 without a dark matter component, so this ultra-diffuse galaxy is not a decisive dark matter detection once modified gravity is allowed.","A constant negative (tangentially biased) anisotropy fits as well as isotropy in every model tested, so velocity-dispersion data alone cannot fix both the gravity law and the orbital structure.","The Osipkov-Merritt radial anisotropy profile is strongly disfavored by Bayesian information criterion in all three gravity models, ruling out a simple radial-orbit alternative for DF44.","The generic f(R) model fits the data but is statistically less competitive than NFW, so this particular Yukawa-type modification is constrained by DF44.","The inferred mass-to-light ratio shifts when anisotropy is introduced, so stellar masses and dark matter fractions derived from isotropic Jeans modeling carry a systematic uncertainty."],"supporting_citations":[{"why":"It supplies the DF44 photometric parameters (Sersic index 0.94, effective radius 4.7 kpc, total luminosity) and the observational target of the study.","marker":"[24]"},{"why":"It provides the globular-cluster line-of-sight velocity dispersion measurements, including the about 41 km/s point at 5.1 kpc, that the fits must match.","marker":"[19]"},{"why":"It gives the NFW dark matter halo modeling approach and the baseline fit the modified gravity models are compared against.","marker":"[35]"},{"why":"It supplies the reduced analytic line-of-sight velocity dispersion kernel used in the paper's central computation.","marker":"[48]"},{"why":"It provides the kernel functions for the constant and Osipkov-Merritt anisotropy profiles used in the analysis.","marker":"[52]"},{"why":"It shows that adding the Coma cluster external field in MOND fails to reproduce DF44, the limitation the paper explicitly sets aside.","marker":"[36]"},{"why":"It is the earlier MOND fit to DF44 that this work extends by adding anisotropic velocity dispersions.","marker":"[32]"},{"why":"It defines the renormalization-group corrected gravity model with a running gravitational coupling and its phenomenological parameter.","marker":"[45]"},{"why":"It supplies the weak-field Yukawa potential for the f(R) expansion that is fitted to DF44 here.","marker":"[57]"}],"fun_headline_variants":["MOND and RGGR rival dark matter in DF44's motion","DF44's velocity dispersion fits MOND and RGGR as well","Modified gravity matches dark matter in DF44's kinematics","Anisotropy lets MOND and RGGR compete with dark matter","DF44 doesn't require dark matter: MOND and RGGR suffice"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The gravity-model fits assume DF44 is isolated, so they leave out the external gravitational field of the Coma cluster in which the galaxy sits; if that external field acts on the galaxy in the usual way, the MOND and RGGR fits lose their physical basis.","fun_headline_variants_meta":{"raw":{"variants":["MOND and RGGR rival dark matter in DF44's motion","DF44's velocity dispersion fits MOND and RGGR as well","Modified gravity matches dark matter in DF44's kinematics","Anisotropy lets MOND and RGGR compete with dark matter","DF44 doesn't require dark matter: MOND and RGGR suffice"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000253,"raw_usage":{"total_tokens":1588,"prompt_tokens":995,"completion_tokens":593,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":611,"completion_tokens_details":{"reasoning_tokens":504}},"tokens_in":611,"tokens_out":593,"duration_ms":6389,"temperature":1.0,"reasoning_tokens":504,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:14:19.544748+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the best-fit MOND and RGGR models with the Coma cluster external field included; if the predicted line-of-sight velocity dispersion at 5.1 kpc falls below the observed about 41 km/s, as the earlier external-field MOND calculation quoted in the paper found, the competitive-fit claim fails. A complementary check is DF44's three-dimensional position and velocity relative to Coma: if the galaxy is not on a first infall, the proposed suppression of the external field cannot rescue the isolated fits.","supporting_citations":[{"cited_title":"Yozin and K","cited_arxiv_id":null,"evidence_quote":"It provides the globular-cluster line-of-sight velocity dispersion measurements, including the about 41 km/s point at 5.1 kpc, that the fits must match."},{"cited_title":"Elliptical galaxies kinematics within general relativity with renormalization group effects","cited_arxiv_id":"1203.2286","evidence_quote":"It supplies the reduced analytic line-of-sight velocity dispersion kernel used in the paper's central computation."},{"cited_title":"Prugniel and F","cited_arxiv_id":null,"evidence_quote":"It provides the kernel functions for the constant and Osipkov-Merritt anisotropy profiles used in the analysis."},{"cited_title":"Enigmatic Velocity Dispersions of Ultra-Diffuse Galaxies in Light of Modified Gravity Theories and Radial Acceleration Relation","cited_arxiv_id":"1910.09726","evidence_quote":"It is the earlier MOND fit to DF44 that this work extends by adding anisotropic velocity dispersions."}],"review_version":1}