REVIEW 3 major objections 4 minor 2 cited by
Modified Newtonian Dynamics: Observational Successes and Failures
T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read MOND remains viable because one galaxy relation is nearly scatter-free, and its failures may be repairable.
desk verdict A useful, honest review of MOND whose central claim rests on a possibly over-fitted RAR scatter. 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 radial acceleration relation (RAR), an empirical curve linking baryonic acceleration $g_{\rm bar}$ to dynamical acceleration $g_{\rm obs}$, which MOND predicts through its interpolating function $\nu(g_{\rm bar}/a_0)$ in the relation $\vec{g}_{\rm obs} = \nu(g_{\rm bar}/a_0)\,\vec{g}_{\rm bar}$. The function $\nu(x)$ approaches 1 for $x\gg1$ and $x^{-1/2}$ for $x\ll1$, producing the deep-MOND scaling $V^4 = a_0 G M$ and flat rotation curves. The constant $a_0 \approx 1.2\times10^{-10}$ m/s$^2$ marks the acceleration dividing the Newtonian and deep-MOND regimes. The paper also introduces four 'fundamentality criteria' for dynamical relations and uses them to argue that the RAR is not a mere fitting artifact but a fundamental law, which is the load-bearing claim for MOND's ongoing relevance.
What would settle it
Measure the RAR with a sample whose distances and masses are known independently of the rotation-curve fit (for example, galaxies with Cepheid or maser distances): if the intrinsic scatter comes out above roughly 0.1 dex, the paper's central claim fails. Conversely, a definitive wide-binary analysis that finds exactly Newtonian kinematics with no residual systematics would falsify the simple MOND prediction, although modified-inertia or frequency-scale variants could survive.
Extended reading notes
Core claim
The central claim is that MOND remains a live research programme because its most distinctive prediction—the radial acceleration relation—passes every test devised for a fundamental dynamical property. The RAR states that the acceleration $g_{\rm obs}$ inferred from a galaxy's rotation curves is a function only of the acceleration $g_{\rm bar}$ sourced by its visible baryons, with slope 1 in the Newtonian regime and slope 1/2 in the deep-MOND regime, and with negligible scatter. When galaxy nuisance parameters (mass-to-light ratios, distances, inclinations) are fitted simultaneously with the RAR parameters using the SPARC sample, the intrinsic scatter drops to $0.034 \pm 0.002$ dex, and independent weak-lensing measurements extend the relation to extremely low accelerations. The author explicitly identifies the RAR as the best evidence for MOND and shows that it satisfies all four proposed criteria for a fundamental dynamical relation: no residual correlations with other variables, being the tightest projection of dynamical parameter space, accounting for all other dynamical correlations, and uniqueness. The admitted failures at cluster, Solar System, and vertical-disk scales are real, but the paper argues they are not necessarily fatal because two modifications—a frequency scale or modified inertia—could resolve them while preserving the RAR.
Load-bearing premise
The entire case rests on the claim that the RAR's intrinsic scatter is about 0.034 dex; that number comes from a particular simultaneous fit to galaxy distances, inclinations, and mass-to-light ratios, and if those systematics are underestimated the RAR would lose its claimed fundamentality.
Editorial extensions
If this is right
- If the RAR is truly fundamental, then any successful dark-matter theory must reproduce it as a rigid prediction rather than as an emergent accident of galaxy formation; current abundance-matching models do not.
- The Cassini quadrupole constraint and the RAR jointly constrain the shape of the MOND interpolation function in conflict at about 9 sigma, implying either that the interpolation function is system-dependent or that new physics modifies the MOND-to-Newton transition.
- The wide-binary test, run four times, has produced results ranging from Newtonian to MONDian, so a blinded reanalysis on mock data is needed before claiming a decisive refutation.
- Introducing a frequency scale $\omega_0$ that restores Newtonian dynamics for high-frequency motions could explain why Solar System and vertical disk observations look Newtonian while galaxy rotation curves do not.
- If modified inertia is the correct branch of MOND, the RAR could be exact rather than approximate, and the effective interpolation function could vary between environments, reconciling apparently conflicting constraints.
Reading between the lines
- The paper does not say so explicitly, but the near-equality $a_0 \approx cH_0 \approx c^2\Lambda^{1/2}$ suggests a deeper connection between MOND and cosmology; if the frequency-scale idea is correct, $a_0$ might vary with cosmic epoch, a prediction future high-redshift rotation-curve surveys could test.
- A sharper test of the RAR's fundamentality would use galaxies with geometric distance indicators (e.g., Cepheids or water masers); if the intrinsic scatter then exceeds about 0.1 dex, the claimed fundamentality would collapse.
- The two proposed escapes—frequency threshold and modified inertia—predict different behavior for high-frequency tracers in low-acceleration environments, so a single measurement of stellar oscillations or asteroseismology in a dwarf galaxy could separate them.
- The frequency-scale proposal implies that globular clusters or satellite galaxies, which orbit at frequencies between galaxies and wide binaries, should show a partial return to Newtonian dynamics; no current dataset resolves this prediction.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper is a review-style assessment of MOND's observational record. It argues that MOND has three main successes: the baryonic Tully-Fisher relation, galaxy rotation curves, and especially the radial acceleration relation (RAR), which the author claims is extremely tight (intrinsic scatter 0.034±0.002 dex), satisfies four 'fundamentality' criteria, and is difficult to reproduce in ΛCDM. The paper then documents MOND's failures: incomplete cluster mass discrepancies, the Bullet Cluster, the Solar System quadrupole constraint (a reported 9σ tension), Milky Way vertical kinematics, and the contradictory wide-binary test results. It concludes that MOND retains ongoing relevance and suggests two speculative avenues for reconciliation: introducing a frequency scale ω0, or pursuing modified-inertia formulations.
Significance. The article is clearly written, well organised, and unusually candid: it states both MOND's strengths and weaknesses, explicitly labels some claims as the author's opinion, and calls for blinded wide-binary-test analyses. If the RAR tightness and fundamentality results are correct, the paper makes a nontrivial case that MOND remains an empirically motivated alternative. The main weakness is that the two load-bearing pieces of evidence—the RAR intrinsic scatter and the four-criteria fundamentality test—come from the author's own previous analyses, with no independent validation presented here. The proposed remedies (frequency scale, modified inertia) are clearly speculative. The paper's value is therefore as a perspective that frames MOND's current standing, not as an independent test of the theory.
major comments (3)
- [Section 2 (The Good)] The central claim that the RAR has an 'extraordinarily small' intrinsic scatter of 0.034±0.002 dex is load-bearing for the paper's conclusion, but it rests on a single reanalysis (Ref. 20) in which galaxy distances, inclinations, and mass-to-light ratios are simultaneously optimised with the RAR parameters. The paper itself reports 0.082±0.003 dex under the fiducial SPARC error model, and no mock-injection or cross-validation test is presented to show that the joint re-fitting recovers a known input scatter rather than absorbing real astrophysical scatter. The independent MIGHTEE estimate of 0.045±0.022 dex is also consistent with the larger 0.082 dex value, so it does not corroborate the low value. The author should either provide such a validation, cite one, or weaken the 'extraordinarily small' claim accordingly.
- [Section 2 (The Good)] The statement that 'Extraordinarily, the RAR satisfies all four' fundamentality criteria is taken from Ref. 4, which uses the same re-fitted SPARC data and assumes the RAR as the relation of interest. This is not an independent check, and the paper should state this dependence explicitly or qualify the claim. As it stands, the fundamentality conclusion inherits the same systematic uncertainties as the scatter measurement, and the text gives the reader no way to assess how much of the result is built into the construction of the test.
- [Section 5 (Summary and Conclusion)] The paper's overall conclusion that MOND has 'ongoing relevance' depends critically on the RAR evidence, yet the two proposed ways of circumventing MOND's failures—a frequency scale ω0 and modified inertia—are not developed into concrete, testable models. They are explicitly speculative, which is acceptable in a review, but the concluding section risks presenting them as part of the empirical case for MOND. The author should more sharply separate the observational evidence from the conjectural remedies, and should explicitly state that without independent validation of the RAR scatter the 'ongoing relevance' conclusion is conditional.
minor comments (4)
- [Section 4 (The Ugly)] The definition of αgrav appears with a typographical issue (the text reads 'αgrav ≡ √η−10.193'); the intended expression should be typeset cleanly.
- [Abstract / running title] The running title and first page header contain 'F ailures' with a space; this should be corrected to 'Failures'.
- [Section 4 (The Ugly)] The observation that 'almost any form of unmodelled contamination will enhance the dynamics and hence masquerade as a MONDian signal' is an important caveat for interpreting the wide-binary results; it would be worth stating more prominently, perhaps in the abstract or conclusion, because it affects how much weight the reader should place on any single WBT outcome.
- [Section 2 (The Good)] The claim that 'it seems quite unlikely' for ΛCDM to produce the RAR's remarkable properties is asserted rather than quantified. A short discussion of the specific difficulties in ΛCDM models, or a pointer to papers that attempt to quantify the fine-tuning, would help the reader evaluate the strength of this claim.
Circularity Check
No significant circularity: MOND's a0 is explicitly empirical and the RAR is measured from external data; only minor self-citation in the 'four criteria' claim.
full rationale
The paper is a review rather than a derivation, and its central quantitative claims are benchmarked against external observations (SPARC, MIGHTEE, Cassini, Gaia). MOND's a0 is explicitly stated to be empirical ('the theory does not provide any way of calculating a0 and hence it must be found empirically'), and the interpolating function is acknowledged to be 'little more than a fitting function'. The RAR is therefore not a free prediction in the sense of being derived from independent parameters; its asymptotic slopes are chosen to reproduce flat rotation curves and Newtonian behaviour. However, the paper's more specific claims, namely the small intrinsic scatter and universality of the RAR across galaxies, are not equivalent to a fitted input by construction. The 0.034 dex scatter is obtained from a joint fit of galaxy nuisance parameters reported in the author's Ref. 20; this is a potential systematic bias (nuisance parameters can absorb real scatter) but not a definitional circularity, and the paper also quotes an independent MIGHTEE measurement. The 'four criteria' argument for the RAR's fundamentality is taken from the author's own prior work (Ref. 4), which is a self-citation, but that work is an empirical analysis of external SPARC data and is not used as an unexamined theorem to forbid alternatives. No equation in the paper is shown to equal its own input by construction, and no fitted parameter is renamed as a prediction without a test against external data. The main circularity-adjacent weakness is the reliance on the author's own joint-fit value of the RAR scatter, which the paper itself flags as dependent on the error model (0.082 vs 0.034 dex). This warrants a low score, not a finding of substantive circularity.
Assumptions & free parameters
free parameters (1)
- a0 (MOND acceleration constant) =
~1.2e-10 m/s^2
assumptions (3)
- domain assumption Equation (1) with the interpolating function limits is an adequate effective description of non-relativistic dynamics.
- domain assumption The external field effect is correctly modeled in the RAR analyses and in the Cassini-Q2 comparison.
- domain assumption The SPARC and KIDS measurements accurately trace the radial acceleration relation.
invented entities (1)
-
Frequency scale omega_0
Cite this review
Pith. "Pith review of Modified Newtonian Dynamics: Observational Successes and Failures." pith.science (2026). https://pith.science/paper/ST2CEYY4
@misc{pith2026250521638,
author = {Pith},
title = {Pith review of: Modified Newtonian Dynamics: Observational Successes and Failures},
year = {2026},
howpublished = {\url{https://pith.science/paper/ST2CEYY4}},
note = {Machine review of arXiv:2505.21638}
}
read the original abstract
Modified Newtonian Dynamics (MOND) is an alternative to the dark matter hypothesis that attempts to explain the "missing gravity" problem in astrophysics and cosmology through a modification to objects' dynamics. Since its conception in 1983, MOND has had a chequered history. Some phenomena difficult to understand in standard cosmology MOND explains remarkably well, most notably galaxies' radial dynamics encapsulated in the Radial Acceleration Relation. But for others it falls flat -- mass discrepancies in clusters are not fully accounted for, the Solar System imposes a constraint on the shape of the MOND modification seemingly incompatible with that from galaxies, and non-radial motions are poorly predicted. An experiment that promised to be decisive -- the wide binary test -- has produced mainly confusion. This article summarises the good, the bad and the ugly of MOND's observational existence. I argue that despite its imperfections it does possess ongoing relevance: there may yet be crucial insight to be gleaned from it.
Forward citations
Cited by 2 Pith papers
-
Evolution of superthin galaxies under Milgromian dynamics
In isolated MOND simulations, the UGC 7321 model keeps h_z/R_D mostly below 0.1 for 5 Gyr, and higher-MOND-depth (more diffuse) models stay thinner than low-depth models.
-
(Exhaustive) Symbolic Regression and model selection by minimum description length
Exhaustive search over simple functions ranked by description length beats the Friedmann equation, MOND, and common inflaton potentials on astrophysical datasets.
Reference graph
Works this paper leans on
- [1]
- [2]
- [3]
- [4]
-
[5]
MOND as manifestation of modified inertia
M. Milgrom,arXiv2310.14334, (2023)
work page Pith review arXiv 2023
- [6]
- [7]
-
[8]
R. H. Sanders and S. S. McGaugh,ARA&A40, 263 (2002)
work page 2002
Show all 44 references
-
[9]
Famaey and S
B. Famaey and S. S. McGaugh,Liv. Rev. Relativity15, 10 (2012)
2012
-
[10]
Banik, H
I. Banik, H. Zhao,Symmetry14, 1331 (2022)
2022
-
[11]
Skordis, T
C. Skordis, T. Z lo´ snik,Phys. Rev. Lett.127, 161302 (2021)
2021
-
[12]
S. S. McGaugh et al.,ApJL533, L99 (2000)
2000
-
[13]
Desmond,MNRAS472, L35 (2017)
H. Desmond,MNRAS472, L35 (2017)
2017
-
[14]
Desmond et al.,MNRAS484, 239 (2019)
H. Desmond et al.,MNRAS484, 239 (2019)
2019
-
[15]
R. H. Sanders,A&ARv2, 1 (1990)
1990
-
[16]
Lelli et al.,ApJ836, 152 (2017)
F. Lelli et al.,ApJ836, 152 (2017)
2017
-
[17]
Lelli et al.,AJ152, 157 (2016)
F. Lelli et al.,AJ152, 157 (2016)
2016
-
[18]
Desmond et al.,MNRAS521, 1817 (2023)
H. Desmond et al.,MNRAS521, 1817 (2023)
2023
-
[19]
Desmond,MNRAS464, 4160 (2017)
H. Desmond,MNRAS464, 4160 (2017)
2017
-
[20]
Desmond,MNRAS526, 3342 (2023)
H. Desmond,MNRAS526, 3342 (2023)
2023
- [21]
-
[22]
M. M. Brouwer et al.,A&A650, A113 (2021)
2021
-
[23]
Mistele et al.,JCAP2024, 020 (2024)
T. Mistele et al.,JCAP2024, 020 (2024)
2024
-
[24]
Kroupa,PASA29, 395 (2012)
P. Kroupa,PASA29, 395 (2012)
2012
-
[25]
Roshan et al.,MNRAS508, 926 (2021)
M. Roshan et al.,MNRAS508, 926 (2021)
2021
-
[26]
Chae et al.,ApJ921, 104 (2021)
K.-H. Chae et al.,ApJ921, 104 (2021)
2021
-
[27]
H. Zhao, B. Famaey,Phys. Rev. D86, 067301 (2012)
2012
-
[28]
Clowe et al.,ApJL648, L109 (2006)
D. Clowe et al.,ApJL648, L109 (2006)
2006
-
[29]
Li et al.,A&A677, A24 (2023)
P. Li et al.,A&A677, A24 (2023)
2023
-
[30]
Freundlich et al.,A&A658, A26 (2022)
J. Freundlich et al.,A&A658, A26 (2022)
2022
-
[31]
Lisanti et al.,Phys
M. Lisanti et al.,Phys. Dark Univ.39, 101140 (2023)
2023
-
[32]
Hees et al.,Phys
A. Hees et al.,Phys. Rev. D89, 102002 (2014)
2014
-
[33]
Hees et al.,MNRAS455, 449 (2016)
A. Hees et al.,MNRAS455, 449 (2016)
2016
-
[34]
Desmond et al.,MNRAS530, 1781 (2024)
H. Desmond et al.,MNRAS530, 1781 (2024)
2024
-
[35]
Vokrouhlick´ y et al.,ApJ968, 47 (2024)
D. Vokrouhlick´ y et al.,ApJ968, 47 (2024)
2024
-
[36]
Banik et al.,MNRAS527, 4573 (2024)
I. Banik et al.,MNRAS527, 4573 (2024)
2024
-
[37]
Pittordis and W
C. Pittordis and W. Sutherland,Open J. Astrophys.6, 4 (2023)
2023
- [38]
-
[39]
Hernandez et al.,EPJ C72, 1884 (2012)
X. Hernandez et al.,EPJ C72, 1884 (2012)
2012
-
[40]
Chae,ApJ952, 128 (2023)
K.-H. Chae,ApJ952, 128 (2023)
2023
-
[41]
Chae,ApJ972, 186 (2024)
K.-H. Chae,ApJ972, 186 (2024)
2024
-
[42]
Milgrom,Phys
M. Milgrom,Phys. Rev. D108, 084005 (2023)
2023
-
[43]
D. W. Sciama,MNRAS113, 34 (1953)
1953
-
[44]
Uruena Palomo,Int
M. Uruena Palomo,Int. J. Theor. Phys.63, 271 (2024)
2024
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.