{"id":"1eac8db5-733c-4862-bc2c-5c280fe77a3a","arxiv_id":"2506.00860","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A cosmographic fit of a Gödel-type rotating universe to supernova data yields a mild hint of cosmic rotation, with Ω0≈0.3 at z≤0.2, but the evidence is not strong.","lead":"The study finds a weak but consistent preference for rotation in a model of the expanding universe, based on the Pantheon+ supernova catalog. If real, such rotation would contradict the standard picture of a universe that looks the same in every direction.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"First-order cosmographic truncation can bias Ω0 at Z≤0.5; no truncation error estimate is given and the signal peaks in the Z≤0.2 bin.","rationale":"The reader's conditional verdict already hinges on the unquantified first-order truncation; our stress-test agrees that this is the most load-bearing assumption. The concern is concrete: the rotation dipole term and the neglected O(Z^2) term are of similar amplitude at Z=0.5, and the signal peaks in the Z≤0.2 bin where the higher-order term first matters. The proposed test—extending Eq. (9) to second order and refitting—would settle whether Ω0 survives. Because the paper does not provide chains or this extension, the conditional verdict is appropriate and no change is needed.","tokens_in":17148,"tokens_out":6551,"duration_ms":63090,"concrete_test":"Add the O(Z^2) term to Eq. (9) with a free jerk-like coefficient (or the full second-order Kristian-Sachs coefficient) and re-run the same five redshift-bin fits on Pantheon+. If Ω0 for Z≤0.2 shifts toward zero or its uncertainty grows so that zero is within 1σ, the rotation preference is explained by truncation bias. As a complement, compute the exact luminosity distance from the null geodesic equations of the metric (2) and compare the first-order approximation at Z=0.2 and Z=0.5; if the magnitude difference exceeds ~0.05 mag, the truncation is unsafe for these cuts.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equation (9) is truncated at first order in Z, yet the fits extend to Z=0.5. In the standard cosmographic series the neglected O(Z^2) term is sizable at these redshifts; for reference, in ΛCDM-like parameters the second-order contribution to the distance modulus is roughly −0.1 to −0.3 mag at Z=0.5, comparable to the rotation dipole term (5/2)(log10 e)(Ω0/h0) Z, which is ≈0.2 mag for Ω0/h0≈0.4. Since Ω0 is consistent with zero in the Z≤0.1 bin (0.24±0.46) and peaks only in the Z≤0.2 bin (0.29+0.21−0.15), the apparent detection coincides with the redshift range where truncation effects become non-negligible. The nested redshift bins also make the 'consistent axis' appear more significant than it is. Without an estimate of the O(Z^2) contribution, the inferred Ω0 cannot be trusted as clean evidence for rotation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper constrains a Gödel-type rotating universe model using the Pantheon+ Type Ia supernova sample. The authors use a magnitude-redshift relation derived via the Kristian-Sachs formalism (Eq. 9), truncated at first order in redshift Z, and fit model parameters with MCMC in five cumulative redshift cutoffs from Z≤0.1 to Z≤0.5. They report a stable h0≈0.73, a deceleration parameter q0 near zero, a rotation parameter Ω0 peaking at 0.29^{+0.21}_{-0.15} in the Z≤0.2 bin, a preferred anisotropy axis around (243°,−49°), and an AIC-based preference for the rotating model over flat ΛCDM. The abstract concludes a 'mild but consistent preference for cosmic rotation'.","tokens_in":17320,"tokens_out":6516,"duration_ms":59932,"significance":"If the rotation signal were robust, it would be a noteworthy indication of anisotropy in the late-time expansion history. The paper benefits from using the public Pantheon+ data with its full covariance matrix, including Cepheid anchors that break the h0–M0 degeneracy, and from following a standard MCMC likelihood procedure with a clear model-comparison framework. The main weakness is that the statistical evidence is marginal: the most favorable bin shows Ω0 only about 2σ from zero, and the analysis relies on a first-order cosmographic expansion at redshifts where neglected higher-order terms are expected to be appreciable. The significance of the claimed preference is also overstated relative to the reported AIC values. The paper is readable and reproduces standard pipelines, but the central claim is not yet firmly supported.","major_comments":[{"comment":"The magnitude-redshift relation is truncated to first order in Z, yet the fits extend to Z=0.5. In a standard cosmographic expansion, the neglected O(Z^2) contribution to the distance modulus is non-negligible at Z=0.5, typically of order −0.1 to −0.3 mag for ΛCDM-like parameters, which is comparable to the rotation dipole term (about 0.2 mag for Ω0/h0≈0.4). Since Ω0 is consistent with zero in the Z≤0.1 bin (0.24±0.46) and peaks only in the Z≤0.2 bin (0.29^{+0.21}_{-0.15}), the apparent signal coincides with the redshift range where truncation effects become important. The paper provides no estimate of this truncation error, such as a jerk term or a residual test, so the inferred Ω0 and axis may be biased by the neglected higher-order terms.","section":"§2.2, Eq. (9)"},{"comment":"The abstract states that the AIC indicates a 'statistically significant preference' for the rotating model, but the reported ΔA values (−0.15, −2.23, −4.60, −3.12, −3.87) correspond to model probabilities of only about 1, 3, 10, 4.8, and 6.9 times, which by standard AIC thresholds (e.g., Burnham & Anderson) is at most moderate evidence. In addition, the text's threshold description ('ΔA ≤2 representing strong support, values in the range 4≤ΔA≤7 suggesting moderate to weak support, and ΔA ≥10 indicating no support') is ill-defined because ΔA is negative, and it does not match the standard interpretation. The 'statistically significant preference' claim in the abstract is therefore not supported by the reported numbers.","section":"§4, Model comparison through AIC and Table 1"},{"comment":"The claimed consistency of the anisotropy axis across redshift bins is weakened by the use of cumulative (nested) redshift cuts. Each bin contains all the supernovae from the lower bins, so the axis measurements are strongly correlated; the apparent agreement across bins is therefore not an independent confirmation. Moreover, the declination uncertainties are very large (roughly ±20°–30°), so the axis is only loosely constrained. The authors should either use disjoint redshift bins or explicitly account for the correlation of the axis estimates before claiming a 'broadly aligned' or 'consistent' axis.","section":"§4, Table 1 and Fig. 5"},{"comment":"The evidence for non-zero rotation is marginal even in the most favorable bin. For Z≤0.2, Ω0 = 0.29^{+0.21}_{-0.15}, which is only about 1.6–2σ from zero, and for Z≤0.1, Ω0 = 0.24±0.46, fully consistent with zero. The abstract's phrase 'mild but consistent preference' overstates the strength of the signal; the paper should emphasize that the detection is tentative and significance depends on the choice of redshift cutoff.","section":"§4, Table 1"}],"minor_comments":[{"comment":"The conversion from the fitted equatorial coordinates (Ra, Da) to the angles (θ, φ) that appear in Eq. (9) is not explicitly described; please state the convention used to compute θ and φ for each supernova given a trial axis.","section":"§2.2"},{"comment":"The probabilities P_i = exp(−ΔA_i/2) are not posterior probabilities but relative likelihood ratios; the text should use the term 'relative likelihood' rather than 'probable' to avoid misinterpretation.","section":"§4, Model comparison through AIC"},{"comment":"The abstract's 'statistically significant preference' should be softened to something like 'weak to moderate preference' in light of the ΔA values reported in Table 1.","section":"Abstract"},{"comment":"Only 1σ contours are shown for the fitted axes; plotting 2σ contours would better illustrate the large declination uncertainty and the degree of overlap between bins.","section":"§4, Fig. 5"},{"comment":"The text states that q0 'approaches the standard ΛCDM expectation' at higher redshifts, but the fitted ΛCDM q0 values (−0.07 to −0.22) are far from the canonical value q0≈−0.55; please clarify that this is a cosmographic, low-redshift estimate rather than the full-sample ΛCDM constraint.","section":"§4"},{"comment":"There are several typographical issues, including 'G¨odel' in the title, 'Panthoen+SH0ES' in §4, and inconsistent use of 'effect' vs. 'affect' in the text; a careful proofreading pass is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically straightforward and uses a standard pipeline with public data, but the central claim is not yet convincing. The truncation issue in Eq. (9) is the most serious, as the signal appears exactly in the redshift range where the neglected second-order term is expected to be significant. A revision that either includes the next-order term, restricts the analysis to lower redshift, or provides a quantitative truncation-error estimate could make the paper publishable. The AIC interpretation should also be corrected. I would not recommend rejection because the issues are fixable within the scope of the paper."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper does something genuinely new—fits the full rotation term of an expanding Gödel-type model to the Pantheon+ sample—and the basic analysis is competently done. But the claim of a “statistically significant preference” for rotation is weaker than the abstract suggests, and the signal could easily be a truncation artifact. Worth a serious referee, not because the detection is solid, but because the test is clean and the caveats are instructive.\n\nWhat’s new: earlier work (Jain et al. 2007 and references therein) used small, old supernova samples and dropped the Ω0 term in Eq. (9). Here they use ~1500 SNe up to z=0.5, include the full dipole term, fit the axis and rotation amplitude with a standard MCMC, use the full Pantheon+ covariance, and anchor with Cepheid distances. The equations look right to me, and the likelihood treatment is textbook. They also compare against a similarly truncated ΛCDM model, which is the correct control.\n\nSoft spots: the load-bearing issue is the linear-order truncation. Eq. (9) is O(Z), but they fit to z=0.5. The omitted O(Z^2) terms—jerk and angular counterparts—are not negligible there, roughly a few tenths of a magnitude, the same order as the claimed rotation dipole. In the z≤0.1 bin, where the expansion is safest, Ω0=0.24±0.46, fully consistent with zero. The signal appears only in the z≤0.2 bin (0.29+0.21−0.15) and then declines. That is the pattern you’d expect from truncation bias, and the paper provides no estimate of it. Second, the abstract’s “statistically significant preference” is overstated: ΔA=−4.6 gives P≈10, moderate evidence at best, and the text’s own AIC thresholds are stated in a confused way. Third, the five redshift cuts are nested, so the apparent consistency of the axis across bins is less meaningful than it looks; no look-elsewhere correction is attempted, and the axis uncertainties are large enough that “broadly aligned” is doing a lot of work.\n\nNone of this makes the paper silly. The authors honestly hedge in the conclusions, and the model-comparison design is fair. Strengthening it is straightforward: add the second-order term or restrict to z≤0.2 with truncation error quantified, correct the AIC language, and release the chains.\n\nThis paper is for people working on isotropy tests and rotating cosmologies. It deserves peer review, not desk rejection, because the referee can push on the truncation and force the authors to do the extra work. I’d bring it to a reading group as a cautionary example of how a valid likelihood analysis can still be hostage to the order of a Taylor expansion.","headline":"A clean, honest cosmographic test of a rotating Gödel-type universe against Pantheon+; the new result is a ~2σ dipole that is plausibly an artifact of first-order truncation, so the abstract overstates the evidence.","tokens_in":17917,"tokens_out":4506,"would_cite":false,"duration_ms":45430,"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":"A Gödel-type rotating universe, fit to 1,491 low-redshift Type Ia supernovae, shows a mild but consistent preference for cosmic rotation with $\\Omega_0 = 0.29^{+0.21}_{-0.15}$ at $Z \\le 0.2$ and a stable axis at $(243^\\circ,-49^\\circ)$.","keywords":["cosmic rotation","Gödel-type universe","cosmography","Type Ia supernovae","Pantheon+","anisotropy axis","Akaike Information Criterion","rotating cosmology"],"falsifier":"Refit the same Pantheon+ data after adding the explicit $O(Z^2)$ Kristian–Sachs terms (jerk and second-order rotation contributions) to Eq. (9); if $\\Omega_0$ drops below about 0.1 in all bins, or if a model with an arbitrary dipole direction but no Gödel rotation fits equally well, the reported rotation preference is a truncation artifact rather than evidence for a rotating spacetime.","tokens_in":16898,"feed_emoji":"🌀","tokens_out":8385,"duration_ms":76269,"temperature":0.7,"pith_summary":"The paper asks whether the universe rotates on cosmological scales and answers with a mild yes from Type Ia supernovae. It fits a Gödel-type spacetime that expands and rotates without shear to 1,491 low-redshift Pantheon+ supernovae, using a magnitude–redshift relation expanded to first order in redshift $Z$. The central result is a consistent preference for a nonzero rotation parameter, peaking at $\\Omega_0 = 0.29^{+0.21}_{-0.15}$ for $Z \\le 0.2$, along with a stable anisotropy axis near $(243^\\circ, -49^\\circ)$ in equatorial coordinates. The Hubble constant stays at $h_0 \\approx 0.73$ in every redshift bin, and AIC comparison favors the rotating model over flat $\\Lambda$CDM by factors of roughly 3–10 at intermediate redshifts. If the preference is physical, the local expansion of the universe is direction-dependent and standard isotropic cosmology omits a small but measurable kinematic term.","feed_headline":"Supernova data point to a slowly rotating universe","feed_subtitle":"A Gödel-type model with rotation parameter 0.29 fits low-redshift Pantheon+ data better than flat ΛCDM.","key_machinery":"The carrying object is the expanding Gödel-type line element $$$ds^{2}$ = $dt^{2}$ - 2\\sqrt{\\$\\sigma$} R(t) $e^{{mx}}$ dt\\,dy - $R^{2}$(t)\\left($dx^{2}$ + k $e^{{2mx}}$ $dy^{2}$ + $dz^{2}$\\right),$$ with $k>0$ excluding closed timelike curves; the rotation rate $\\omega = \\frac{m}{2R}\\sqrt{\\sigma/(k+\\sigma)}$ decays as the universe expands and defines the anisotropy axis. From the Kristian–Sachs expansion of area distance in powers of redshift, the paper uses the first-order apparent magnitude–redshift relation (its Eq. 9), which adds to the usual Hubble and deceleration terms a directional dipole controlled by $\\Omega_0/h_0$ and by $\\tilde{\\rho} = \\sqrt{\\sigma/(k+\\sigma)}$. This equation is what converts a hypothetical cosmic rotation into a redshift- and direction-dependent brightness offset, and the MCMC fit of its parameters to the Pantheon+ data is the core of the analysis.","core_discovery":"On its own terms, the paper establishes that the directional terms in the Gödel-type magnitude–redshift relation are preferred by the data: the dimensionless rotation parameter $\\Omega_0$ is nonzero out to about $2\\sigma$ at $Z \\le 0.2$, and the inferred rotation axis $(R_a, D_a) \\approx (243^\\circ, -49^\\circ)$ is consistent across all five redshift cutoffs. The same fit gives a stable $h_0 \\approx 0.73$ and a deceleration parameter $q_0$ that moves from near zero to mildly negative values with increasing redshift. The paper reads this not as a detection but as a consistent mild preference, strong enough that a rotating, shear-free Gödel-type cosmology is statistically competitive with, or favored over, flat $\\Lambda$CDM at intermediate redshifts and deserves a test beyond the first-order cosmographic expansion.","pith_inferences":["Because the fit uses only the first-order expansion, the reported $\\Omega_0$ could be absorbing neglected $O(Z^2)$ terms such as the jerk and second-order rotation corrections; a concrete next step is to compute those terms from the Kristian–Sachs series and see whether the dipole survives.","If the rotation signal is real, it may be relevant to the Hubble tension: a direction-dependent expansion changes how local distance-ladder measurements and early-universe anchors are compared, since they effectively average over different sky directions.","The closeness of the inferred axis to other reported cosmic dipoles suggests a joint test: fitting the rotation axis together with CMB, radio, and fine-structure dipole data in one model would reveal whether these are the same underlying anisotropy or unrelated alignments.","Future low-redshift supernova samples could distinguish the Gödel-type model from a generic dipole by checking the predicted redshift dependence $\\omega \\propto 1/R$: rotation should weaken with distance in a specific way."],"forward_implications":["If the preference is physical, the local expansion rate is direction-dependent at the level $\\Omega_0/h_0 \\approx 0.40$ in the $Z\\le0.2$ bin, largest for sources near the inferred rotation axis.","The steady decline of $\\Omega_0$ and $\\Omega_0/h_0$ with increasing redshift cutoff means rotation, if present, is a relatively local effect that fades as more distant supernovae enter the sample.","A rotation axis stable near $(243^\\circ, -49^\\circ)$ across all bins predicts a specific sky direction for other anisotropy probes, close to axes previously reported from radio polarization and other data.","The stable $h_0 \\approx 0.73$ implies that allowing for rotation does not wash out the local Hubble rate, while $q_0$ becomes less negative than in a no-rotation fit at the same redshift cuts."],"supporting_citations":[{"why":"Supplies the Kristian–Sachs power-series formalism used to expand the area distance and derive the magnitude–redshift relation.","marker":"[65]"},{"why":"Together with [64], source of the first-order m–Z relation (Eq. 9) for the Gödel-type rotating model.","marker":"[63]"},{"why":"Derives the observational effects of cosmic rotation and the m–Z expansion used here.","marker":"[64]"},{"why":"Provides the Pantheon+ SNIa light curves, covariance matrices, and corrected magnitudes that are fitted.","marker":"[66]"},{"why":"Supplies the SH0ES Cepheid-anchor distances that break the h0–M0 degeneracy in the fit.","marker":"[69]"},{"why":"Earlier analysis of the same Gödel-type model on gold/silver SN data, which this work extends to Pantheon+ with the full rotation term.","marker":"[78]"},{"why":"Defines the Akaike Information Criterion used to compare the rotating model against ΛCDM.","marker":"[80]"},{"why":"Gives the ΔA thresholds and model probabilities used to interpret the AIC preference.","marker":"[81]"}],"fun_headline_variants":["Supernovae hint at a rotating universe","Gödel-type model favored by Pantheon+ supernovae","Rotating universe: mild preference from cosmic data","A rotating universe fits supernova data slightly better","Cosmic rotation gets a quiet boost from SNIa data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the first-order Taylor expansion of the Gödel-type magnitude–redshift relation is accurate for redshifts up to $Z=0.5$; if the omitted second-order terms are not negligible, the fitted rotation parameter can absorb them and mimic a rotation signal that is not in the metric.","fun_headline_variants_meta":{"raw":{"variants":["Supernovae hint at a rotating universe","Gödel-type model favored by Pantheon+ supernovae","Rotating universe: mild preference from cosmic data","A rotating universe fits supernova data slightly better","Cosmic rotation gets a quiet boost from SNIa data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000554,"raw_usage":{"total_tokens":2647,"prompt_tokens":963,"completion_tokens":1684,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":579,"completion_tokens_details":{"reasoning_tokens":1608}},"tokens_in":579,"tokens_out":1684,"duration_ms":13267,"temperature":1.0,"reasoning_tokens":1608,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:56:30.517017+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Refit the same Pantheon+ data after adding the explicit $O(Z^2)$ Kristian–Sachs terms (jerk and second-order rotation contributions) to Eq. (9); if $\\Omega_0$ drops below about 0.1 in all bins, or if a model with an arbitrary dipole direction but no Gödel rotation fits equally well, the reported rotation preference is a truncation artifact rather than evidence for a rotating spacetime.","supporting_citations":[{"cited_title":"Kristian and R","cited_arxiv_id":null,"evidence_quote":"Supplies the Kristian–Sachs power-series formalism used to expand the area distance and derive the magnitude–redshift relation."},{"cited_title":"Korotky and Yuri N","cited_arxiv_id":null,"evidence_quote":"Together with [64], source of the first-order m–Z relation (Eq. 9) for the Gödel-type rotating model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Derives the observational effects of cosmic rotation and the m–Z expansion used here."},{"cited_title":"Riess et al","cited_arxiv_id":null,"evidence_quote":"Supplies the SH0ES Cepheid-anchor distances that break the h0–M0 degeneracy in the fit."},{"cited_title":"Modgil, and John P","cited_arxiv_id":null,"evidence_quote":"Earlier analysis of the same Gödel-type model on gold/silver SN data, which this work extends to Pantheon+ with the full rotation term."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the Akaike Information Criterion used to compare the rotating model against ΛCDM."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the ΔA thresholds and model probabilities used to interpret the AIC preference."}],"review_version":1}