{"id":"9b7a5659-0246-4f09-a158-8822ea3d3bb2","arxiv_id":"2501.12264","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A cosmological fit with Planck CMB and DESI BAO data rules out purely bosonic neutrinos at 95% confidence and favors Fermi-Dirac statistics.","lead":"This paper fits the latest cosmic microwave background and galaxy clustering data to test whether neutrinos obey Fermi-Dirac or Bose-Einstein statistics. The data exclude purely bosonic neutrinos at 95% confidence and prefer fermionic statistics, with a three-way degeneracy among neutrino statistics, neutrino mass, and the Hubble constant.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Exclusion of bosonic neutrinos rests on Planck's H0 value; adding a SH0ES H0 prior lets κν=-1 into the 95% interval, so the claim is prior-dependent until this degeneracy is tested.","rationale":"The paper is honest and the physics is standard: changing κν changes the neutrino phase-space distribution and hence the early-universe radiation density and the late-time matter density, and the CMB peak parameter θs shows a three-way degeneracy among κν, Σmν, and H0 (Fig. 3c). The reader's weakest-assumption diagnosis matches the authors' own statement that the exclusion is driven by the Planck H0 constraint. My stress-test sharpens this into a quantitative fragility: the entire reversal of the central claim occurs with a ~1 km/s/Mpc change in H0 and only Δχ²_CMB ≈ 3.8, which is within the 95% confidence region of the CMB-only fit. Therefore the conclusion 'purely bosonic neutrinos are ruled out at 95% CL' is not a pure statement about neutrino statistics; it is a joint statement about neutrino statistics and the Hubble scale. The CMB+BAO+SNe exception is not a minor footnote but the same model with a different H0 assumption, so the abstract should either state the H0 dependence or restrict the claim to the Planck-calibrated H0 scale. I do not see an internal inconsistency in the analytic argument: the relativistic and non-relativistic density ratios (8/7 and 4/3) are correct, and the numerical robustness check with non-degenerate masses in Sec. 5 supports the modeling. The main missing piece is verification: the modified CLASS code is not released, and the perturbation treatment is not described, but that affects reproducibility rather than the logic of the H0-based concern. A clean H0-prior test would settle whether the 95% exclusion is a robust cosmological constraint or a prior-driven selection. Since the reader already conditioned acceptance on code release and abstract qualification, my read does not move the verdict; it strengthens the reason for the H0 condition.","tokens_in":12192,"tokens_out":16242,"duration_ms":191650,"concrete_test":"Re-run the Section 3 analysis with the CMB and DESI BAO likelihoods unchanged, but add a Gaussian prior on H0 of 73.04 ± 1.04 km/s/Mpc (SH0ES) instead of the Pantheon SN data and supernova magnitude prior used in CMB+BAO+SNe. Keep all other priors identical to Table I. If the 95% lower bound on κν from this CMB+BAO+SH0ES fit moves to or below -1, the exclusion of purely bosonic neutrinos is an artifact of the Planck H0 scale and the abstract's claim should be qualified as H0-dependent. If the bound stays above, say, -0.5, the central claim survives this challenge.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that CMB and CMB+BAO data rule out κν=-1 at 95% CL is carried by the κν-H0 degeneracy identified in Sec. 4 and Fig. 5. The paper itself says the exclusion 'can be attributed to the stringent constraint on H0 by the CMB data.' This attribution is load-bearing because the same CMB+BAO data, augmented only by the SH0ES local H0 determination and Pantheon SNe (the CMB+BAO+SNe row of Table II), shift the mean κν from 0.354 to -0.051 and make the 95% lower bound -1, so purely bosonic neutrinos are no longer excluded. Table III shows this reversal costs only Δχ²_CMB ≈ 3.8 (2799.22 vs 2795.44) while H0 moves from 68.36 to 69.40 km/s/Mpc. A 95% exclusion that can be erased by a ~1 km/s/Mpc upward shift in the assumed H0 is not robust to the unresolved Hubble tension. The paper discounts the CMB+BAO+SNe result as statistically ill-founded, but that is an assertion: no calculation is given to show how the κν lower bound varies with the H0 prior, so the claim 'purely bosonic neutrinos can be ruled out' is conditional on adopting the Planck H0 scale.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends ΛCDM by a neutrino statistics parameter κν, interpolating between Fermi-Dirac (κν=1) and Bose-Einstein (κν=-1) occupation, and fits the model to CMB-only (Planck PR3 + ACT/PR4 lensing) and CMB+BAO (DESI DR1/DR2) datasets. The headline finding, stated in the abstract and Table II, is that purely bosonic neutrinos are excluded at 95% confidence for the CMB and CMB+BAO analyses, with lower bounds κν > -0.317 and κν > -0.489, respectively. The CMB+BAO+SNe combination, which includes a local H0 constraint via the supernova absolute magnitude, shifts the lower bound to -1 and no longer excludes bosonic neutrinos; the paper attributes this reversal to the Hubble tension and cautions that this combination is not statistically well-founded. A secondary conclusion is that CMB+BAO data prefer fermionic statistics, while mixed statistics remain possible.","tokens_in":1441,"tokens_out":1496,"duration_ms":75116,"significance":"If the central claim holds, the paper would provide a novel cosmological constraint on a fundamental property of neutrinos, going beyond earlier bounds that allowed bosonic statistics at 95% CL. The paper is transparent about the CMB+BAO+SNe reversal, gives a clear physical discussion of the κν–Σmν–H0 degeneracy, and includes a post-submission check with DESI DR2. These are genuine strengths. However, as argued below, the headline exclusion is conditional on the assumed H0 scale and on an unverified implementation of the modified statistics in the Boltzmann code, so the significance is currently provisional.","major_comments":[{"comment":"The paper's central claim that purely bosonic neutrinos are ruled out at 95% CL is not robust to the assumed H0 scale. The text itself states that the exclusion 'can be attributed to the stringent constraint on H0 by the CMB data,' and Table II shows that the CMB+BAO+SNe fit, which adds the local H0 information, shifts the 95% lower bound from -0.489 to -1, i.e., bosonic neutrinos become allowed. The paper dismisses this dataset as 'not statistically well-founded' because of the Hubble tension, but no calculation is given to show how the κν lower bound varies with the H0 prior. Because the Hubble tension is an unresolved systematic, the exclusion claim is conditional on the Planck H0 scale. The authors should quantify the prior dependence, for example by re-running the fits with a wide H0 prior or with a Gaussian prior centered on the local value, and should report the Δχ2 between κν=-1 and the best-fit for the CMB and CMB+BAO analyses. Without such a test, the abstract's unconditional statement is not supported.","section":"Sec. 4, Fig. 5, Tables II and III"},{"comment":"The analysis uses a modified version of CLASS, but the paper neither provides the code nor describes how κν enters the perturbation equations. The CMB constraints come from the full angular power spectra, which depend on the neutrino perturbation hierarchy (density, velocity, anisotropic stress, free-streaming), not only on the background energy density computed in Eqs. (3)–(8). If the modification changes only the background cosmology, the resulting CMB spectra may be inconsistent with the assumed statistics. The authors should release the code or provide a complete derivation of the perturbation equations for the modified distribution, and validate the implementation, for instance by reproducing standard results for κν=1 and showing the response of the CMB spectra to κν in a test case. This is load-bearing because the exclusion claim relies on the CMB power spectra.","section":"Sec. 3 and Sec. 2, Eqs. (3)–(8)"}],"minor_comments":[{"comment":"There are several typos and notation errors: 'appoaches' in Sec. 2, 'unlablled' in Fig. 3 captions, 'adpoted' in Table I, and 'kν' in the Fig. 3 captions where κν is meant. These should be corrected.","section":"Throughout"},{"comment":"Reference [35] is given the same arXiv number (2211.04492) as reference [20]; the Herold & Kamionkowski entry appears to have the wrong identifier.","section":"References"},{"comment":"The label CMB+BAO+SNe is used for a dataset that includes a Gaussian constraint on the supernova absolute magnitude from [14], which effectively incorporates the local H0 measurement. The caption of Fig. 5 calls this 'local H0 measurement'; the terminology should be unified so readers can clearly see that the local H0 prior is included in the CMB+BAO+SNe combination.","section":"Sec. 3 and Fig. 5"},{"comment":"Table III reports χ2 values 'obtained from the MCMC chains'; for best-fit comparisons, a proper χ2 minimization would be more reliable than values read off from MCMC chains.","section":"Table III"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents an interesting and potentially important result, but two issues need to be addressed before publication: (i) the robustness of the κν exclusion to the H0 scale, given the Hubble tension and the paper's own admission that the exclusion is driven by Planck's H0; and (ii) the lack of detail or code release for the modified CLASS implementation. The authors have been transparent about the CMB+BAO+SNe reversal, which is commendable, but the abstract currently overstates the robustness of the conclusion. I would encourage the editor to request a detailed response to the two major comments and to ask for the modified code or a full description of the perturbation-scheme changes."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a careful, honest update of de Salas et al. (2018) with newer CMB and BAO data, and the headline result—bosonic neutrinos excluded at 95% from CMB and CMB+BAO—is real but rests on Planck's H0. If you put a SH0ES-style H0 prior into the fit, the lower bound on κν drops to −1 and the exclusion evaporates. The paper says as much in the text, but the abstract does not.\n\nWhat is new: they redo the variable-statistics model of [8] with Planck PR3, ACT DR6 lensing, and DESI DR1 BAO, plus a DESI DR2 check in a note added. The bounds are genuinely stronger: κν > −0.317 (CMB) and > −0.489 (CMB+BAO), so purely bosonic neutrinos are out at 95%. That is a legitimate quantitative result. They also lay out the κν–Σmν–H0 degeneracy clearly, and they are transparent that the CMB+BAO+SNe combination, which would allow κν = −1, is statistically ill-founded because of the Hubble tension.\n\nSoft spots, in order of severity. First, the central claim is prior-dependent. The exclusion is driven by the Planck H0 measurement through the degeneracy they themselves map. Adding a local H0 prior shifts the mean κν from 0.354 to −0.051 and makes the 95% lower bound −1. They dismiss that combination as ill-founded, which is a defensible statistical position when datasets are in tension, but they do not show how the bound varies with the H0 prior, so \"ruled out\" is conditional on the Planck H0 scale. Second, the modified CLASS code is not released or described; the perturbation implementation is a black box. For a paper whose whole point is a non-standard distribution, that matters. Third, the abstract overstates: it says \"ruled out at 95%\" with no mention of the SNe exception or the H0 dependence. The text is more careful.\n\nNone of this is fatal. The paper is honest about its main weakness, the numbers are plausible, and the DR2 check is a nice touch. It deserves a serious referee, with code release and abstract qualification as conditions.","headline":"A solid, honest update of the de Salas et al. neutrino-statistics analysis with newer data, but the headline exclusion of bosonic neutrinos rests on Planck's H0 and evaporates if H0 shifts upward by ~1 km/s/Mpc.","tokens_in":13002,"tokens_out":2324,"would_cite":false,"duration_ms":22561,"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":"Cosmological data can now exclude purely bosonic neutrinos at 95% confidence.","keywords":["neutrino statistics","Bose-Einstein statistics","Fermi-Dirac statistics","cosmological constraints","cosmic microwave background","baryon acoustic oscillations","neutrino mass","Hubble constant"],"falsifier":"Fix the Hubble constant at values 2, 5, and 8 km s$^{-1}$ Mpc$^{-1}$ above the CMB-inferred one and rerun the same fit; if the 95% lower bound on $\\kappa_\\nu$ crosses $-1$ for any of these shifts, the exclusion claim fails as stated. The authors' CMB+BAO+SNe fit, which effectively includes such a shift, already produces a bound of $-1$, so this comparison is directly checkable with published chains.","tokens_in":11974,"feed_emoji":"⚛️","tokens_out":8185,"duration_ms":78129,"temperature":0.7,"pith_summary":"The paper asks whether cosmology can tell us not just how heavy neutrinos are but what quantum statistics they obey. It extends the standard cosmological model with a single parameter $\\kappa_\\nu$ that interpolates between Fermi-Dirac statistics ($\\kappa_\\nu=1$) and Bose-Einstein statistics ($\\kappa_\\nu=-1$), and fits the resulting model to CMB and baryon acoustic oscillation (BAO) data. The central claim is that purely bosonic neutrinos are excluded at 95% confidence, while purely fermionic and mixed-statistics neutrinos remain viable. The exclusion works through a degeneracy among $\\kappa_\\nu$, the sum of neutrino masses, and the Hubble constant, so the CMB's tight constraint on the Hubble constant is what pushes the low-statistics branch out. If the claim holds, cosmology becomes a direct probe of the spin-statistics connection for neutrinos.","feed_headline":"Bosonic neutrinos ruled out by cosmology at 95% confidence","feed_subtitle":"One parameter bridging Fermi and Bose statistics is pinned close to the fermionic side by CMB and BAO data.","key_machinery":"The load-bearing object is the statistical parameter $\\kappa_\\nu$ entering the neutrino phase-space distribution $f = 1/(e^{E/T} + \\kappa_\\nu)$, which controls the neutrino energy density at every epoch: Bose-Einstein neutrinos carry $8/7$ times the photon-scaled energy density of Fermi-Dirac neutrinos when relativistic and $4/3$ times when non-relativistic. The argument runs through the angular scale of the sound horizon at recombination, $\\theta_s = r_s/D_A$, because increasing $\\kappa_\\nu$, increasing the neutrino mass sum, and increasing the Hubble constant all move $\\theta_s$ in the same direction. That near-degeneracy is the machinery: the CMB's sharp measurement of $\\theta_s$ and of $H_0$ converts into a lower bound on $\\kappa_\\nu$, especially once small masses and BAO shrink the other degeneracy directions.","core_discovery":"On the paper's own terms, the discovery is that current CMB data alone, and CMB combined with BAO distances, rule out the possibility that neutrinos obey Bose-Einstein rather than Fermi-Dirac statistics. In the authors' fit, the 95% lower bounds on the statistical parameter are $\\kappa_\\nu > -0.317$ for CMB only and $\\kappa_\\nu > -0.489$ for CMB+BAO, both excluding the purely bosonic value $\\kappa_\\nu=-1$. Adding supernova and local-$H_0$ information reverses this exclusion, but the paper treats that combination as statistically compromised by the Hubble tension and reports it only for comparison. The fits also favor small neutrino masses, and allowing mixed statistics raises the BAO-based neutrino-mass upper limit slightly, from 0.072 eV to 0.078 eV, which modestly eases the tension with oscillation lower bounds. The second conclusion is that the data prefer fermionic or nearly fermionic statistics, with mixed statistics still open below the 95% boundary.","pith_inferences":["As an editorial extension, the same degeneracy implies that if the Hubble tension is eventually resolved toward the higher local value, the $\\kappa_\\nu$ bound would probably cross $-1$ on its own; the paper's own CMB+BAO+SNe fit is a preview of that regime and is why the authors discount it.","Because the degeneracy runs through the sound-horizon scale, any early-universe mechanism that changes the sound horizon (extra relativistic species, early dark energy) could mimic part of the neutrino-statistics signal, so the bound is model-dependent rather than pure spin-statistics evidence.","A direct quantitative test: refit with a Gaussian prior on $H_0$ whose central value is varied from 67 to 73 km s$^{-1}$ Mpc$^{-1}$ and map the 95% lower bound on $\\kappa_\\nu$; the curve would identify exactly which future $H_0$ measurement would flip the exclusion."],"forward_implications":["Purely bosonic neutrinos are excluded at the 95% confidence level by both CMB-only and CMB+BAO fits, so any particle-physics model that predicts $\\kappa_\\nu=-1$ is now in tension with cosmology.","Mixed-statistics neutrinos with $\\kappa_\\nu \\gtrsim -0.5$ at 2$\\sigma$ remain allowed, meaning the observationally viable distributions are closer to fermionic than bosonic.","Allowing $\\kappa_\\nu$ to vary relaxes the BAO-based upper bound on the neutrino mass sum from 0.072 eV to 0.078 eV, making the cosmology-oscillation mass tension slightly smaller.","Tighter future measurements of the Hubble constant and of BAO distances should sharpen the bound on $\\kappa_\\nu$ and could eventually distinguish fermionic from mixed statistics."],"supporting_citations":[{"why":"This earlier cosmological analysis of neutrino statistics supplies the baseline bounds that the paper's new datasets and fits strengthen.","marker":"[8]"},{"why":"This CMB temperature and polarization likelihood provides the cosmological observations whose sharp Hubble-constant constraint drives the exclusion of bosonic neutrinos.","marker":"[15]"},{"why":"These baryon acoustic oscillation distance measurements, combined with the CMB data, produce the 95% lower bound on $\\kappa_\\nu$ reported for the CMB+BAO fit.","marker":"[24, 25]"},{"why":"This BAO data release supplies the neutrino-mass upper limit that the paper compares against to show how mixed statistics relax the bound.","marker":"[5]"},{"why":"This Boltzmann solver is the code the authors modify to compute power spectra and distance quantities with variable neutrino statistics.","marker":"[13]"}],"fun_headline_variants":["Cosmology rules out bosonic neutrinos at 95% CL","Neutrinos are fermions, say CMB and BAO","Bose neutrinos dead: cosmology's 95% verdict","Fermionic neutrinos preferred at 95% CL"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Hubble constant derived from CMB data is accurate; if the true value is instead closer to the higher number favored by local distance measurements, the degeneracy could let purely bosonic neutrinos survive.","fun_headline_variants_meta":{"raw":{"variants":["Cosmology rules out bosonic neutrinos at 95% CL","Neutrinos are fermions, say CMB and BAO","Bose neutrinos dead: cosmology's 95% verdict","Fermionic neutrinos preferred at 95% CL"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000628,"raw_usage":{"total_tokens":2853,"prompt_tokens":841,"completion_tokens":2012,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":457,"completion_tokens_details":{"reasoning_tokens":1939}},"tokens_in":457,"tokens_out":2012,"duration_ms":16592,"temperature":1.0,"reasoning_tokens":1939,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T17:20:14.104094+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fix the Hubble constant at values 2, 5, and 8 km s$^{-1}$ Mpc$^{-1}$ above the CMB-inferred one and rerun the same fit; if the 95% lower bound on $\\kappa_\\nu$ crosses $-1$ for any of these shifts, the exclusion claim fails as stated. The authors' CMB+BAO+SNe fit, which effectively includes such a shift, already produces a bound of $-1$, so this comparison is directly checkable with published chains.","supporting_citations":[{"cited_title":"Cosmological bounds on neutrino statistics","cited_arxiv_id":null,"evidence_quote":"This earlier cosmological analysis of neutrino statistics supplies the baseline bounds that the paper's new datasets and fits strengthen."},{"cited_title":"Planck2018 results: VI. Cosmological parameters","cited_arxiv_id":null,"evidence_quote":"This CMB temperature and polarization likelihood provides the cosmological observations whose sharp Hubble-constant constraint drives the exclusion of bosonic neutrinos."},{"cited_title":"The Cosmic Linear Anisotropy Solving System (CLASS). Part II: Approximation schemes","cited_arxiv_id":null,"evidence_quote":"This Boltzmann solver is the code the authors modify to compute power spectra and distance quantities with variable neutrino statistics."}],"review_version":1}