{"id":"f9774338-3729-4824-9b67-ce929ae9bc5f","arxiv_id":"2607.26141","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Electromagnetic sectors built from projectively invariant torsion and non-metricity can modify light cones, intensity transport, and polarization structure in geometric optics.","lead":"This paper constructs two ways to couple electromagnetism to a spacetime geometry that includes torsion and non-metricity, guided only by standard gauge symmetry and projective invariance. It derives how light rays, their intensity, and their polarization would behave in such geometries, setting up future tests with black-hole images and lensing.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Birefringence/mode-mixing claims rest on the uncomputed tensor B; Eq. (60)'s effective metric is not established as the physical light cone until B is shown not to cancel or dominate S.","rationale":"The reader's weakest assumption is that the minimal constitutive Ansatz (38) is rich enough and that the uncomputed tensorial part B would produce the claimed birefringence and mode mixing. My read identifies the same load-bearing gap and sharpens it: the problem is not merely that some admissible terms were omitted, but that the paper's own displayed equations cannot determine whether even the terms it keeps produce the effects. Equation (58) separates P into S and B, and Eq. (60) is offered as an effective metric, yet the actual characteristic surfaces are determined by det P = 0, which requires B. The pure-trace torsion example shows that S alone can contain terms with no physical effect, because T_f=0 makes χ equal to Maxwell while S written in full torsion is nonzero. Hence S and g_eff are not physical until B is computed. This is a correctness risk in the central claim, not merely a missing generality proof. I credit the scalar-coupled model: the derivation of I=ΞA² and the conservation law is coherent and constitutes a valid formal result. I also note the paper's own honesty in deferring B to future papers, which supports a conditional rather than rejection verdict. The recommended action is therefore to keep the reader's CONDITIONAL verdict unchanged, with the concrete task above as the minimal check that would turn the conditional status into a definite one.","tokens_in":12344,"tokens_out":11248,"duration_ms":102686,"concrete_test":"Take flat spacetime with a single nonzero trace-free torsion component, e.g. T^0_{12} = τ, Q = 0. Insert Ansatz (38) into Eq. (54), form the 4×4 matrix P^α_μ of Eq. (58), impose Lorenz k·a=0, and solve det P = 0 for, say, wave vectors k=(ω,k_x,0,k_z). Compare the roots for the two transverse polarization modes against the single root predicted by S=0 from Eq. (59). If det P factorizes into two distinct branches with different phase speeds, birefringence and mode mixing are genuine consequences of Ansatz (38). If the two branches coincide or if the transverse projection of B vanishes on-shell, then the constitutive model as written does not yield the advertised effects, and the conditional verdict should be revised downward.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central load-bearing step is the constitutive-tensor model's claim that torsion and non-metricity produce modified light cones, birefringence, and mode mixing. At O(ω²) the propagation operator is P^α_μ = δ^α_μ S + B^α_μ (Eq. 58). The paper never computes B, stating that its explicit form is lengthy and deferring it to future work. But without B the physical dispersion relation det P = 0 cannot be evaluated, and S=0 (Eq. 59) is not the light cone: B can shift the determinant and, in the physical transverse subspace, can even cancel terms in S. This is not hypothetical. If the torsion is pure trace, then T_f = 0, the Ansatz (38) reduces to Maxwell, and the physical cone is k²=0; nevertheless the S written in Eq. (59) contains (k·T)² terms because it is expressed in full torsion. Thus S alone is projectively variant and unphysical; B must remove those terms. The advertised interpretation of Eq. (60) as the 'scalar sector' effective metric is therefore premature. The paper also gives no census of admissible quadratic χ terms; the 'minimal Ansatz (38)' is asserted, so even if B is nonzero for this particular Ansatz, the claim that these effects are generic features of the symmetry-guided framework is unsupported. The scalar-coupled half, by contrast, is internally sound: I = ΞA² and its conservation along rays follow from the stated equations. The problem is specifically that the paper's headline birefringence and mode-mixing conclusions are not derived from the displayed expressions.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs two symmetry-guided electromagnetic actions in metric-affine geometry, using U(1) gauge invariance and projective invariance as selection principles. The first model multiplies the Maxwell term by a scalar Ξ built from projectively invariant torsion and non-metricity invariants; the second replaces the kinetic term with a constitutive tensor χ^{μναβ}. The geometric-optics limit is derived in each case. For the scalar model the light cone is unchanged and the intensity I = Ξ A² satisfies the standard transport law. For the constitutive model the paper claims modified light cones, birefringence, and polarization mode mixing, and interprets the scalar part S of the polarization operator as defining an effective metric (Eq. 60). The tensorial piece B of the polarization operator is, however, not computed, and the manuscript explicitly defers it to future work. The constitutive-model claims therefore rest on an unverified assumption about B and on an asserted minimal Ansatz.","tokens_in":12784,"tokens_out":3094,"duration_ms":32857,"significance":"If fully established, the framework would give a formal starting point for connecting torsion and non-metricity to a broad set of electromagnetic observables (black-hole shadows, birefringent lensing, CMB polarization, luminosity-distance anomalies). The scalar-coupled model is worked out completely and is internally consistent; the intensity-transport result I = Ξ A² with IΣ = const is a clean, falsifiable prediction. The constitutive-tensor model, however, has an explicit gap: the central birefringence/mode-mixing conclusions are not derived from the displayed equations because B in Eq. (58) is never computed and Eq. (60) is only a 'scalar sector' object, not the physical dispersion relation det P = 0. The paper also ships no code or machine-checked derivations; its strength is the symmetry classification, but the classification itself is asserted rather than proved. For these reasons the significance is conditional on closing the B gap.","major_comments":[{"comment":"The physical light cone is det P^α_μ = 0, not S = 0. The manuscript computes only S and states that B is too lengthy to be displayed (after Eq. 59), yet immediately interprets Eq. (60) as an effective inverse metric. This is not justified: B contributes to the determinant and can shift, cancel, or dominate the S terms. A concrete failure mode is pure-trace torsion: T_f = 0 implies Q_f = 0 and the Ansatz (38) reduces to Maxwell, so P = δ k², but Eq. (59) written in full torsion contains 2/9 c̄_T k_μ k_ν T^μ T^ν. Thus S alone is not projectively invariant and is not the physical cone; B must cancel such terms. The advertised 'scalar sector effective metric' is therefore not established as a light cone. To support the paper's headline claims, B (or at least det P) must be computed for the explicit Ansatz, or the claims must be restricted to the scalar sector in a way that makes clear they a","section":"§IV.B, Eqs. (58)–(60)"},{"comment":"The paper presents Eq. (34) as 'the most general Ansatz' of parity-even, derivative-free quadratic invariants, and Eq. (38) as a 'minimal' projectively invariant constitutive Ansatz. No derivation or census of admissible terms is given. For the constitutive tensor, the paper acknowledges a 'substantially larger set of tensorial structures' but asserts that (38) is 'already rich enough' to exhibit the claimed effects. This assertion is load-bearing: the claim that torsion and non-metricity generically induce birefringence and mode mixing in the constitutive framework requires either a complete classification or a demonstration that omitted terms cannot cancel the effects produced by (38). Without this, the conclusions in §V that the model 'can affect light propagation in several distinct ways' are not supported for the constitutive model.","section":"§III.B, Eqs. (34) and (38)"},{"comment":"The transport analysis is incomplete in a way that affects the intensity and mode-mixing claims. Equation (65) is derived only under the eigenmode assumption (63), and Eq. (67) is schematic, with the mixing matrix M^(i)_(j) never defined from the explicit χ. The later phenomenological statements about 'energy leakage', 'depolarization', and 'photon-number loss' (see also §V) are therefore not consequences of the displayed derivation. The manuscript should either derive the mode-mixing transport equation for the concrete Ansatz (38) or clearly label these effects as conjectures to be studied in future work.","section":"§IV.B, O(ω) transport, Eqs. (61)–(67)"}],"minor_comments":[{"comment":"The notation ˚F_{μν} = ˚∇_μ A_ν − ˚∇_ν A_μ is introduced, but later the same symbol is used for the Levi-Civita covariant derivative while F is also written as dA in Eq. (11). Please make the distinction between F and ˚F consistent throughout, especially in Eqs. (13), (33), and (35).","section":"§II.B, Eq. (12)"},{"comment":"The first form of the constitutive Ansatz has an index placement that is hard to parse: T_{f δ}^{μν} T_{f δ}^{αβ} in Eq. (37) is not explicitly defined. The equivalence to Eq. (38) is asserted but not demonstrated; a short derivation or a footnote would help.","section":"§III.B, Eq. (37)"},{"comment":"The non-affinely parametrized geodesic equation is written with k^μ_eff; it would be clearer to show that ˚∇_k_eff k_eff^μ ∝ k_eff^μ and state the reparametrization explicitly, since this is a standard but easily misread point.","section":"§IV.A, Eq. (44)"},{"comment":"There is a typographical issue in the email address 'pastor c@thphys.uni-heidelberg.de' and a missing space in 'Institut f¨ ur'. Also, the arXiv identifier in the header (2607.26141) appears several times; please ensure the final published version has the correct metadata.","section":"Author affiliations and abstract"},{"comment":"Ref. [2] is a 2026 preprint; if available, add a more complete reference. The list is otherwise adequate, but the text would benefit from explicit page or equation references when citing previous GO derivations, e.g., for the standard result (18).","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The scalar-model part of the paper is sound and publishable in principle. The constitutive-model part, however, is the source of the most interesting and most prominently advertised conclusions, and those conclusions are not derived: B is uncomputed, the minimal Ansatz is asserted, and the effective metric (60) is not shown to be the physical light cone. The stress-test concern about pure-trace torsion lands decisively: Eq. (59) cannot be the full dispersion relation unless B cancels the (k·T)^2 terms, and that cancellation is not shown. This is a fixable gap — the authors could compute B for the stated Ansatz or substantially weaken the claims — but it is not a presentation issue. I therefore recommend major revision rather than rejection, with the expectation that the constitutive-model claims be either fully derived or explicitly labeled as conditional on future work."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read this one for the scalar model, not for the advertised birefringence. The projective-symmetry + U(1) organizing principle is genuinely useful, and the geometric-optics derivation for the scalar-dressed Maxwell term is clean: light cone stays null (conformal rescaling) and the conserved beam quantity is the dressed intensity I=ΞA², with IΣ=const. That is a real result and it is fully derived from the stated equations.\n\nThe constitutive-tensor model is where the paper's headline claims live, and the stress-test is right: they are not actually derived. The propagation operator is decomposed as P=δ S + B, but B is never computed. The paper says the explicit form is lengthy and defers it to a future paper. Without B, det P=0 — the true dispersion relation — cannot be evaluated, so Eq. (59)/(60) is not demonstrated to be the physical light cone. The pure-trace-torsion sanity check makes the problem concrete: T_f=0 reduces the Ansatz to Maxwell, the physical cone is k²=0, yet S as written still contains (k·T)² terms. B would have to cancel those terms; since B is not given, the displayed S is at best an unphysical intermediate. The \"scalar sector effective metric\" interpretation is therefore premature, and the birefringence/mode-mixing conclusions in the abstract and conclusions are promissory notes, not results of this paper.\n\nTwo smaller issues. The \"most general\" quadratic Ξ in Eq. (34) and the \"minimal Ansatz\" (38) are asserted without a completeness proof; a census of admissible index contractions would be straightforward and should be added. And the photon-number non-conservation claim is only speculative — it is tied to the derivative term ∇χ and is not demonstrated for either model. On the citation pattern: self-citations are background, not load-bearing; nothing circular there.\n\nThe framework is worth engaging. The scalar half stands, and the constitutive half is a plausible route that needs one explicit computation — B for a concrete background, or at least a demonstration that the physical dispersion relation is projectively invariant before interpreting S. Send it to a serious referee, but the referee should demand that computation before the birefringence claims are allowed to stand.","headline":"A clean scalar-model derivation wrapped around a constitutive-tensor model whose headline birefringence and mode-mixing claims are not actually derived — the gap is load-bearing and needs to be fixed before the paper does what it advertises.","tokens_in":13230,"tokens_out":4129,"would_cite":false,"duration_ms":37144,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["83C10","83C22","83C50","78A05"],"pacs":["04.20.Cv","04.40.Nr","41.20.Jb","42.25.Lc","98.62.Sb"],"model":"deepseek-v4-flash","headline":"Torsion and non-metricity can bend light, alter intensity, and mix polarizations in metric–affine spacetimes, this paper argues, deriving the geometric-optics limit for two symmetry-guided electromagnetic models.","keywords":["metric-affine geometry","torsion","non-metricity","geometric optics","birefringence","polarization mixing","light cone","constitutive tensor"],"falsifier":"A concrete calculation for a specific background—say, a Schwarzschild-like metric with a non-trivial torsion or non-metricity profile—that computes the full dispersion relation and the polarization transport matrix M would either confirm or rule out the claimed birefringence and mode mixing. If the tensorial piece B vanishes for all physically acceptable backgrounds, or if the omitted terms in the constitutive tensor cancel the polarization dependence, the central claim would be falsified.","tokens_in":12243,"feed_emoji":"🌗","tokens_out":1499,"duration_ms":14690,"temperature":0.7,"pith_summary":"This paper asks how light behaves when spacetime has not just curvature but also torsion and non-metricity—geometric properties that are absent in general relativity. The authors argue that, once the connection is allowed to be independent of the metric, there is no unique way to couple electromagnetism to geometry; they choose two guiding symmetries—standard U(1) gauge invariance and projective invariance—to pick out consistent couplings. They present two representative models: one where a scalar prefactor multiplies the Maxwell term, and one where a rank-four constitutive tensor acts as an anisotropic medium. In the geometric-optics limit, they show these couplings can modify the effective light cone, change the relation between field amplitude and intensity, induce polarization-dependent propagation, and generate birefringence and mode mixing. If correct, these results provide the formal basis for connecting metric–affine geometry to observable phenomena like black-hole images, birefringent lensing, and polarization patterns in the cosmic microwave background.","feed_headline":"Light keeps its gauge, but spacetime can bend it, split it, mix it","feed_subtitle":"Torsion and non-metricity modify the light cone, intensity, and polarization in the geometric-optics limit—opening new tests for metric-affi","key_machinery":"The key objects are the scalar prefactor Ξ (built from projectively invariant quadratic contractions of torsion and non-metricity) and the rank-four constitutive tensor χ^{μναβ} (with a minimal projectively invariant Ansatz). The geometric-optics limit is the machinery: the field is expanded as A^μ = a^μ e^{iωΨ}, and the equations of motion are expanded in powers of ω. At O(ω²) one obtains the dispersion relation and polarization eigenvalue problem; at O(ω) one obtains amplitude transport equations. The distinction between the scalar and tensorial parts of the polarization operator P^α_μ = δ^α_μ S + B^α_μ is what separates light-cone modification from birefringence.","core_discovery":"The central claim is that torsion and non-metricity can affect light propagation in several distinct ways, even in the geometric-optics limit, without breaking U(1) gauge invariance. In the scalar-coupled model (LΞ = -¼ Ξ F^2), the light cone is not deformed at leading order—the effect is a conformal rescaling—but the conserved intensity becomes I = Ξ A², so amplitude and intensity are no longer directly proportional. In the constitutive-tensor model (Lχ = -¼ χ F F), the leading-order GO equation becomes a polarization-dependent eigenvalue problem kμ H^{μν} = 0, leading to an effective inverse metric with anisotropic corrections and a tensorial piece that induces birefringence and mode mixin","pith_inferences":["The paper's symmetry principles (U(1) gauge and projective invariance) effectively single out the Levi-Civita Faraday tensor; one could test whether other gauge-preserving constructions (e.g., involving the non-metricity in the field strength) would introduce qualitatively different propagation effects.","The minimal Ansatz (38) for χ is explicitly acknowledged as not exhaustive; a full classification of admissible quadratic terms in T_f and Q_f might reveal that some effects (e.g., birefringence) are not generic but model-dependent, or might introduce additional polarization couplings.","The paper does not compute the explicit tensorial piece B; a natural extension is to compute it for a specific background (e.g., a spherically symmetric spacetime with torsion) and derive concrete shadow or lensing predictions, which would make the formalism falsifiable.","The intensity transport law I Σ = const along eigenmodes suggests that the ether drift of polarization eigenbases could be observable; one could propose a null test looking for differential photon flux between polarization modes in astrophysical sources."],"forward_implications":["If the constitutive-tensor model holds, torsion and non-metricity can make spacetime behave like an anisotropic optical medium, leading to two distinct photon speeds (birefringence) and a superposition of two images in black-hole observations.","The scalar-coupled model predicts a renormalization of the effective intensity I = ΞA² along a ray bundle, meaning that observations assuming standard photon-number conservation (e.g., luminosity distances) could be misinterpreted if torsion and non-metricity are present.","The ∇χ term acts as an effective current, implying possible energy exchange between photons and the non-Riemannian background, which could lead to departures from photon-number conservation.","Polarization mixing between transverse modes, governed by a matrix M in the transport equation, can cause depolarization of light from behind an accretion disk compared to the front, providing a testable asymmetry.","If χ breaks time-reversal invariance, co-rotating and counter-rotating photons could experience different optical paths, adding an asymmetry to black-hole shadows beyond frame dragging."],"fun_headline_variants":["Torsion and non-metricity bend light without breaking gauge","Light intensity and polarization shift in twisted spacetime","Gauge-safe light feels torsion and non-metricity","Torsion and non-metricity decouple light amplitude from intensity","Twisted spacetime: light bends, splits, and mixes, gauge intact"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The minimal constitutive-tensor Ansatz (38) is assumed to be rich enough to exhibit the claimed birefringence and mode mixing; the paper does not compute the explicit tensorial piece B (Eq. 58) or rule out that omitted admissible terms change or cancel these effects.","fun_headline_variants_meta":{"raw":{"variants":["Torsion and non-metricity bend light without breaking gauge","Light intensity and polarization shift in twisted spacetime","Gauge-safe light feels torsion and non-metricity","Torsion and non-metricity decouple light amplitude from intensity","Twisted spacetime: light bends, splits, and mixes, gauge intact"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001114,"raw_usage":{"total_tokens":4480,"prompt_tokens":753,"completion_tokens":3727,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":497,"completion_tokens_details":{"reasoning_tokens":3641}},"tokens_in":497,"tokens_out":3727,"duration_ms":23280,"temperature":1.0,"reasoning_tokens":3641,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T00:39:21.363259+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete calculation for a specific background—say, a Schwarzschild-like metric with a non-trivial torsion or non-metricity profile—that computes the full dispersion relation and the polarization transport matrix M would either confirm or rule out the claimed birefringence and mode mixing. If the tensorial piece B vanishes for all physically acceptable backgrounds, or if the omitted terms in the constitutive tensor cancel the polarization dependence, the central claim would be falsified.","supporting_citations":[],"review_version":1}