{"id":"ee05a0dc-2143-4f48-b122-61151f8a6222","arxiv_id":"1909.01248","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A two-field mimetic gravity model has two scalar degrees of freedom, not one, and the extra entropy mode is a ghost when the fields have opposite-sign kinetic terms.","lead":"This paper re-examines a two-field modified gravity model and shows it contains two scalar modes, one more than earlier work claimed. For one sign of the fields' kinetic terms, the extra mode is a ghost, so that version of the model is unstable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central 4-DOF claim rests on the unverified first-class status of {π_N, π_i, H_R, H_i}; the crucial matter bracket is only cited to Ref. [36], and the c=-1 square-root Hamiltonian is not globally real.","rationale":"The reader identified the same weakest assumption: the reduced Hamiltonian analysis assumes the eight constraints are first-class and delegates the key Poisson bracket calculation to Ref. [36]. My read agrees. The linear perturbation argument in Sec. 3 is self-contained and supports the two-scalar-DOF claim: the adiabatic sector has one physical DOF after correctly handling the Ṙ=0 equation, and the entropy sector contributes one propagating mode (a ghost for c=-1). The nonlinear Hamiltonian analysis is presented as the full confirmation, but its decisive step, Eq. (4.12), is asserted rather than computed. For c=-1 this is more than a bookkeeping gap: the reduced Hamiltonian is not a globally real function on phase space, so the standard Dirac counting procedure needs additional justification. This is a concrete, addressable gap rather than a demonstrated contradiction, so the appropriate verdict remains CONDITIONAL, as the reader already concluded.","tokens_in":11615,"tokens_out":10131,"duration_ms":101031,"concrete_test":"Perform a direct Dirac-Bergmann analysis of the action (2.9) without eliminating λ: keep (λ, π_λ) in the phase space, compute the full constraint chain for c=1 and c=-1, and classify each constraint. Separately, symbolically evaluate the matter contribution to {H[M],H[N]} using H_m from Eq. (4.7) with the exact square-root branch for c=-1. If the constraint chain yields exactly four first-class constraints and two second-class constraints (from the λ sector), or if the matter bracket equals D_m[h^{ij}(M∂_jN−N∂_jM)] on the constraint surface, the 4-DOF count is confirmed; otherwise the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the nonlinear Hamiltonian DOF count of Sec. 4. After eliminating λ, the system has 12 canonical pairs and the constraints π_N, π_i, H[N], D[N]. The count (2×12−2×8)/2 = 4 is valid only if all eight constraints are first-class. The decisive relation is Eq. (4.12), {H[M],H[N]} = D[h^{ij}(M∂_jN−N∂_jM)], whose matter-sector part is not demonstrated. The text after Eq. (4.9) refers to Ref. [36], but that reference covers the single-field mimetic theory; it does not cover the two-field matter Hamiltonian (4.7), let alone the c=-1 branch. For c=-1 the reduced Hamiltonian contains sqrt(π_φ^2 − π_ψ^2) (up to spatial terms), which is not real everywhere and is non-differentiable where π_φ^2 = π_ψ^2; a Poisson bracket involving this square root may fail to close or may acquire non-weak terms. If any of the four constraints is actually second-class, or if a secondary constraint was missed, the DOF count changes and the disagreement with [46] is not established at the full nonlinear level. The 'always ghost' conclusion for c=-1 inherits the same gap, since it assumes the reduced phase-space description is valid.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript revisits the two-field mimetic gravity model with shift symmetries, defined by the action (2.9). It performs a linear cosmological perturbation analysis in comoving gauge and finds two scalar degrees of freedom (DOFs) — an adiabatic mode with \\dot{R}=0 and a propagating entropy mode — contrary to Ref. [46], which claimed only one. It then attempts a full nonlinear Hamiltonian analysis, asserting eight first-class constraints and four DOFs in total. It also analyzes the c=-1 branch (opposite-sign kinetic terms in the mimetic constraint) and concludes that the entropy mode is always a ghost.","tokens_in":11860,"tokens_out":13898,"duration_ms":125986,"significance":"If the results hold, the paper corrects the DOF count in the two-field mimetic model and identifies a new ghost instability. The linear perturbation derivation in Sec. 3.3 is explicit, and the methodological criticism of Ref. [46] is well founded: substituting \\dot{R}=0 into the action before varying is invalid because it is a differential equation, not an algebraic constraint. The Hamiltonian reduction for the adiabatic mode is also internally consistent, and the c=-1 ghost conclusion is a concrete falsifiable prediction at the linear level. However, the nonlinear confirmation is incomplete: the first-class nature of the reduced constraints is asserted rather than demonstrated, and the c=-1 branch introduces a square-root Hamiltonian with possible non-differentiable points. The central claim is defensible on the linear side, but the nonlinear section needs substantial completion before the paper can be accepted.","major_comments":[{"comment":"The claim that the eight constraints {\\pi_N, \\pi_i, H_R, H_i} are all first-class is the load-bearing step for the nonlinear DOF count (2\\times12-2\\times8)/2 = 4, but the matter-sector Poisson brackets are not computed. In particular, {H_m[M], H_m[N]} in (4.12) is only asserted to equal D_m[h^{ij}(M\\partial_jN-N\\partial_jM)], with the calculation deferred to Ref. [36]. That reference treats the single-field mimetic model and does not cover the two-field matter Hamiltonian (4.7) or the c=-1 branch. Without this calculation, the first-class status of the reduced Hamiltonian constraint is unestablished, and the claimed nonlinear confirmation of the four-DOF result is not supported.","section":"Sec. 4, Eqs. (4.10)-(4.12)"},{"comment":"The reduction that eliminates \\lambda and \\pi_\\lambda treats the pair (\\pi_\\lambda, \\Phi_2) as second-class, but the subsequent constraint algebra (4.10)-(4.12) is computed with ordinary Poisson brackets. In a system with second-class constraints, the reduced Hamiltonian (4.7) must be used with Dirac brackets; the naive Poisson bracket algebra is not sufficient to establish that H_R and H_i are first-class. The paper should either compute the Dirac brackets or demonstrate that they coincide with the Poisson brackets on the relevant variables. In addition, for c=-1 the argument of the square root in (4.7) vanishes on a locus (e.g., where the spatial kinetic terms cancel), making the Hamiltonian non-differentiable there; the paper does not discuss whether the constraint algebra closes on that locus, which is directly relevant to the c=-1 ghost claim in Sec. 5.","section":"Sec. 4, Eq. (4.7)"}],"minor_comments":[{"comment":"The linear Hamiltonian reduction for the adiabatic mode imposes two constraints, \\pi_B\\approx0 and \\pi_R+2a^3k^2B\\approx0, and concludes that this sector contributes one DOF. This is correct, but only because the two constraints are second-class (their Poisson bracket is -2a^3k^2, nonzero for k\\neq0). Adding this one-line check would make the counting explicit and avoid a possible misreading as first-class constraints.","section":"Sec. 3.3, Eqs. (3.25)-(3.28)"},{"comment":"The notation \"\\delta \\dot{s}^2\" is ambiguous; it should be written as \\delta\\dot{s}^2 or \\dot{\\delta s}^2 to avoid confusion between the square of the time derivative and the time derivative of the squared perturbation.","section":"Sec. 3.3, below Eq. (3.17)"},{"comment":"The phrase \"the rest four primary constraints\" is imprecise: \\pi_N is one constraint and \\pi_i are three, and the secondary constraints H_R and H_i arise from requiring consistency of these primary constraints. Please rephrase for clarity.","section":"Sec. 4, before Eq. (4.8)"},{"comment":"There are minor typos: \"orignial\" in Sec. 1 should be \"original,\" and \"inlcude\" in footnote 3 should be \"include.\" Also, Eq. (2.12) appears to have a missing parenthesis in the derivative term.","section":"Sec. 2, footnote 3 and Sec. 1, paragraph 1"}],"recommendation":"major_revision","confidential_remarks":"The linear perturbation analysis appears sound and likely corrects the literature, so the paper should not be rejected. The nonlinear section, however, is not sufficient to confirm the four-DOF claim, as the matter-sector Poisson bracket is not computed and the Dirac-bracket issue is unaddressed. The missing pieces are calculations that are within the scope of a revision, so major revision is the appropriate recommendation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this paper does a real service by catching an actual mistake in Ref. [46]. The authors show that the two-field mimetic model has two scalar DOFs, not one: the curvature mode is non-propagating but it is a genuine degree of freedom, and the entropy mode propagates. Their complaint against [46] is right—substituting Ṙ=0 into the Lagrangian before varying is not legitimate because it is a differential equation, not an algebraic constraint. That point alone justifies the paper.\n\nWhat is new: the corrected DOF count, the explicit quadratic action and Hamiltonian derivation at linear order, and the c=-1 analysis that finds a ghost entropy mode. The linear perturbation section is the strongest part. It is explicit, the decomposition into adiabatic and entropy modes is clean, and the Hamiltonian treatment of the reduced quadratic action is consistent.\n\nWhere it gets soft: Section 4, the full nonlinear Hamiltonian analysis, is supposed to confirm the four-DOF count, but the decisive Poisson bracket algebra is not actually done. The paper says the matter-sector commutators are 'similar to' the single-field case and sends you to Ref. [36]. That reference covers the one-field theory; it does not cover the two-field matter Hamiltonian (4.7), and it certainly does not cover the c=-1 branch. So the claim that all eight constraints are first-class is an assumption, not a demonstrated result. If a secondary constraint is hiding there, the count changes.\n\nThere is also a regularity issue the authors do not address. For c=-1, the reduced Hamiltonian contains sqrt(π_φ² - π_ψ²), which is not real over the whole phase space and is non-differentiable where the argument vanishes. That matters for the Hamiltonian analysis and for the 'always' ghost conclusion, which is drawn from a linearized treatment. These concerns do not obviously break the central claim—the linear analysis is convincing on its own—but they are real gaps.\n\nBottom line: this is a paper that correctly identifies an error in a published result, gives a clear linear-order derivation of the right DOF count, and adds a new instability result. It deserves a serious referee. The referee should ask for the matter-sector Poisson brackets to be shown or for the nonlinear section to be flagged as provisional, and should ask the authors to address the square-root phase-space issue in the c=-1 case. I would bring it to a reading group for the [46] correction alone.","headline":"A careful correction of the DOF count in two-field mimetic gravity, with a solid linear-perturbation core and a nonlinear Hamiltonian section that leans too heavily on a cited single-field calculation.","tokens_in":12407,"tokens_out":1673,"would_cite":true,"duration_ms":14203,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The two-field mimetic gravity model contains two scalar degrees of freedom plus two tensor modes, not the single scalar mode claimed earlier, and with opposite-sign kinetic terms its entropy mode is a ghost.","keywords":["mimetic gravity","degrees of freedom","Hamiltonian analysis","cosmological perturbations","adiabatic mode","entropy mode","ghost instability","shift symmetry"],"falsifier":"Compute the full Poisson bracket $\\{H[M],H[N]\\}$ for the matter sector of the two-field reduced Hamiltonian without borrowing the single-field result; the four-degree-of-freedom count stands only if the bracket closes on the constraint surface with no tertiary constraints. Separately, for $c=-1$, check whether the reduced Hamiltonian $\\sqrt{\\pi_\\phi^2-\\pi_\\psi^2}$ is real on the allowed phase space; if not, the Hamiltonian analysis is incomplete for that branch.","tokens_in":11369,"feed_emoji":"🌌","tokens_out":8812,"duration_ms":84809,"temperature":0.7,"pith_summary":"The paper revisits the two-field mimetic gravity model with shift-symmetric scalars and argues that it contains two scalar degrees of freedom on top of the two tensor modes, for a total of four. This corrects an earlier study that reported only one scalar mode. The claim is established in two independent ways: quadratic perturbation theory around a cosmological background, and a full nonlinear Hamiltonian analysis that counts constraints. The paper also shows that when the two scalar kinetic terms enter the mimetic constraint with opposite signs, the entropy mode is a ghost, so the model is quantum-mechanically unstable.","feed_headline":"Two-field mimetic gravity has two scalar modes, not one","feed_subtitle":"A full Hamiltonian count finds an extra frozen curvature mode; opposite-sign fields make the entropy mode a ghost.","key_machinery":"The central device is the decomposition of the two scalar fluctuations into an adiabatic mode along the background trajectory and an orthogonal entropy mode, together with the Hamiltonian 3+1 split of the metric. Varying the Lagrange multiplier imposes $A=0$, removing the lapse perturbation and leaving the reduced quadratic Lagrangian from which the equations of motion are derived without substituting $\\dot{R}=0$ before varying the curvature perturbation $\\mathcal{R}$. At the nonlinear level the machinery is the reduced Hamiltonian density obtained by eliminating $\\lambda$, whose smeared Hamiltonian and momentum constraints are claimed to close as first-class constraints; that closure fixes the degree-of-freedom count.","core_discovery":"The paper claims that the two-field mimetic action does not reduce to a single propagating scalar. Decomposing the two scalars into adiabatic and entropy directions, the quadratic action gives a decoupled entropy mode that propagates with unit sound speed and a curvature perturbation that is frozen, $\\dot{R}=0$; both modes contribute phase-space degrees of freedom, so the scalar sector has two DOFs. In the full nonlinear reduced Hamiltonian, obtained after eliminating the Lagrange multiplier, the paper finds eight first-class constraints on a twelve-pair phase space and counts $(2\\times12-2\\times8)/2=4$ degrees of freedom. In the opposite-sign case $c=-1$, the entropy mode has a negative kinetic term and is a ghost.","pith_inferences":["If the four-degree-of-freedom count is right, the earlier one-scalar result likely came from substituting $\\dot{\\mathcal{R}}=0$ into the action before varying the curvature mode, a step the paper identifies as invalid; analogous premature substitutions may affect other multi-field mimetic models.","Adding more mimetic scalars would probably add one entropy mode per scalar while leaving the adiabatic curvature mode frozen, so the instability structure of the single-field theory in the presence of matter would persist.","The $c=-1$ ghost appears already at quadratic order around any background satisfying the constraint, so avoiding it would require adding higher-derivative or curvature-coupling terms beyond the minimal shift-symmetric action.","Because the reduced Hamiltonian for $c=-1$ contains a square root of $\\pi_\\phi^2-\\pi_\\psi^2$, a direct test is whether the physical phase space is restricted to the region where that quantity is nonnegative; if not, the Hamiltonian formulation itself may need revision."],"forward_implications":["At the background level the two mimetic fields still behave as pressureless dark matter, with $\\rho\\propto a^{-3}$.","The comoving curvature perturbation is frozen, $\\dot{\\mathcal{R}}=0$, exactly as in the single-field mimetic theory.","The entropy perturbation decouples from the curvature mode at linear order, is massless, and propagates with unit sound speed when $c=1$.","For $c=-1$ the entropy mode is a ghost, so the model is quantum-mechanically unstable in that branch.","The total degree-of-freedom count is four, one more than the earlier analysis concluded, which changes the counting of initial conditions for cosmological perturbations."],"supporting_citations":[{"why":"introduces the original mimetic dark matter construction from which this two-field model descends","marker":"[1]"},{"why":"proposes the two-field mimetic model and claims one scalar DOF, the conclusion this paper argues against","marker":"[46]"},{"why":"provides the single-field reduced Hamiltonian and Poisson algebra used for the key constraint brackets","marker":"[36]"},{"why":"shows mimetic gravity arises from a singular conformal transformation","marker":"[54]"},{"why":"further develops the non-invertible transformation origin of mimetic gravity","marker":"[55]"},{"why":"supplies the adiabatic/entropy decomposition used for the two-field perturbations","marker":"[60]"},{"why":"provides the total-Hamiltonian and smeared-constraint formalism for the nonlinear analysis","marker":"[65]"},{"why":"supplies the standard constrained-system counting formula for degrees of freedom","marker":"[66]"}],"fun_headline_variants":["Two-field mimetic gravity has two scalar modes, not one","Hamiltonian analysis reveals extra scalar in two-field mimetic","Opposite-sign mimetic fields cause ghost instability","Two-field mimetic: entropy mode ghostly when signs oppose"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the eight constraints of the reduced nonlinear theory are all first-class, with no hidden secondary constraints, and that the matter-sector Poisson algebra matches the single-field result to which the paper refers.","fun_headline_variants_meta":{"raw":{"variants":["Two-field mimetic gravity has two scalar modes, not one","Hamiltonian analysis reveals extra scalar in two-field mimetic","Opposite-sign mimetic fields cause ghost instability","Two-field mimetic: entropy mode ghostly when signs oppose"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000254,"raw_usage":{"total_tokens":1496,"prompt_tokens":802,"completion_tokens":694,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":418,"completion_tokens_details":{"reasoning_tokens":627}},"tokens_in":418,"tokens_out":694,"duration_ms":7017,"temperature":1.0,"reasoning_tokens":627,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:22:54.842499+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the full Poisson bracket $\\{H[M],H[N]\\}$ for the matter sector of the two-field reduced Hamiltonian without borrowing the single-field result; the four-degree-of-freedom count stands only if the bracket closes on the constraint surface with no tertiary constraints. Separately, for $c=-1$, check whether the reduced Hamiltonian $\\sqrt{\\pi_\\phi^2-\\pi_\\psi^2}$ is real on the allowed phase space; if not, the Hamiltonian analysis is incomplete for that branch.","supporting_citations":[{"cited_title":"Bojowald, Canonical gravity and applications: cosmology, black holes, and quantum gravity","cited_arxiv_id":null,"evidence_quote":"provides the total-Hamiltonian and smeared-constraint formalism for the nonlinear analysis"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the standard constrained-system counting formula for degrees of freedom"}],"review_version":1}