{"id":"5eb0c038-2a0c-4536-9618-181bc8108206","arxiv_id":"2505.01038","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A self-contained review of black hole no-hair proofs in GR and scalar-tensor theories, including a new theorem that stationary axisymmetric multiscalar black holes with a semidefinite-positive Hessian potential must be Kerr with constant scalars.","lead":"This review collects the main no-hair theorems for black holes in general relativity and scalar-tensor gravity, with detailed proofs, and adds a new uniqueness result for rotating multiscalar black holes with a scalar potential. It gives researchers a single map of what is rigorously known about when black holes can carry scalar hair.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The new theorem in Sec. 5.5 is not proven as stated: with only a semi-positive definite Hessian, the vanishing of the integrand in Eq. (158) does not force the scalar gradients to vanish.","rationale":"The reader's weakest assumption is exactly the step I would flag: the passage from vanishing of the non-negative integrand in Eq. (158) to \\hat D_i phi^a = 0 requires positive definiteness, while the theorem only assumes semi-positive definiteness. I checked whether the scalar field equation or asymptotic conditions could close the gap. Asymptotic flatness gives D_a V(phi_infinity) = 0 for a scalar field tending to a constant, but D_a V = 0 on the whole domain is a conclusion, not an assumption. If the image lies in a critical submanifold of V, the potential is constant along the image and equal to its asymptotic value 0; the local field equations then reduce to the V = 0 system of Sec. 5.4. That suggests the theorem's conclusion may be true, but the paper does not supply this reduction, so the proof as written has a real gap. A secondary issue in Sec. 6 is also real: the theorem conditions (160)-(161) have F_K and phi F_phi with the same sign, whereas the proof of Eq. (163) needs opposite signs for a definite-sign integrand. I do not think either issue invalidates the review's expository content, which is substantial and largely standard, but the paper's stated new theorem should be regarded as conditional pending repair.","tokens_in":33848,"tokens_out":10458,"duration_ms":116076,"concrete_test":"Check the logical step by computing Eq. (158) for the flat-direction model: target space E^2 with flat metric, V(phi^1, phi^2) = (phi^1)^2, and a nonconstant factor-space map with phi^1 = 0, phi^2 = psi(rho,z) satisfying the same fall-off. The integrand vanishes identically while \\hat D_i phi^2 != 0, so the proof's inference is invalid. Then attempt the missing repair: prove that D_a V = 0 plus V(phi_infinity) = 0 implies V = 0 along the scalar image and apply the Sec. 5.4 V = 0 theorem; if this reduction cannot be carried out, the Sec. 5.5 theorem remains unproven.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central original claim is the theorem in Sec. 5.5. Its proof hinges on the inference after Eq. (158): the boundary terms vanish, the right-hand side is a sum of non-negative terms, so the integrand must vanish, hence '\\hat D_i phi^a = 0 and D_a V = 0'. The second conclusion follows from the |DV|^2 term. The first does not follow from the stated hypothesis. Semi-positive definiteness of D_b D_c V allows D_b D_c V \\hat D_i phi^b \\hat D_i phi^c = 0 while \\hat D_i phi^a != 0, whenever the gradients lie in the kernel of the Hessian, i.e. in a flat direction of V. Example: on E^2 with flat target metric and V(phi^1, phi^2) = (phi^1)^2, any nonconstant map with phi^1 = 0 has D_a V = 0 and zero Hessian term, yet \\hat D_i phi^2 != 0. Since V is not identically zero, this configuration is not excluded by the theorem's assumptions. The proof could perhaps be repaired by showing that D_a V = 0 plus asymptotic flatness forces V = 0 along the scalar image and then invoking the V = 0 Kerr theorem of Sec. 5.4, but that step is absent. As written, the theorem is unproven.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper is a review of no-hair and uniqueness theorems for stationary, asymptotically flat black holes in general relativity and in scalar-tensor theories. It presents detailed proofs of the GR uniqueness theorems for Schwarzschild and Kerr black holes, then surveys and reproves results for classical single-field and multiscalar scalar-tensor theories, noncanonical scalar fields, and shift-symmetric Horndeski theories. The paper also claims one new theorem, in Section 5.5, for stationary and axisymmetric black holes in multiscalar theories with nonzero potential and a semi-positive definite Hessian of the potential. The review is largely self-contained, and the authors state that their aim is both to collect known results and to demonstrate the underlying mathematical techniques.","tokens_in":34200,"tokens_out":11797,"duration_ms":125114,"significance":"If fully correct, the new theorem in Section 5.5 would be a substantive extension of the no-hair paradigm to rotating black holes in multiscalar theories with a class of nonzero potentials, and the review would provide a useful pedagogical survey of the available uniqueness techniques. The paper's strengths include its detailed, largely self-contained derivations of the classical GR uniqueness arguments via the positive mass theorem and the Mazur identity, and its explicit treatment of several nontrivial scalar-field cases. However, the central new claim is not proven as stated, and a theorem in Section 6 is stated with sign conditions that contradict the proof. These issues affect the reliability of the paper's original and review content, and they require correction before the manuscript can be accepted.","major_comments":[{"comment":"The inference after Eq. (158) that the vanishing of the integrated non-negative terms forces \\hat D_i\\varphi^a = 0 is not valid under the stated hypothesis. Semi-positive definiteness of D_bD_cV only gives D_bD_cV \\hat D_i\\varphi^b \\hat D_i\\varphi^c = 0; this allows nonzero scalar gradients lying in the kernel of the Hessian. For example, with a flat target metric and V(\\varphi^1,\\varphi^2)=(\\varphi^1)^2, any configuration with \\varphi^1=0 and nonconstant \\varphi^2 has D_aV=0 and a vanishing Hessian term, while \\hat D_i\\varphi^2 \\neq 0. The proof would need an additional argument, such as using D_aV=0 and asymptotic flatness to show that V vanishes on the scalar image and then invoking the V=0 theorem of Sec. 5.4, but no such step is supplied. The theorem is therefore not proven as stated.","section":"Sec. 5.5, Eq. (158)"},{"comment":"The sign conditions in the theorem do not match the proof. In Eq. (163) the integrand is \\partial_KF\\,\\nabla_\\mu\\varphi\\nabla^\\mu\\varphi - \\varphi\\partial_\\varphi F. If \\partial_KF>0 and \\varphi\\partial_\\varphi F\\ge 0, this is a difference of two nonnegative terms and can vanish for nontrivial configurations, so the vanishing of the integral does not imply that the scalar field is trivial. The argument requires opposite signs, namely \\partial_KF>0 with \\varphi\\partial_\\varphi F\\le 0, or \\partial_KF<0 with \\varphi\\partial_\\varphi F\\ge 0. As printed, the theorem's hypotheses are insufficient for the stated conclusion.","section":"Sec. 6, Eqs. (160)-(163)"},{"comment":"The theorem's conclusion k^2(\\varphi_0)=0 is not derived in the case \\omega=0. The integration of Eq. (132) forces P=0 and \\omega^2|k|^2=0; when \\omega=0, this imposes no condition on |k|^2. A constant scalar map at a point with V=0 but k^2\\neq 0 satisfies L_\\xi\\varphi=-\\omega k with \\omega=0 and reduces the field equations to vacuum GR, so such a configuration is not excluded by the stated hypotheses. The theorem should either assume \\omega\\neq 0 or drop the condition k^2(\\varphi_0)=0 from the conclusion.","section":"Sec. 5.3, theorem and Eq. (132)"}],"minor_comments":[{"comment":"In the second integrand of Eq. (149), the first term is printed as (\\partial_\\rho h_\\phi)^2 twice; it should be (\\partial_\\rho h_\\phi)^2 + (\\partial_z h_\\phi)^2.","section":"Sec. 5.4, Eq. (149)"},{"comment":"The Christoffel term in the reduced scalar equation should read \\gamma^a_{bc}(\\partial_\\rho\\varphi^b\\partial_\\rho\\varphi^c + \\partial_z\\varphi^b\\partial_z\\varphi^c); as printed, the index structure with \\gamma^a_{bc}\\partial_\\rho\\varphi^a\\partial_\\rho\\varphi^b is inconsistent.","section":"Sec. 5.4, Eq. (139)"},{"comment":"The boundary values of the twist potential on the axis are quoted as \\chi=\\pm J + O(\\rho^2), whereas Eq. (68) and the Komar integral give \\chi=\\pm 4J + O(\\rho^2); the factor does not affect the boundedness argument but should be made consistent.","section":"Sec. 3.2, after Eq. (77)"},{"comment":"The proof explicitly assumes that \\varphi\\to 0 at infinity, while the theorem only states that \\varphi is constant; because the authors note this limitation at the end of the proof, the theorem statement should either include the asymptotic condition or the proof should be extended to constants at other zeros of the potential.","section":"Sec. 6, proof of the theorem"},{"comment":"There are numerous typographical and grammatical slips, such as \"passting\" in Sec. 5.1, \"the filed equations\" in Sec. 3.1, and a duplicated coefficient in the paragraph containing Eq. (149); these should be corrected in a final pass.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper's main added value is the claimed new theorem in Section 5.5, and that theorem is currently unproved. The Section 6 sign-condition mismatch is also a substantive correctness issue in a stated theorem. Both problems appear repairable, so I recommend major revision rather than rejection. Given that this is a review article, the authors should also ensure that all reproduced theorems match their cited sources."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a review with one genuinely new result. The theorem in Sec. 5.5 extends Heusler's rotating no-hair theorem to multiscalar theories with a nonzero potential whose Hessian is positive semidefinite. The review part is solid: it gives a self-contained account of the GR uniqueness theorems (Bunting-Masood-ul-Alam, Mazur's identity) and the main scalar-tensor no-hair results, with proofs sketched at a level that is actually usable. This is the kind of review people will keep on their desks.\n\nThe soft spots are in the new material. The proof of Theorem 5.5 has a genuine gap. After integrating the divergence identity, the vanishing of the right-hand side of (158) only gives D_a V = 0 and D_b D_c V \\hat D_i phi^b \\hat D_i phi^c = 0. With a semidefinite Hessian, that second condition does not force \\hat D_i phi^a = 0—the gradients can lie in the kernel of the Hessian. The example V = (phi^1)^2 on a flat two-scalar target, with phi^1 = 0 and phi^2 nonconstant, satisfies all the stated assumptions and makes the integrand vanish without a constant scalar map. The theorem may still be true: since D_a V = 0 along the scalar image and V tends to 0 at infinity, the solution satisfies the V = 0 field equations, so the Sec. 5.4 theorem should apply. But that reduction is not in the paper, and the assertion as written is not derived. This is fixable, but the proof is incomplete.\n\nSection 6 has a smaller but real sign problem. The theorem states conditions (160)-(161) with phi ∂_phi F having the same sign as ∂_K F, but the proof requires opposite signs to make both terms in (162) non-negative. As stated, the theorem's hypotheses do not imply the conclusion; the inequalities need to be swapped.\n\nThe citation practice is fine; the authors cite their own numerical work for context, and the new theorem does not depend on it. The review claims are mostly faithful to the original proofs. For a reader wanting a map of the no-hair landscape, this is a good entry point. I would send it to a competent referee after a light revision: fix the 5.5 proof, correct the Section 6 sign, and it becomes a solid contribution.","headline":"A dependable, self-contained review with one new theorem in Sec. 5.5 whose proof has a fixable gap, plus a sign error in the noncanonical section.","tokens_in":34656,"tokens_out":5067,"would_cite":true,"duration_ms":50067,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["04.70.Bw","04.50.Kd"],"model":"deepseek-v4-flash","headline":"The review establishes that most known no-hair theorems for black holes in scalar-tensor gravity hold, and adds a new theorem: in classical multiscalar theories with a nonzero potential whose target-space Hessian is semi-positive…","keywords":["no-hair theorems","black hole uniqueness","scalar-tensor gravity","multiscalar theories","Kerr black holes","Horndeski gravity","scalar-Gauss-Bonnet gravity","stationary axisymmetric black holes"],"falsifier":"Consider a two-scalar theory with potential $V(\\varphi^1,\\varphi^2)=(\\varphi^1)^2$, so $D_bD_cV$ has eigenvalues $2$ and $0$ and is semi-positive definite, and numerically search for a stationary, axisymmetric, asymptotically flat, regular-horizon black hole whose $\\varphi^2$ profile is nonconstant outside the horizon and tends to a constant at infinity; such a solution would refute the theorem.","tokens_in":33651,"feed_emoji":"🕳️","tokens_out":12954,"duration_ms":120573,"temperature":0.7,"pith_summary":"This review argues that the black-hole classification problems of general relativity and of scalar-tensor gravity are largely governed by the same no-hair mechanism: under suitable conditions on the scalar potential and on the scalar fields' symmetries, the only stationary, asymptotically flat vacuum black holes are the Kerr or Schwarzschild solutions with a constant scalar field. The paper's own new contribution is a theorem in Section 5.5 covering rotating black holes in classical multiscalar theories with a nonzero potential: if the target-space Hessian of the potential is semi-positive definite everywhere, then every stationary and axisymmetric solution with symmetry-compatible scalar fields and a regular horizon must be Kerr with a constant scalar map. The review also demonstrates that the bulk of the known no-hair statements, including static spherical cases with time-dependent scalars and several shift-symmetric Horndeski subclasses, follow from the same family of techniques. A sympathetic reader would care because the result draws a boundary around where beyond-GR scalar hair can exist and reduces a large part of the black-hole uniqueness problem back to the Kerr family.","feed_headline":"The new theorem: rotating black holes with several scalars are Kerr","feed_subtitle":"When the scalar potential curves only upward, scalar hair cannot survive around rotating black holes.","key_machinery":"The engine of the argument is the target-space Hessian of the scalar potential, $D_bD_cV(\\phi)$, together with the divergence identity $$\\nabla_\\mu(D_a V(\\phi)\\nabla^\\mu\\phi^a) = D_bD_cV(\\phi)\\nabla_\\mu\\phi^b\\nabla^\\mu\\phi^c + \\frac{1}{4}\\$gamma^{{ab}}$D_aV(\\phi)D_bV(\\phi).$$ After dimensional reduction to the two-dimensional factor space of the stationary and axial Killing fields, this becomes an elliptic identity in which $\\rho = \\sqrt{-\\det \\Gamma_{IJ}}$ multiplies both sides; the boundary terms drop out because $\\rho$ vanishes on the horizon and the rotation axis and the potential decays at infinity, leaving a manifestly nonnegative bulk integral. The same toolbox recurs through the review: conformal doubling with factors $\\Omega_\\pm = \\frac{1}{4}(1\\pm N)^2$ combined with the positive-mass theorem for static cases, the Mazur identity for the vacuum Kerr uniqueness proof, a scaling argument on the mass function of scalar hair, and the conserved-current integral that kills the scalar charge in shift-symmetric Horndeski theories.","core_discovery":"The central claim, stated as a new theorem, is this: consider the vacuum field equations of the classical multiscalar theories of gravity with potential $V(\\phi)\\neq 0$, and assume the tensor $D_bD_cV(\\phi)$ is semi-positive definite everywhere on the target space $\\mathbb{E}^N$. Then every stationary and axisymmetric black hole solution with regular horizon and stationary, axisymmetric scalar fields, $L_\\xi\\phi^a = L_\\eta\\phi^a = 0$, consists of the Kerr solution and a constant scalar map with $D_aV(\\phi)=0$. The paper presents the proof by reducing the problem to a Riemannian one on the factor space of the two Killing fields, integrating a divergence identity weighted by $\\rho$, and concluding that the scalar gradients and the potential derivative must vanish; the remaining equations are the vacuum Einstein equations, whose unique stationary black hole is Kerr. Alongside this theorem the review maintains that the known single-scalar results, the static spherical multiscalar results, the $V=0$ rotating result, and the shift-symmetric Horndeski scalar-charge result all hold, and it gives the proofs in detail.","pith_inferences":["The flat-direction gap in the Section 5.5 proof suggests that the most likely place for a counterexample is a potential that is nonnegative but flat in some direction, where the Hessian condition cannot see the hair.","If one could close that gap, the same Hessian condition would give a clean convexity criterion separating Kerr-only multiscalar theories from theories that can host scalar hair.","The divergence-identity technique appears general enough to carry over to electrovacuum multiscalar theories or to higher-dimensional black holes with interval structure data, though the paper does not develop those extensions."],"forward_implications":["In every multiscalar vacuum theory whose potential has a semi-positive definite Hessian and at least one zero, a stationary axisymmetric black hole with symmetry-compatible scalar fields must be Kerr with a constant scalar field.","For the classical theories with a single scalar field, the no-hair theorem applies whenever $\\phi V'(\\phi)\\ge 0$ or $V''(\\phi)\\ge 0$, so any observed deviation from Kerr inside those classes would have to come from time-dependent scalars or from matter.","Static, spherically symmetric multiscalar black holes in the vacuum with $V(\\phi)\\ge 0$ are Schwarzschild with a constant scalar map, including the cases where the scalars are time-dependent along a target-space Killing flow with nonempty axis.","In shift-symmetric Horndeski theories with regular coupling functions, static spherical black holes have no nontrivial scalar profile and rotating circular black holes carry zero scalar charge, so hairy solutions require exceptional pole-type couplings or explicit time dependence.","The explicit scalar-Gauss-Bonnet scalarization branches show that no-hair theorems do not extend to every beyond-GR theory: multiple hairy branches and nonuniqueness appear once the potential or coupling allows a different effective structure."],"supporting_citations":[{"why":"the rotating multiscalar $V=0$ no-hair theorem whose factor-space and Gauss-law strategy Section 5.5 adapts to $V\\neq 0$.","marker":"[79]"},{"why":"the Mazur identity used in the vacuum Kerr uniqueness proof that the no-hair reductions fall back on.","marker":"[7]"},{"why":"one step of the classical Kerr uniqueness proof the review reproduces and extends.","marker":"[5]"},{"why":"the Robinson step completing the Kerr uniqueness theorem, the endpoint of several no-hair results.","marker":"[6]"},{"why":"the conformal-doubling construction with $\\Omega_\\pm$ used with the positive-mass theorem in static proofs.","marker":"[55]"},{"why":"Hawking's scalar no-hair integral argument adapted in Section 4 for single-field theories.","marker":"[61]"},{"why":"the scaling argument for static spherically symmetric multiscalar black holes with $V\\ge 0$.","marker":"[65]"},{"why":"the time-dependent multiscalar no-hair theorem for static spherical black holes.","marker":"[73]"},{"why":"the Horndeski no-hair result and the exceptional pole-type coupling that permits Gauss-Bonnet hair.","marker":"[83]"}],"fun_headline_variants":["Scalar hair dies: rotating black holes are Kerr after all","New theorem: no scalar hair for rotating black holes with convex potential","Multiscalar no-hair: Kerr is the only stationary black hole","Proof: rotating black holes in scalar-tensor theory are Kerr","When scalar potential curves up, black holes lose their hair"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing step is that the vanishing of the integrated nonnegative terms in Eq. (158) forces the scalar-field gradients to be zero; with only a semi-positive definite Hessian, those gradients could instead lie in a flat direction of $V$, and the paper gives no further argument excluding that possibility.","fun_headline_variants_meta":{"raw":{"variants":["Scalar hair dies: rotating black holes are Kerr after all","New theorem: no scalar hair for rotating black holes with convex potential","Multiscalar no-hair: Kerr is the only stationary black hole","Proof: rotating black holes in scalar-tensor theory are Kerr","When scalar potential curves up, black holes lose their hair"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00016,"raw_usage":{"total_tokens":1214,"prompt_tokens":906,"completion_tokens":308,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":522,"completion_tokens_details":{"reasoning_tokens":220}},"tokens_in":522,"tokens_out":308,"duration_ms":3419,"temperature":1.0,"reasoning_tokens":220,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:28:53.525041+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Consider a two-scalar theory with potential $V(\\varphi^1,\\varphi^2)=(\\varphi^1)^2$, so $D_bD_cV$ has eigenvalues $2$ and $0$ and is semi-positive definite, and numerically search for a stationary, axisymmetric, asymptotically flat, regular-horizon black hole whose $\\varphi^2$ profile is nonconstant outside the horizon and tends to a constant at infinity; such a solution would refute the theorem.","supporting_citations":[{"cited_title":"Heusler, Class","cited_arxiv_id":null,"evidence_quote":"the rotating multiscalar $V=0$ no-hair theorem whose factor-space and Gauss-law strategy Section 5.5 adapts to $V\\neq 0$."},{"cited_title":"Bunting and A","cited_arxiv_id":null,"evidence_quote":"the conformal-doubling construction with $\\Omega_\\pm$ used with the positive-mass theorem in static proofs."},{"cited_title":"Heusler, J","cited_arxiv_id":null,"evidence_quote":"the scaling argument for static spherically symmetric multiscalar black holes with $V\\ge 0$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the time-dependent multiscalar no-hair theorem for static spherical black holes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the Horndeski no-hair result and the exceptional pole-type coupling that permits Gauss-Bonnet hair."}],"review_version":1}