{"id":"e6bd5a71-b02d-4ae9-8b7a-afe3e3ea99a7","arxiv_id":"2608.04461","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Inserting tanh((r-r_h)/delta) into the inverse radial BTZ metric produces a black bounce whose throat sits exactly at an extremal null horizon, with no Lorentzian-to-Riemannian signature change.","lead":"A deformed BTZ black hole metric with a smooth transition at the horizon is shown to extend into a Lorentzian black bounce with a degenerate null throat, not a signature change. The paper gives explicit curvature, energy conditions, entropy, and an Aretakis instability for this geometry.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The analytic black-bounce construction is well supported, but the Aretakis growth rate in Eq. (25) is conditional on an unproved settling assumption, so the instability claim as stated is not yet established for generic data.","rationale":"Reading the paper in good faith, the analytic extension is carefully constructed and the geometric black-bounce claims are well supported by explicit coordinate transformations, determinant checks, Christoffel regularity, and geodesic integration. The main risk is precisely the one the Reader identified: the linear Aretakis growth is derived under an explicit settling assumption that is not proven and is checked numerically for only one initial-data family on a finite domain. The exact conservation of H0 is independent of settling, so the instability claim reduces to a weaker statement if settling fails. This concern is real but secondary to the central geometry; it does not overturn the paper, but it justifies a conditional verdict rather than full acceptance of the Aretakis claim as stated. I see no basis for rejection: the effective-source and entropy arguments are explicitly kinematic or consistency checks, and the endpoint classification issue mentioned by the Reader is not the decisive obstacle here. The proposed test would settle whether the settling condition is generic and thus whether Eq. (25) should be presented as a theorem or as a conditional numerical observation.","tokens_in":16015,"tokens_out":52291,"duration_ms":457305,"concrete_test":"Run the L=0 scalar evolution (21) with at least three independent smooth initial-data families, including compactly supported data with support away from the throat and non-compact data with different decay rates, on a much larger domain (for example q∈[0,20]) with an absorbing or transparent outer boundary. For each family with H0≠0, monitor ∂vψ|q=0 over extended advanced time v and compare the late-time slope of ∂²qψ|q=0 with -rh/(ℓ²√δ) H0. If any family with non-vanishing H0 fails to settle, so that ∂vψ|q=0 does not tend to zero, and the slope of ∂²qψ|q=0 deviates from the predicted value or becomes non-linear, then Eq. (25) is not a generic statement and the Aretakis-growth claim must be weakened accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The core geometry is solid: the coordinate change r-rh=q^2 with signed root sigma(q) is real-analytic, the q-chart metric is non-degenerate and Lorentzian at q=0, and the two isometric sheets are a genuine black-bounce structure. The load-bearing risk lies in Sec. 7.2. Eq. (21) gives the exact conservation law H0 = ∂qψ|q=0, which is rigorous. However, Eq. (24) for ∂v(∂²qψ)|q=0 contains the source term -∂vψ|q=0/rh. The linear-growth prediction (25) follows only under the unproved condition stated in Sec. 7.2: 'If ψ settles to a constant on the throat at late advanced time, ∂vψ|0 → 0.' The abstract and conclusions present linear growth of ∂²qψ|q=0 as an unconditional feature of the throat, but no proof of settling is given. The numerical confirmation uses a single initial-data family, ψ(0,q)=q exp[-q²/(2(0.3)²)], on a finite q-interval [0,1.1] with an unspecified outer boundary treatment. If generic admissible data do not settle on the throat, then ∂²qψ|q=0 need not grow linearly; only conservation of H0 remains. Since the Aretakis-type instability is part of the headline claim, this conditional step is the weakest load-bearing point, even though it does not affect the analytic-extension or second-sheet results.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the static, circularly symmetric BTZ deformation (2), in which a tanh transition function multiplies the inverse radial metric component, g^{rr}=tanh((r-r_h)/δ)F(r). The central claim is that the naive Lorentzian-to-Riemannian signature-change interpretation is incorrect: in coordinates r-r_h=q^2 and with an advanced time v, the metric extends real-analytically across q=0 as a regular, Lorentzian, degenerate Killing horizon, beyond which lies a second isometric exterior; the areal radius bounces at the throat. The paper also gives the curvature and effective source in closed form, analyzes energy conditions, proves analytic regularity of the geodesic crossing, identifies the near-throat AdS_2×S^1 geometry, computes the throat entropy by three routes, proves (or claims to prove) strict positivity of the scalar effective potential, and derives an Aretakis-type instability with a conserved first transverse derivative and a linearly growing second derivative. It also records a negative result: smoothing g_{tt} instead is singular at the horizon for every finite width. The paper is explicit about several limitations, including the absence of a first law, the kinematic character of the model, and the deferred construction of the maximal extension.","tokens_in":16315,"tokens_out":36534,"duration_ms":298402,"significance":"The analytic-extension construction is the paper's strongest contribution: the coordinate transformation and signed-root argument in Sec. 2 and Appendix A are explicit, the q-chart metric is non-degenerate and Lorentzian at q=0, and the existence of a second isometric exterior is established by real analyticity. This part is convincing and likely to be useful to the black-bounce and signature-change community. The closed-form effective source, the energy-condition accounting, and the entropy concordance (minimal-surface length, Wald--Noether charge, and a Cardy estimate explicitly labeled a consistency check) are also valuable and are presented with unusual honesty. The scalar-sector claims are less robust: the derivation of the effective potential in Appendix C contains an algebraic inconsistency, and the Aretakis linear-growth statement in Sec. 7.2 is conditional on an unproved settling assumption. These issues affect load-bearing parts of the mode-stability and instability headlines, even though the core geometric results are not in question.","major_comments":[{"comment":"The reduction from Eq. (C1) to Eq. (C2) is not correct as written. With A=√(SF), dz=dr/A, substituting into (C1) does not produce (1/r)∂_z(r∂_z R). For the representative case S=1, F=r^2/ℓ^2, the exact transformed equation is (r^2/ℓ^2) R_zz + (4r^2/ℓ^3) R_z + (ω^2ℓ^2/r^2) R=0, whereas Eq. (C2) gives (r^2/ℓ^2) R_zz + (3r^2/ℓ^3) R_z + (ω^2ℓ^2/r^2) R=0; the coefficient of R_z differs. In addition, the appendix defines A=√(SF) but then uses A^2=SF^2 when computing AA′, so the printed computation of ∂_z^2√r does not correspond to the stated tortoise coordinate. Consequently Eq. (20) is not the actual effective potential for the scalar field in the stated coordinates, and the proof that V_L>0 for every mode is not valid as written. This is load-bearing for the mode-stability claim in Sec. 7.1 and must be corrected, or the claim must be withdrawn or qualified.","section":"Appendix C, Eqs. (C1)–(C2) and Eq. (20)"},{"comment":"The statement that the AdS endpoint is limit-circle appears incorrect. For the potential (20) as given, V_L∼3r^2/(4ℓ^4) as r→∞; in the standard tortoise coordinate for BTZ, x≈−ℓ^2/r, this is V_L∼3/(4x^2), which is the borderline limit-point case (one independent solution behaves as x^{-1/2} and is not square-integrable near x=0), not the limit-circle case. If the effective potential is corrected, the divergence at the AdS boundary remains, and the classification should be re-examined. The discussion of which self-adjoint extension must be chosen, and the claim that the Dirichlet (Friedrichs) extension is forced by normalizability, therefore need to be revisited.","section":"Sec. 7.1, endpoint classification"},{"comment":"The linear-growth formula (25) is derived from Eq. (24) under the explicitly unproved assumption that ∂_vψ|_0→0 on the throat at late advanced time. The exact conservation of H_0=∂_qψ|_0 is rigorous, but the linear growth of ∂_q^2ψ|_0 is conditional on this settling assumption. The abstract and conclusions present the linear growth as an unconditional property of the throat. The numerical confirmation uses a single initial-data family, ψ(0,q)=q exp[−q^2/(2(0.3)^2)], on the finite interval q∈[0,1.1], with no statement of the outer boundary condition or a convergence study. The authors should either prove the settling for a suitable class of admissible data or explicitly state the result as conditional, describing the numerical run as an illustrative example rather than generic evidence.","section":"Sec. 7.2, Eqs. (24)–(25), and Fig. 6"}],"minor_comments":[{"comment":"At δ=0.1 the expansion is quoted as σ=√10 q − (50√10/3) q^5 + O(q^7), but the series (A4) has no q^7 term; the next nonzero term is O(q^9). This is a harmless typo but should be fixed.","section":"Appendix A, after Eq. (A4)"},{"comment":"Please specify the outer boundary condition used in the finite-difference evolution of Eq. (21) and report a convergence test, since boundary reflections could affect the late-time settling of ∂_vψ|_0 on a finite q-interval.","section":"Sec. 7.2 and Fig. 6"},{"comment":"The abstract says the entropy is 'reproduced independently' by the Wald–Noether charge and by a Cardy estimate, but Sec. 6(iii) and Sec. 8 explicitly label the Cardy route as a consistency check rather than a derivation. The abstract should carry the same caveat to avoid overstating the microscopic status of the result.","section":"Abstract and Sec. 6(iii)"}],"recommendation":"major_revision","confidential_remarks":"The geometric core of the paper—the analytic extension, the degenerate null throat, the bounce structure, and the entropy concordance—is strong and well presented. My main concern is the scalar-sector analysis: the derivation of the effective potential in Appendix C contains a clear algebraic inconsistency, and the Aretakis linear-growth claim in Sec. 7.2 is conditional on an unproved settling assumption that is presented as a theorem in the abstract and conclusions. These issues are fixable within the manuscript's scope, but they are load-bearing for the mode-stability and instability claims, so I cannot recommend acceptance until they are addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The first thing to know: the core construction is sound and genuinely new. The coordinate change r-rh=q^2 with the signed root sigma turns the apparent signature-change chart into a smooth Lorentzian metric, the throat is a regular degenerate null horizon, and a second isometric exterior appears. I read the appendices carefully; the analytic continuation, geodesic regularity, closed-form source, and entropy computations all hold together. The negative result for smoothing gtt (finite-width singularity) is a useful caution for that whole ansatz family.\n\nThe main soft spot is the Aretakis claim. The conservation of H0 = ∂qψ|q=0 is exact and follows immediately from the equation for L=0. But the linear growth of ∂²qψ|q=0, Eq. (25), depends on the phrase 'If ψ settles to a constant on the throat at late advanced time, ∂vψ|0 -> 0' — a condition that is stated once and then dropped. The abstract and conclusions present linear growth as an unconditional feature. The numerics use one initial-data family, q exp(-q²/2(0.3)²), on a finite q-interval with the outer boundary handling unspecified. That is evidence, not proof, that generic data settle. As far as I can tell the instability reduces to 'H0 is conserved' for any smooth L=0 initial data; the linear growth needs a separate argument. A referee should ask for either a settling proof for the relevant class or a careful qualifier on the claim. This is a real flaw in presentation, not in the geometry.\n\nOne more thing to check: Sec. 7.1 calls the AdS endpoint limit-circle. I'd verify that against Ishibashi-Wald before relying on it; with a growing potential in tortoise coordinates, the endpoint classification is subtle. It's not load-bearing for the bounce construction, but it's the kind of claim that will get flagged.\n\nWho gets value: black-bounce and near-horizon holography people, and anyone studying extremal horizon instabilities in 2+1 dimensions. It deserves a serious referee: the core is careful, the derivations are explicit, and the construction is likely correct. I would not desk-reject. But I would not accept without revision either; the Aretakis overstatement and the endpoint classification need to be fixed or qualified.","headline":"Solid black-bounce construction with a rigorous analytic extension; the Aretakis instability headline overreaches because the linear growth rests on an unproved settling assumption.","tokens_in":16853,"tokens_out":5099,"would_cite":true,"duration_ms":43624,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["04.70.Bw","04.70.Dy","04.60.Kz","04.20.Jb"],"model":"deepseek-v4-flash","headline":"A smoothed BTZ horizon is a black bounce, not a signature change","keywords":["BTZ black hole","black bounce","degenerate Killing horizon","AdS2 x S1 throat","Wald entropy","Aretakis instability","signature change","lower-dimensional gravity"],"falsifier":"Evolve Eq. (21) from a family of smooth $L=0$ initial data with $H_0\\neq0$ for which $\\partial_v\\psi|_{q=0}$ does not decay; if the late-time slope of $\\partial_q^2\\psi|_{q=0}$ deviates from $-\\frac{r_h}{\\ell^2\\sqrt{\\delta}}H_0\\,v$, the claimed Aretakis growth rate is falsified.","tokens_in":15802,"feed_emoji":"🕳️","tokens_out":12854,"duration_ms":103393,"temperature":0.7,"pith_summary":"This paper establishes that inserting a smooth tanh transition function into the inverse radial component of the non-rotating BTZ metric, $g^{rr}=S_\\delta(r)F(r)$, does not produce a Lorentzian-to-Riemannian signature change at the horizon. In coordinates $r-r_h=q^2$ with an advanced null time, the metric extends real-analytically through $q=0$ to a second isometric exterior, so the would-be horizon becomes a degenerate null throat, a regular black bounce with vanishing surface gravity. The construction matters because it provides a fully controlled example of a BTZ black bounce whose bounce surface sits exactly at the horizon, with closed-form curvature, effective source, energy-condition, and entropy data, and because it exposes a generic obstruction: smoothing $g_{tt}$ instead is singular for every finite width. The paper also proves strict positivity of every scalar mode's effective potential and derives an Aretakis-type instability on the throat, with an exactly conserved leading transverse derivative and a linearly growing subleading one.","feed_headline":"Smoothed BTZ horizon becomes a black bounce, not a signature change","feed_subtitle":"The would-be horizon is a degenerate null throat with a second exterior, and probes on it can grow linearly.","key_machinery":"The machine that carries the argument is the square-root coordinate transformation $r-r_h=q^2$ together with the signed root $\\sigma=\\sqrt{S_\\delta}$ and the advanced time $dv=dt+dr/(F\\sigma)$. Because $\\tanh(y)=y\\,g(y)$ with $g$ analytic and $g(0)=1$, the root $\\sigma$ is analytic and odd in $q$, and the cross term $\\beta=4\\sqrt{\\delta}/\\sqrt{g(q^2/\\delta)}$ is analytic and nonzero at $q=0$; this cancels the would-be divergence of $dr^2$ exactly. The same analyticity makes the extension unique, makes all Christoffel symbols analytic near $q=0$ so that geodesics cross the throat analytically for every conserved charge, and turns the scalar wave equation into the exact form whose evaluation at $q=0$ yields the conserved Aretakis constant $H_0=\\partial_q\\psi|_{q=0}$ and the linear growth law for $\\partial_q^2\\psi|_{q=0}$.","core_discovery":"At $r=r_h$ the original chart is singular, but the regular chart $r-r_h=q^2$, $dv=dt+dr/(F\\sigma)$ with $\\sigma=\\sqrt{S_\\delta}$, turns the metric into $ds^2=-F(r_h+q^2)\\,dv^2+\\beta(q^2)\\,dv\\,dq+(r_h+q^2)^2\\,d\\varphi^2$, with $\\beta(0)=4\\sqrt{\\delta}\\neq0$. This line element is real-analytic at $q=0$, so it has a unique analytic extension to $q<0$, where $r>r_h$ again. The surface $q=0$ is a null hypersurface, a degenerate Killing horizon with vanishing surface gravity, and the two exterior sheets are isometric. The areal radius has a strict minimum there, so the geometry is a member of the black-bounce family rather than a signature-changing spacetime; near the throat it is $\\mathrm{AdS}_2\\times S^1$, and the throat circle is a minimal surface.","pith_inferences":["If the settling assumption in Sec. 7.2 fails for generic smooth data, the instability reduces to exact conservation of $H_0$ without the linear growth rate; this is a direct numerical test across a wider space of initial data.","The rotating or charged generalisation, where a Kaluza–Klein gauge field makes Sen's entropy function nondegenerate, may be where a first law for this class of extremal throats can actually be formulated.","The same $r-r_h=q^2$ analytic-extension mechanism likely applies to any metric with $g^{rr}=S(r)F(r)$ where $S$ vanishes linearly at the zero of $F$, so the no-signature-change conclusion may hold for a whole class of smooth transitions, not only tanh.","If late-time waves probe the throat, the linearly growing $\\partial_q^2\\psi|_0$ may feed nonlinear or backreaction effects, and the exponentially decaying tails of the energy-condition violation make the transition shell effectively infinite; both are testable in time-domain evolutions beyond linear order."],"forward_implications":["If the central claim is right, the smoothing ansatz $g^{rr}=S_\\delta F$ does not connect the Lorentzian sector to a Riemannian interior; the spacetime consists of two isometric exteriors joined at a degenerate null throat.","At the throat all curvature invariants are finite, with $R=-r_h/(\\ell^2\\delta)$ and $K=r_h^2/(\\ell^4\\delta^2)$, so the family is a one-parameter regularisation whose curvature grows as $\\delta\\to0$ and which never converges smoothly to BTZ.","The throat entropy $\\pi r_h/2G$ is obtained independently from the minimal-surface length, the Wald–Noether charge, and a Cardy estimate using the Brown–York mass, although no first law exists because $\\kappa=0$.","The scalar effective potential is strictly positive for every angular mode, so under the Dirichlet (Friedrichs) extension there are no exponentially growing exterior test-scalar modes; the instability that does exist is the transverse-derivative, Aretakis type at the throat.","The alternative ansatz that smooths $g_{tt}$ instead of $g^{rr}$ produces a curvature singularity at the horizon for every finite smoothing width."],"supporting_citations":[{"why":"The Lorentzian–Euclidean Schwarzschild ansatz that this construction adapts and ultimately negates: smoothing $g_{tt}$ instead is shown singular.","marker":"[7]"},{"why":"Supplies the degeneracy-of-the-dual-metric distinction used to show the chart (2) is singular while the metric itself is regular.","marker":"[8]"},{"why":"Defines the BTZ background whose inverse radial component is deformed.","marker":"[9, 10]"},{"why":"Defines the black-bounce family and its characteristic energy-condition violations, to which this geometry belongs.","marker":"[12, 13]"},{"why":"Establishes that a BTZ black bounce was already known, so the new feature is the degenerate throat at the horizon.","marker":"[14]"},{"why":"Provide the covariant minimal-surface prescription used to assign the throat entropy.","marker":"[17, 18]"},{"why":"Give the Noether-charge formula used to reproduce the entropy.","marker":"[20, 21]"},{"why":"Supply the asymptotic central charge and quasilocal mass used in the Cardy check.","marker":"[22, 23]"},{"why":"Give the Cardy formula and its BTZ application used for the consistency check.","marker":"[24, 25]"},{"why":"Origin of the transverse-derivative instability mechanism that the throat analysis reproduces.","marker":"[29, 30]"}],"fun_headline_variants":["Smooth BTZ yields black bounce, no Lorentzian flip","Degenerate throat in BTZ bounce, Aretakis instability","No signature change: BTZ bounce has extremal null throat","BTZ bounce: extremal null throat, zero surface gravity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The headline instability rate rests on an unproved settling condition: the paper shows $\\partial_q\\psi|_{q=0}$ is exactly conserved, but converting Eq. (24) into linear growth of $\\partial_q^2\\psi|_{q=0}$ assumes that $\\partial_v\\psi$ tends to zero on the throat at late advanced time, a settling that is demonstrated only for one numerical initial-data family.","fun_headline_variants_meta":{"raw":{"variants":["Smooth BTZ yields black bounce, no Lorentzian flip","Degenerate throat in BTZ bounce, Aretakis instability","No signature change: BTZ bounce has extremal null throat","BTZ bounce: extremal null throat, zero surface gravity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000843,"raw_usage":{"total_tokens":3760,"prompt_tokens":1123,"completion_tokens":2637,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":739,"completion_tokens_details":{"reasoning_tokens":2566}},"tokens_in":739,"tokens_out":2637,"duration_ms":18242,"temperature":1.0,"reasoning_tokens":2566,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:44:07.853805+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evolve Eq. (21) from a family of smooth $L=0$ initial data with $H_0\\neq0$ for which $\\partial_v\\psi|_{q=0}$ does not decay; if the late-time slope of $\\partial_q^2\\psi|_{q=0}$ deviates from $-\\frac{r_h}{\\ell^2\\sqrt{\\delta}}H_0\\,v$, the claimed Aretakis growth rate is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The Lorentzian–Euclidean Schwarzschild ansatz that this construction adapts and ultimately negates: smoothing $g_{tt}$ instead is shown singular."},{"cited_title":"14 Not a signature-changing spacetime","cited_arxiv_id":null,"evidence_quote":"Supplies the degeneracy-of-the-dual-metric distinction used to show the chart (2) is singular while the metric itself is regular."},{"cited_title":"dv = 0 (ingoing) , dq dv = F β (outgoing)","cited_arxiv_id":null,"evidence_quote":"Establishes that a BTZ black bounce was already known, so the new feature is the degenerate throat at the horizon."}],"review_version":2}