{"id":"008bd113-d1ea-49cd-8ab0-39741db17543","arxiv_id":"1908.02044","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The first law of entanglement, applied to a putative holographic dual of Minkowski spacetime, is shown to imply the linearized gravitational equations of motion in 3d and 4d, modulo assumed boundary stress-tensor conditions.","lead":"This paper tries to extend the holographic principle to flat spacetimes like our own, deriving gravity around Minkowski space from quantum entanglement. It shows that, under assumptions about a hypothetical boundary theory, the first law of entanglement is equivalent to Einstein's equations linearized around Minkowski spacetime in three and four dimensions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof assumes boundary stress-tensor conservation and trace conditions that are themselves the large-r components of the linearized Einstein equations, so the claimed equivalence passes part of the target equations in through the holographic input.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the boundary stress-tensor conservation and trace conditions are inputs that are themselves components of the linearized Einstein equations at asymptotic infinity, and C0 is left undetermined in 3d while the 4d analog is simply assumed. I agree with that assessment. I do not see a contradiction or a fatal mathematical error in the RT construction, the Wald-form derivation, or the isometry bookkeeping; the argument is coherent as far as it goes. The problem is logical status: in the AdS derivation, boundary stress-tensor conservation is an independent CFT Ward identity, but in this paper it is obtained from the flat limit of the residual AdS Einstein equations and is therefore not an independent flat-holographic input. This weakens the headline claim of equivalence, but it does not invalidate the conditional statement because the paper explicitly lists the stress-tensor assumptions and the conclusion is already framed conditionally. The concrete test would show whether the first law over an enlarged class of RT surfaces can close the gap on its own, which is the natural and decisive next step. Since the reader's verdict is CONDITIONAL and this concern supports that verdict rather than moving it, no change to the verdict is recommended.","tokens_in":35324,"tokens_out":7293,"duration_ms":85409,"concrete_test":"Perform the following analytic check in the 3d setting of Sec. 4.2: use the generalized light sheaf (3.23) with independent shifts Y+ and Y−, together with the 𝓁u ≠ 0 configurations of App. C, to build an enlarged family of first-law constraints ∫Σ_{Y+,Y−} dχ = 0. Differentiate these constraints with respect to Y+ and Y− exactly as done for Y in eqs. (4.20)–(4.23), and read off the resulting local equations for δEur, δEuφ, and δEuu. If these equations force C0 = C1 = C2 = 0 without invoking (5.12)–(5.13) or (5.29)–(5.30), the circularity is resolved and the holographic input is derivable from the first law alone. If they only reproduce constraints on δErr and δErφ, the theorem must be restated as 'first law + boundary asymptotic Einstein equations ⇒ bulk Einstein equations,' and the analogous 4d check should be performed with deformed watermelons.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central theorem is that the first law of entanglement, δS_A = δE_A in eq. (2.1), implies the linearized gravitational equations. In the 3d proof, after RT configurations and bulk isometries give δErr = δErφ = 0 (eqs. (4.29), (4.32)), the remaining components are fixed by the conservation identity ∇_a(δE^ab) = 0 only up to integration constants: δEur = C0(u,φ), δEφφ = −r²C0(u,φ), δEuφ = C2(u,φ)/r, and δEuu = C1(u,φ)/r. The paper then invokes the Carrollian stress-tensor conservation and trace equations (5.27)–(5.30) to set C0 = C1 = C2 = 0. But in the flat-space Bondi analysis of Sec. 5.2, those conservation equations are exactly the (u,u) and (u,φ) components of the linearized Einstein equations at leading order in r, eqs. (5.12)–(5.14), and the trace conditions come from the same flat limit of AdS. So the proof does not derive the full bulk equations from the first law alone; it assumes the boundary components of the very equations it aims to prove. The manuscript itself flags the gap: Sec. 4.2 states that C0 = 0 is expected from trace conditions but was not shown conclusively, and Sec. 6.3 states that in 4d 'we will assume that these boundary conditions ensure the vanishing of the components δEua at leading asymptotic order.' In 4d the Carrollian stress tensor and its conservation/trace identities are not even constructed. The result is therefore better stated as: first law plus asymptotic boundary Einstein constraints implies bulk linearized Einstein equations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an analog of the Ryu-Takayanagi prescription for asymptotically flat spacetimes in three and four dimensions, building on prior work by Jiang-Song-Wen and others. For a special class of boundary regions, the authors define gravitational entropy and energy in terms of Wald's Noether charge, and study the implication of the first law of entanglement δS_A = δE_A for linearized perturbations of Minkowski spacetime. The main claimed result is that this first law is equivalent to the linearized gravitational equations of motion. The proof proceeds by using the first law over a family of RT surfaces generated by bulk isometries and light-sheaf deformations to set certain components of δE_{ab} to zero, then invoking conservation equations and boundary stress-tensor constraints to kill the remaining integration constants. The paper also develops a Carrollian stress-tensor formalism for 3d flat space via the flat limit of AdS_3, and proposes a 4d generalization, with explicit computations for on-shell perturbations in Bondi gauge and a positivity constraint on allowed perturbations.","tokens_in":35796,"tokens_out":3746,"duration_ms":40576,"significance":"If the central claim were fully established, this would be a valuable extension of the 'gravity from entanglement' program to flat space, with several novel ingredients: a refined RT prescription featuring a light-sheaf choice, a derivation of Carrollian stress-tensor conservation laws from the flat limit, and explicit first-law checks for both 3d and 4d. The paper is careful and explicit in its analytic steps, and it is commendably transparent about the places where assumptions are needed (Sec. 4.2, Sec. 6.3). The Carrollian stress-tensor decomposition and the flat-limit analysis in Sec. 5 are useful contributions in their own right. However, the central theorem as stated in the abstract and Sec. 2 is stronger than what is actually proven: the derivation passes part of the target equations through the boundary stress-tensor conservation and trace conditions, and the 4d proof relies on an unproven assumption about boundary conditions. The paper is therefore best read as establishing that the first law, together with asymptotic boundary Einstein constraints, implies the full bulk linearized equations, not that the first law alone does.","major_comments":[{"comment":"The 3d proof is circular in a load-bearing way. After the light-sheaf and radial-translation arguments give δE_rr = δE_rφ = 0 (eqs. (4.29), (4.32)), the conservation equation ∇_a δE^{ab}=0 leaves the remaining components as δE_ur = C_0(u,φ), δE_φφ = −r²C_0, δE_uφ = C_2(u,φ)/r, and δE_uu = C_1(u,φ)/r. The paper then uses the Carrollian conservation equations (5.12)–(5.13) and trace conditions (5.29)–(5.30) to set C_1=C_2=0 and expects C_0=0. But Eq. (5.14) identifies C_1=0 and C_2=0 with the (u,u) and (u,φ) components of the linearized Einstein equations at leading order in r. Thus the proof assumes the large-r components of the very equations it aims to derive. Moreover, C_0=0 is not demonstrated; Sec. 4.2 explicitly states 'we have not been able to show it conclusively.' The abstract's claim of equivalence between the first law and the full linearized equations is therefore not established; the established statement is that the first law plus asymptotic boundary Einstein constraints implies the bulk equations.","section":"§4.2 and §5.2"},{"comment":"The 4d proof is conditional on an unproven and unconstructed boundary input. After reducing δE_rr and δE_rθ to zero, the remaining components are solved in terms of integration constants C_0(u,φ), C_1(u,θ,φ), C_2(u,θ,φ), C_3(u,θ,φ). The paper states 'we will assume that these boundary conditions ensure the vanishing of the components δE_ua at leading asymptotic order' and expects the trace condition to imply C_0=0, but no 4d Carrollian stress tensor, conservation equations, or trace identities are derived. Unlike the 3d case, where Sec. 5 provides explicit flat-limit equations, the 4d boundary conditions are purely assumed. This is a load-bearing gap: without these conditions, the first law only constrains a subset of the linearized Einstein equations. The conclusion in Sec. 7 that 'we have shown that the first law of entanglement is equivalent to the linearized gravitational equations of motion' is too strong for the 4d case.","section":"§6.3"},{"comment":"The corner-regulated prescription (3.48) is an additional input that is not derived from the five assumptions in Sec. 2. The paper motivates it by the discrepancy between δS_A=0 and δE_A≠0 for the on-shell perturbation (3.44), and argues that a smooth curve ~A_ε arbitrarily close to the cornered surface produces the correct first law. However, this is a new rule for computing the entropy, not a consequence of the stated assumptions, and the first-law equivalence δS_A=δE_A depends on it. The claim that the corner has a non-trivial contribution is supported only by an analogy with polar coordinates (footnote 6) and a formal limit argument, not by a systematic derivation from the presymplectic structure. Since the central theorem relies on this prescription, it should either be derived from the assumptions or explicitly incorporated as a sixth working assumption; the current status is a gap between the stated axioms and the proof.","section":"§3.3–3.4"}],"minor_comments":[{"comment":"Footnote 1 cites reference [27] as 'To appear' with no author list or preprint number; this should be completed or removed before publication.","section":"Footnote 1"},{"comment":"The Bondi-gauge perturbation in Eq. (3.41) contains the term '−2r²U dudr', which appears to be a typo for '−2r²U dudφ'; the same pattern appears in the flat-limit metric (5.1).","section":"§3.3, Eq. (3.41)"},{"comment":"Eq. (4.31) displays an expression without an equality sign or a right-hand side; it should read '= 0' for the constraint following from the radial translation.","section":"§4.2, Eq. (4.31)"},{"comment":"The derivation of the 4d light-sheaf parametrization (6.38) is quite dense; the claim that modular-flow tangency forces each light sheaf to intersect the bifurcation surface at a single point (Sec. 6.2, after Eq. (6.37)) would benefit from an explicit algebraic derivation, as the current text leaves room for doubt.","section":"§6.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest about its two main gaps, and the reframed theorem ('first law + asymptotic boundary Einstein constraints implies bulk linearized Einstein equations') is still a substantial and publishable result, especially given the novel Carrollian stress-tensor analysis and the explicit light-sheaf RT construction. I therefore recommend major revision rather than rejection: the authors should either close the 3d C_0 gap and derive the 4d boundary conditions, or clearly restate the central claim with the boundary conditions as part of the hypothesis. I would also ask them to check the concurrent-work footnote and to correct the minor typographical issues noted above."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nI read Godet–Marteau (arXiv:1908.02044) carefully, and the stress-test note is right: it lands on the central theorem. Here is my take.\n\nWhat is new: the authors refine the existing 3d RT prescription by making the light-sheaf ambiguity explicit (Sec. 3.2), propose a genuine 4d generalization with 'watermelon' regions and RT surfaces (Sec. 6.1–6.2), and show that, under their five working assumptions plus boundary stress-tensor input, the first law δS_A = δE_A is equivalent to the linearized Einstein equations around Minkowski. The 4d construction is a real step forward. The flat-limit derivation of the Carrollian stress tensor in Sec. 5 is also solid: eqs. (5.12)–(5.13) and (5.29)–(5.30) are carefully obtained from the AdS3 limit. Credit is due: they state Assumptions 1–5 plainly and explicitly flag the C0 gap in Sec. 4.2 and the assumed 4d boundary conditions in Sec. 6.3.\n\nThe soft spot is not minor. In 3d, after the first law gives δErr = δErφ = 0, the conservation identity leaves C0, C1, C2 undetermined. The paper kills C1 and C2 with eqs. (5.12)–(5.13), which are the leading large-r (u,u) and (u,φ) components of the linearized Einstein equations (5.14). The trace conditions (5.29)–(5.30) are expected to kill C0, but the authors admit they could not prove it. So the derivation is not 'first law alone ⇒ Einstein'; it is 'first law + boundary Einstein constraints ⇒ bulk Einstein.' In 4d the situation is weaker: the analogous boundary conditions are simply assumed, and no Carrollian stress tensor is constructed. That is a load-bearing gap, not a cosmetic one. The abstract's phrase 'under reasonable assumptions on the holographic stress tensor' is doing real work.\n\nStill, this is a valuable structural result. It identifies exactly which boundary data must supplement the first law to recover bulk dynamics, and the 4d RT construction is worth having. The paper is honest about its gaps; I would not call it circular in a deceptive sense, but the theorem should be restated as conditional on the boundary Einstein constraints. For people working on flat-space holography or Carrollian CFTs, this is a paper to read and to build on. I would want to cite it for the 4d prescription. My recommendation: send it to a serious referee, but expect a major revision—either remove the circularity or state the theorem with the boundary conditions as explicit assumptions.","headline":"A genuine flat-space extension of gravity-from-entanglement, but the central equivalence is conditional: the proof imports boundary components of Einstein's equations as 'holographic' input, and one 3d integration constant is never shown to vanish.","tokens_in":36222,"tokens_out":4323,"would_cite":true,"duration_ms":46753,"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":"Under holographic assumptions, the first law of entanglement is equivalent to linearized gravitational equations around Minkowski spacetime in three and four dimensions.","keywords":["flat holography","entanglement entropy","Ryu-Takayanagi prescription","first law of entanglement","Carrollian geometry","Minkowski spacetime","linearized gravity","BMS symmetry"],"falsifier":"Take a linearized perturbation of 3d Minkowski with δErr = δErφ = 0 but δEur = C0(u,φ) and δEφφ = −r²C0(u,φ), and check whether the boundary trace conditions (5.29)–(5.30) permit C0 ≠ 0; if they do, the paper's conclusion δEur = 0 fails while its stated assumptions still hold, which is exactly the gap the paper flags in Sec. 4.2.","tokens_in":35091,"feed_emoji":"🌌","tokens_out":6362,"duration_ms":64525,"temperature":0.7,"pith_summary":"This paper works within flat-space holography, where gravity in asymptotically flat spacetime is dual to a quantum system living on null infinity, and entanglement entropies of certain boundary regions are computed by an analog of the Ryu-Takayanagi formula. For linearized perturbations of three- and four-dimensional Minkowski spacetime, it claims that the first law of entanglement, the equality between entropy change and modular-energy change, holds if and only if the linearized gravitational field equations hold in the bulk. This matters because it extends to flat space the AdS/CFT result that gravity emerges from entanglement thermodynamics, without needing a cosmological constant or an AdS boundary. The cost is a set of assumptions about the dual theory, and in particular boundary stress-tensor conservation and trace conditions that are the only remaining holographic input.","feed_headline":"Entanglement first law yields flat-space Einstein equations","feed_subtitle":"In 3d and 4d Minkowski, the entanglement first law plus boundary stress-tensor conservation gives linearized gravity.","key_machinery":"The central object is the generalized Rindler transformation: a symmetry of null infinity that maps the domain of dependence of a boundary region A to a spacetime with an imaginary-time circle, so that the generator of the circle is the modular flow. Its bulk extension is a Killing vector of Minkowski spacetime that vanishes on the Rindler bifurcation surface, and the RT surface is the union of an infalling light sheaf with the portion of the bifurcation surface it bounds. The identity dχ = −2ξ^a δE_ab ε^b carries the argument: integrating χ between the boundary region and the RT surface turns the first law into vanishing integrals of components of the equations of motion, and bulk isometries generate enough such integrals to force those components to zero. The remaining constants are killed by Carrollian boundary stress-tensor conservation and trace conditions, which the paper derives in three dimensions from the flat limit of AdS and assumes in four dimensions.","core_discovery":"The central claim is that, for linearized perturbations of Minkowski spacetime, the gravitational equations of motion are equivalent to the first law δS^grav_A = δE^grav_A for all boundary regions A in a special class. In three dimensions, the proof uses RT surfaces made of two light rays and a curve on the Rindler bifurcation surface; variations of interval size, bulk isometries, and deformations of the infalling light sheaf force δErr = δErφ = 0, and the remaining components vanish once the conservation equation is combined with Carrollian boundary stress-tensor conditions. In four dimensions, the entangling regions are generalized watermelon slices and the same strategy yields δErr = δErθ = 0 everywhere, after imposing the analogous boundary stress-tensor conditions at leading asymptotic order. The result is phrased for general theories of gravity through Wald's Noether charge, so it covers higher-derivative corrections to Einstein gravity.","pith_inferences":["Editorial inference: the admitted three-dimensional gap means the cleanest honest reading is that first law plus boundary Einstein equations implies bulk Einstein equations; if the boundary conditions are regarded as part of the holographic dictionary rather than as consequences of bulk dynamics, the result still stands as an equivalence.","Editorial inference: the light-sheaf dependence of the entropy suggests that flat-space entanglement entropy is inherently cutoff dependent, so quantitative matches with conjectured dual theories should specify the sheaf; this may be a feature rather than a bug.","Editorial inference: one could test the positivity constraints directly by computing δE_A for explicit Bondi perturbations and asking whether the relevant modular-flow integral can become negative for some region A; a negative value would not refute the paper but would tell which perturbations admit a consistent modular Hamiltonian.","Editorial inference: the four-dimensional proof uses perturbations with vanishing gravitational-wave aspect Cij, so closing the remaining gap would involve checking whether the first law also forces Cij = 0 or whether the equivalence extends to those modes as well."],"forward_implications":["If correct, linearized gravity around Minkowski needs no cosmological constant or AdS boundary to emerge from entanglement; the boundary dual is Carrollian and lives on null infinity.","The derivation applies to any diffeomorphism-invariant theory of gravity because both entropy and energy are defined through Wald's Noether charge, so it covers higher-derivative corrections.","The RT prescription is not unique: the freedom in choosing the infalling light sheaf changes the entropy by a regulator-dependent amount, which the paper interprets as a choice of UV cutoff in the dual theory.","Positivity of von Neumann entropy imposes constraints on allowed perturbations, selecting a code subspace on which the modular Hamiltonian is bounded below.","In four dimensions the equivalence holds for entangling regions shaped like deformed watermelon slices rather than spheres, showing the first-law logic does not rely on conformally flat boundary geometry."],"supporting_citations":[{"why":"Supplies the original flat-space analog of the Ryu-Takayanagi prescription that this paper refines and extends to four dimensions.","marker":"[25]"},{"why":"Provides the AdS derivation that gravitational equations follow from the first law of entanglement, the strategy adapted here.","marker":"[19]"},{"why":"Gives the conformal transformation and Rindler method that motivates the generalized Rindler transformation.","marker":"[26]"},{"why":"Provides the flat limit of AdS3 in Bondi gauge used to identify the Carrollian stress tensor and its conservation equations.","marker":"[6]"},{"why":"Independent field-theoretic computation in flat holography that the 3d RT entropy formula reproduces.","marker":"[24]"},{"why":"Provides the asymptotic BMS solution class used for explicit on-shell perturbations and energy variations.","marker":"[10]"},{"why":"Supplies the 4d asymptotically flat Bondi-gauge perturbation and asymptotic charges used in the 4d energy computation.","marker":"[43]"},{"why":"Shows the flat RT surface corresponds to an extremal surface, supporting the light-sheaf picture.","marker":"[32]"}],"fun_headline_variants":["Entanglement first law equates to linearized Einstein equations","Entanglement first law gives linearized gravity in flat spacetime","Minkowski gravity from the entanglement first law","Entanglement first law yields linearized Einstein gravity","Flat-space Einstein equations from entanglement first law"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the boundary of the putative dual theory obeys stress-tensor conservation and trace conditions; in three dimensions these conditions are themselves large-radius components of the linearized Einstein equations, so the derivation of those components is not independent of what it aims to prove.","fun_headline_variants_meta":{"raw":{"variants":["Entanglement first law equates to linearized Einstein equations","Entanglement first law gives linearized gravity in flat spacetime","Minkowski gravity from the entanglement first law","Entanglement first law yields linearized Einstein gravity","Flat-space Einstein equations from entanglement first law"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000829,"raw_usage":{"total_tokens":3555,"prompt_tokens":813,"completion_tokens":2742,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":429,"completion_tokens_details":{"reasoning_tokens":2668}},"tokens_in":429,"tokens_out":2742,"duration_ms":62583,"temperature":1.0,"reasoning_tokens":2668,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:55:31.208362+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a linearized perturbation of 3d Minkowski with δErr = δErφ = 0 but δEur = C0(u,φ) and δEφφ = −r²C0(u,φ), and check whether the boundary trace conditions (5.29)–(5.30) permit C0 ≠ 0; if they do, the paper's conclusion δEur = 0 fails while its stated assumptions still hold, which is exactly the gap the paper flags in Sec. 4.2.","supporting_citations":[],"review_version":1}