On the Natural Equivalence Between Canonical and Hilbert Energy Momentum Tensors via Noether's Theorem

Peir-Ru Wang

REVIEW 4 major objections 5 minor 12 references

On the Natural Equivalence Between Canonical and Hilbert Energy Momentum Tensors via Noether's Theorem

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Noether’s theorem, applied to gauge potentials as 1-forms, yields symmetric gauge-invariant canonical energy-momentum tensors and, in the vielbein formalism, the Einstein tensor, so the canonical and Hilbert tensors coincide.

desk verdict The advertised new derivation of Einstein's equations is circular and rests on a false structural assumption; the first half restates known results but rests on a flawed step. read the letter →

arxiv 2507.05780 v1 pith:VQDEOUEQ submitted 2025-07-08 gr-qc physics.class-ph

Peir-Ru Wang This is my paper · ORCID

classification gr-qcphysics.class-ph

keywords Noether'stheoremcanonicalenergy-momentumtensorHilbertgaugeinvariance1-formpotentialvielbeinformalismEinsteinfieldequationsPalatinivariation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Treating the gauge potential as a 1-form rather than a vector field changes how Noether’s theorem responds to spacetime translations: the paper shows that the normally troublesome term combines with the field strength, so the canonical energy–momentum tensor comes out symmetric and gauge-invariant for electromagnetism and Yang–Mills theory without any improvement terms. Extending the same variational logic to gravity through the vielbein and Palatini formalisms makes the Noether current for diffeomorphisms coincide with the Einstein tensor times the volume element. The paper then argues that the canonical and Hilbert energy–momentum tensors are naturally equivalent, and that Einstein’s field equations, including the cosmological constant, follow from translation symmetry alone. If correct, this gives a single Noether-based route to energy–momentum tensors for gauge fields and spacetime, removing one of the oldest discrepancies between the canonical and metric constructions.

What carries the argument

The load-bearing device is the complete variation ΔAν = δAν plus the Lie derivative of the 1-form A along the translation vector field, with the identical treatment of the spin connection in the gravity case. In the Noether calculation this Lie-derivative term supplies Aγ δx^γ,ν, and an integration by parts combines it with Aν,γ to form the field strength Fγν, so the canonical tensor naturally reads −(∂L/∂(∂μAν)) Fγν + δ^μ_γ L. The same maneuver in the vielbein/Palatini formulation turns derivatives of the spin connection into the curvature R^b_{cγν} and finally into −$g^{{μα}}$G_{αγ}√−g. A second mechanism, given in Section VI, identifies the metric variation of the action with the canonical tensor through T^μ_γ = $g^{{μα}}$(−2/√−g δL/$δg^{{γα}}$), which is what makes the canonical and Hilbert tensors equivalent.

What would settle it

Compute the electromagnetic stress-energy tensor on a curved background: it is covariantly conserved and symmetric, yet it is not proportional to δ^μ_γ, so checking whether it satisfies the Section VII classification gives a direct, decisive test of the paper’s derivation.

Watch

Extended reading notes

Core claim

The central claim is that Noether’s theorem, applied to the complete variation of an action under spacetime translations, already contains the physics usually added by hand. When the electromagnetic or Yang–Mills potential is varied as a differential 1-form, the contribution that previously destroyed symmetry and gauge invariance of the canonical tensor recombines into the field strength Fγν, leaving a canonical energy–momentum tensor that is symmetric and gauge invariant on arbitrary curved backgrounds. For general relativity, the paper performs the same computation in the vielbein formalism with the spin connection as a gl(4)-valued 1-form; the resulting Noether current is, up to the gravitational constant and the volume element, the Einstein tensor, which is symmetric, manifestly covariant, and zero in vacuum. The paper concludes from this that the canonical Noether tensor and the Hilbert tensor are the same object, and it derives the Einstein field equations, with a cosmological constant appearing as an integration constant, from the conservation of the total Noether current alone.

Load-bearing premise

The load-bearing assumption is the Section VII statement that every symmetric, covariantly conserved tensor in general relativity must be proportional to the identity tensor times a constant; if that classification is not true, the Noetherian derivation of Einstein’s equations does not go through.

Editorial extensions

If this is right

For any gauge theory written in 1-form variables, the Noether energy–momentum tensor is symmetric and gauge invariant by construction, so improvement terms are unnecessary.
The construction works on curved backgrounds without assuming Minkowski spacetime, extending the canonical Noether prescription beyond its usual flat-space setting.
In the vielbein treatment of Einstein–Hilbert gravity, the Noether current is manifestly covariant, symmetric, and vanishes in vacuum, matching the properties usually reserved for the Hilbert tensor.
The equivalence between canonical and Hilbert tensors implies that energy–momentum conservation ∇_μ T^μ_ν = 0 follows from diffeomorphism invariance rather than from a separate physical postulate.
Einstein’s field equations with a cosmological constant can be presented as a consequence of Noether translation symmetry, with Λ emerging as the integration constant of the conservation law.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The 1-form prescription suggests a testable rule for other fields: any theory whose fundamental variable is a connection or a form should yield the Hilbert tensor from the Noether current without symmetrization, and a non-minimally coupled scalar field would be a quick check.
The Section VII derivation stands or falls on the unstated classification of symmetric covariantly conserved tensors in general relativity; a reader who wants to convert the paper’s conclusion into a theorem would need to prove that classification.
If the classical equivalence survives, the natural follow-up is to ask whether the improvement-term-free canonical tensor remains equal to the Hilbert tensor after quantization, where trace anomalies and renormalization could break the simple equality.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Punchline: this paper does not deliver what it promises. The claim that Noether's theorem alone yields an alternative derivation of Einstein's equations collapses in Sec. VII, where the author assumes without proof that any conserved symmetric tensor in GR is proportional to δ^μ_γ Λ. The paper's own electromagnetic stress tensor is a counterexample, so the assumption is not just unproved—it is false. The final equation is obtained by substituting the previously constructed T_GR = −(1/κ)g^{μα}G_{αγ} into that ansatz, which makes the Einstein equation an input rather than a consequence.

What the paper does well: the first half captures a real idea. Treating the gauge potential as a 1-form, so that its Lie derivative under diffeomorphisms contains the A_γ δx^γ_{,ν} term, is the right way to avoid ad hoc improvement terms. The final expressions for the electromagnetic and Yang–Mills stress tensors are the standard symmetric, gauge-invariant ones. As a pedagogical exposition of that particular trick, the paper is on the right track.

The soft spots are serious. Equation (1) claims that (B^{μν} φ)_{,νμ} vanishes because B^{μν} is antisymmetric and partial derivatives commute. Only the double contraction B^{μν}_{,νμ} vanishes; cross terms like B^{μν}_{,ν} φ_{,μ} remain and are never handled. The same error propagates to the Yang–Mills and gravitational derivations. In Sec. V.B.2, the claim that metric compatibility implies torsion-freeness is false in general; a metric-compatible connection can carry torsion, and the Palatini variation with an independent connection does not force it in the way the paper states. The 'global translation' step in Sec. VII also needs ∇_μ δx^γ = 0, which in curved spacetime means δx^γ is a Killing vector, not a general translation.

The net result is a restatement of standard results with new parts that are either wrong or circular. I would not cite it, and I would not bring it to a reading group except as an example of how not to structure a Noether derivation. A serious editor should desk-reject it; the correct parts are textbook material, and the incorrect parts are load-bearing. The author might salvage a tutorial note from the first half after fixing Eq. (1) and dropping the Einstein-equation derivation, but as submitted the paper is not publishable.

Referee Report

4 major / 5 minor

Summary. The paper proposes a unified Noether-theoretic construction of energy-momentum tensors. It argues that treating the gauge potential as a differential 1-form, so that its transformation under spacetime diffeomorphisms is the Lie derivative of a 1-form, produces a canonical EMT for Maxwell and Yang–Mills theory that is symmetric and gauge-invariant without artificial improvement terms. The same construction is then applied to the spin connection in the vielbein/Palatini formalism, with the claim that the resulting Noether current equals the Einstein tensor times sqrt(-g), that this canonical EMT is naturally equivalent to the Hilbert EMT, and that imposing conservation of the total Noether current yields the Einstein field equations with a cosmological constant. The final section advertises an 'alternative derivation of the Einstein field equations via Noether's theorem' that avoids varying the metric.

Significance. If the central claims were correct, the paper would offer a notable unification: one Noetherian recipe giving a symmetric, gauge-invariant canonical EMT for gauge fields and the Einstein tensor for gravity, together with a derivation of Einstein's equations from symmetry principles. The gauge-field section contains a genuine and partially correct observation that the 1-form Lie derivative in the Noether procedure yields the standard symmetric Maxwell tensor without ad hoc symmetrization. The paper also correctly identifies the vielbein and Palatini frameworks as the natural arena for a gravitational analogue. However, the advertised conclusions are not established: the gravitational EMT identification rests on an unsupported torsion claim, the equivalence theorem in Sec. VI does not follow from the stated variation, and the Sec. VII derivation of Einstein's equations is circular. The paper provides no machine-checked proofs, reproducible code, or falsifiable predictions; its positive value is confined to the gauge-field calculation and to the framing of the problem.

major comments (4)

[Section VI] The conclusion 'Metric compatible then torsion-free' is not a consequence of metric compatibility: any contorsion tensor K_mu nu rho antisymmetric in its first two indices can be added to the Levi-Civita connection while preserving the condition nabla_lambda g_mu nu = 0, and it carries arbitrary torsion. The Palatini equation of motion might in principle eliminate this contorsion, but the displayed 'solution' to Eq. (10) does not demonstrate that; the expression contains index errors, such as g^{mu gamma} g^{mu nu} T^gamma_{alpha gamma}, which repeats mu in a contracted term and is not a well-formed tensor equation. Since Sec. V.C.2 subsequently invokes 'the metric-compatible connection must be torsion-free' in order to replace g^{beta nu} R^mu{}_{beta gamma nu} with g^{alpha mu} R_{alpha gamma} in Eqs. (15)-(16), the central identification (t_GR)^mu_gamma = -g^{mu alpha} G_{alpha gamma} is not established.
[Section VI] The claimed proof of equivalence between the canonical and Hilbert EMTs is logically invalid. From Delta S = 0 for all delta x^gamma, the equation integral (2 delta L/delta g^{gamma alpha} g^{mu alpha} + sqrt(-g) T^mu_gamma) delta x^gamma_{;mu} + integral sqrt(-g) nabla_mu T^mu_gamma delta x^gamma = 0 yields a differential identity after integration by parts; it does not imply that the coefficient of delta x^gamma_{;mu} vanishes pointwise. The relation T^mu_gamma = g^{mu alpha}(-2/sqrt(-g) delta L/delta g^{gamma alpha}) is the standard definition of the Hilbert EMT, so the purported 'equivalence' is either tautological or requires the metric equations of motion and a discussion of the superpotential ambiguity of Noether currents; neither appears in the paper.
[Section VII] The derivation of the Einstein equations is circular. The assertion 'By the structure of conserved symmetric tensors in general relativity, we have (T_GR)^mu_gamma + (T_EM)^mu_gamma = delta^mu_gamma Lambda' is asserted without proof and is false: the Maxwell stress-energy tensor is a symmetric, covariantly conserved tensor that is generically not proportional to delta^mu_gamma, for example a Coulomb field has T^00 = -T^11, which is not of that form. Substituting the previously derived identity (T_GR)^mu_gamma = -g^{mu alpha} G_{alpha gamma} into this ansatz then returns G_{alpha gamma} + g_{alpha gamma} Lambda = kappa g_{mu alpha} (T_EM)^mu_gamma by algebra, so the Einstein equation is an input rather than a consequence of Noether's theorem. The argument also restricts to 'global translations' with nabla_mu delta x^gamma = 0, which on a generic curved spacetime are only Killing vector fields, not arbitrary translations; this is an additional unsupported assumption.
[Section V.C.2] The derivation of the gravitational Noether current in Sec. V.C.2 is structurally parallel to the gauge-field calculation, but its final identification of (t_GR)^mu_gamma with -g^{mu alpha} G_{alpha gamma} depends on the torsion-free claim from Sec. V.B.2 and on Riemann symmetries that hold only for a Levi-Civita connection. Because the torsion claim is not established, Eq. (16) is not a reliable result. Furthermore, the step from partial_mu{(t_GR)^mu_gamma + (t_EM)^mu_gamma] delta x^gamma} = 0 to sqrt(-g) nabla_mu{(T_GR)^mu_gamma + (T_EM)^mu_gamma] delta x^gamma} = 0 assumes the identification t^mu_gamma = T^mu_gamma sqrt(-g), which is part of what the paper is trying to prove rather than an independent input.

minor comments (5)

[Section III.A] The identity in Eq. (1) is actually correct as an identity for any antisymmetric C^{nu mu} times a scalar f under commuting partial derivatives, so the stress-test concern about this equation does not land; the sentence explaining the vanishing could be expanded for clarity, but the step itself is not erroneous.
[Section V.B.2] The sentence 'This relation shows the mutual implication between torsion and metric compatibility. We summarize it as follows:' is repeated verbatim; the duplicate should be removed.
[Section IV] The two displayed Lagrangians L[A_mu] and L[A^mu] are written identically, so the claimed distinction between the 1-form and vector-field formulations is invisible in the equations; the argument that only the 1-form formulation yields the correct Noether current needs clearer notation and a more explicit contrast.
[Section V.C.1] The index contractions in e eta_{ae} e^sigma_e e^omega_c R^c{}_{a omega sigma} are unclear; please spell out the contractions or use a more standard notation for the determinant of the vielbein and the curvature two-form.
[Section VII] The term 'global translation' is misleading when nabla_mu delta x^gamma = 0 is required; in a curved spacetime such vector fields are Killing vectors, and their existence is not guaranteed. The paper should either justify the existence of such a symmetry or phrase the conservation argument in terms of arbitrary compactly supported vector fields.

Circularity Check

1 steps flagged · score 7.0 of 10

The §VII 'derivation' of Einstein's field equations is circular: an unproved structural ansatz (total conserved symmetric tensor = δ^μ_γ Λ) is substituted into Eq. (16), where T_GR was already constructed as the Einstein tensor, so the Einstein equation is an input rewritten, not a consequence of Noether symmetry.

self definitional [Sec. VII, after Eq. (16); paragraph beginning 'Therefore, the total energy-momentum tensor must be conserved.']
"By the structure of conserved symmetric tensors in general relativity, we have: (TGR)µγ + (TEM)µγ = δµγ Λ where Λ is an integration constant, interpreted as the cosmological constant. Substituting the explicit form of the gravitational energy-momentum tensor from Eq. (16), we obtain: Gαγ + gαγΛ = κgµα (TEM)µγ"

This is not a Noether consequence but an ansatz that already encodes the desired conclusion. Eq. (16) fixed the gravitational canonical tensor to be proportional to the Einstein tensor, (TGR)^μ_γ = -κ^{-1} g^{μα} G_{αγ} (up to the paper's normalization). Because ∇_μ G^μ_γ = 0 is the Bianchi identity, conservation alone cannot force G_{αγ} to be proportional to g_{αγ}; and the Maxwell tensor is a standard symmetric covariantly conserved tensor not of the form δ^μ_γ Λ (e.g., Coulomb field has T^00 = -T^11 = E^2/8π). Therefore substituting Eq. (16) into the assumed form merely rewrites the assumption as G_{αγ}+g_{αγ}Λ = κ g_{μα}(TEM)^μ_γ. The Einstein field equation is inserted as the structural premise, not derived from translation symmetry.

full rationale

The gauge-theory and general-relativity Noether-current computations (Secs. II-VI) are direct algebraic manipulations from the stated Lagrangians; they contain no self-citation chain and no fitted parameter, and I find no circularity there aside from the usual need to check index/sign conventions. The circularity is concentrated in the advertised final claim, the 'alternative derivation of the Einstein field equations via Noether's theorem' in Sec. VII. After deriving conservation of the total Noether current, the paper asserts without proof that any conserved symmetric tensor in GR has the form δ^μ_γ Λ. This assertion is (i) unsupported and in fact false for the electromagnetic stress tensor, and (ii) equivalent, once Eq. (16) is substituted, to the Einstein equation with cosmological constant. The gravitational tensor is conserved identically by the Bianchi identity, so the conservation law carries no information forcing G_{αγ} ∝ g_{αγ}; the only content comes from the assumed structural form. Hence the central derivation reduces to an input assumption, meriting a high circularity score, while the rest of the paper retains independent computational content; score 7 rather than 8-10 because the earlier Noether-to-Einstein-tensor identity (Eq. 16) is a genuine computation, and only the final 'derivation of EFE' is circular.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central derivation rests on two false structural assumptions, the torsion equivalence and the conserved-tensor form, plus a domain assumption about the spin connection. The only free parameter is the cosmological constant introduced in the final Einstein equation step.

free parameters (1)

cosmological constant Λ = integration constant, value undetermined
Introduced in Sec. VII as an 'integration constant' inside the assumption that the total conserved symmetric tensor equals δ^μ_γ Λ. No derivation fixes its value; it is an extra parameter pulled into the Einstein equation derivation.

assumptions (3)

ad hoc to paper Metric compatibility implies torsion-freeness (and vice versa) under Palatini variation
Stated as a conclusion in Sec. V.B.2 from Eq. (10). This is false in general: metric-compatible connections with nonzero torsion exist. The paper relies on this to simplify the spin-connection derivation to the Einstein tensor.
ad hoc to paper Any covariantly conserved symmetric tensor in general relativity is proportional to δ^μ_γ times a constant
Invoked in Sec. VII as 'the structure of conserved symmetric tensors.' This is false; the electromagnetic stress tensor is conserved and not of that form. It is the step that converts the conservation law into the Einstein equations.
domain assumption The spin connection transforms under diffeomorphisms exactly like a gauge connection 1-form with the same Lie derivative rule
Used in Sec. V.C.2 to apply the gauge-theory Noether manipulation to ω^b_{cν}. The paper notes subtleties from footnote 9 of [10] but assumes the rule holds anyway.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On the Natural Equivalence Between Canonical and Hilbert Energy Momentum Tensors via Noether's Theorem." pith.science (2026). https://pith.science/paper/VQDEOUEQ

@misc{pith2026250705780,
  author       = {Pith},
  title        = {Pith review of: On the Natural Equivalence Between Canonical and Hilbert Energy Momentum Tensors via Noether's Theorem},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VQDEOUEQ}},
  note         = {Machine review of arXiv:2507.05780}
}

read the original abstract

In this work, we investigate the structure and properties of the canonical energy momentum tensor (EMT) across a range of field theories. We begin by developing a unified and systematic method that naturally yields the canonical EMT for gauge theory, without the need for artificial symmetrization or improvement terms. Our analysis highlights how Noether's theorem intrinsically emphasizes the 1-form nature of gauge potentials. We further extend to general relativity and demonstrate that the assumption of metric compatibility naturally implies a torsion free connection. Building upon variational symmetry principles, we establish the equivalence between the Einstein Hilbert EMT and the canonical EMT, thereby clarifying their respective roles in field dynamics and conservation laws. Lastly, we present an alternative derivation of the Einstein field equations via Noether's theorem.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 11 canonical work pages

[1]

Invariant variation problems

Emmy Noether. Invariant variation problems. Transport theory and statistical physics, 1(3):186–207, 1971

work page 1971
[2]

CRC press, 2018

Richard Feynman.Feynman lectures on gravitation. CRC press, 2018

work page 2018
[3]

Elsevier, 2013

Lev Davidovich Landau.The classical theory of fields, volume 2. Elsevier, 2013

work page 2013
[4]

Noether’s theorems and the energy-momentum tensor in quantum gauge theories.Physical Review D, 106(12):125012, 2022

Adam Freese. Noether’s theorems and the energy-momentum tensor in quantum gauge theories.Physical Review D, 106(12):125012, 2022

work page 2022
[5]

Cambridge university press, 1996

Lewis H Ryder.Quantum field theory. Cambridge university press, 1996

work page 1996
[6]

Princeton University Press, 1996

Paul Adrien Maurice Dirac.General theory of relativity, volume 14. Princeton University Press, 1996

work page 1996
[7]

Ostrogradsky

M. Ostrogradsky. Mémoires sur les équations différentielles, relatives au problème des isopérimètres. Mem. Acad. St. Petersbourg, 6(4):385–517, 1850. 13

work page
[8]

Richard P. Woodard. Ostrogradsky’s theorem on Hamiltonian instability.Scholarpedia, 10(8):32243, 2015

2015

Show all 12 references

[9]

Deduzione invariantiva delle equazioni gravitazionali dal principio di hamilton.Rendi- conti del Circolo Matematico di Palermo (1884-1940), 43(1):203–212, 1919

Attilio Palatini. Deduzione invariantiva delle equazioni gravitazionali dal principio di hamilton.Rendi- conti del Circolo Matematico di Palermo (1884-1940), 43(1):203–212, 1919

1940
[10]

Symmetry and physics.Proceedings of the American Philosophical Society, 140(3):267– 288, 1996

Chen Ning Yang. Symmetry and physics.Proceedings of the American Philosophical Society, 140(3):267– 288, 1996

1996
[11]

Courier Dover Publications, 2020

Kentaro Yano.The theory of Lie derivatives and its applications. Courier Dover Publications, 2020

2020
[12]

Gravitation, gauge theories and differential geometry

Tohru Eguchi, Peter B Gilkey, and Andrew J Hanson. Gravitation, gauge theories and differential geometry. Physics reports, 66(6):213–393, 1980. 14

1980

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Invariant variation problems

Emmy Noether. Invariant variation problems. Transport theory and statistical physics, 1(3):186–207, 1971

work page 1971

[2] [2]

CRC press, 2018

Richard Feynman.Feynman lectures on gravitation. CRC press, 2018

work page 2018

[3] [3]

Elsevier, 2013

Lev Davidovich Landau.The classical theory of fields, volume 2. Elsevier, 2013

work page 2013

[4] [4]

Noether’s theorems and the energy-momentum tensor in quantum gauge theories.Physical Review D, 106(12):125012, 2022

Adam Freese. Noether’s theorems and the energy-momentum tensor in quantum gauge theories.Physical Review D, 106(12):125012, 2022

work page 2022

[5] [5]

Cambridge university press, 1996

Lewis H Ryder.Quantum field theory. Cambridge university press, 1996

work page 1996

[6] [6]

Princeton University Press, 1996

Paul Adrien Maurice Dirac.General theory of relativity, volume 14. Princeton University Press, 1996

work page 1996

[7] [7]

Ostrogradsky

M. Ostrogradsky. Mémoires sur les équations différentielles, relatives au problème des isopérimètres. Mem. Acad. St. Petersbourg, 6(4):385–517, 1850. 13

work page

[8] [8]

Richard P. Woodard. Ostrogradsky’s theorem on Hamiltonian instability.Scholarpedia, 10(8):32243, 2015

2015