REVIEW 2 major objections 5 minor 20 references
An impediment to torsion from spectral geometry
T0 review · 2 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read Spectral geometry imposes a strong obstruction on torsion in gravity: the spectral Einstein functional is well defined only when the torsion tensor vanishes.
desk verdict Solid computation of the torsion Einstein functional, but the no-go for all modifications rests on an unproved classification; the strongest conclusion should be weakened. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is the Wodzicki residue, the unique trace on classical pseudodifferential operators, used to turn operator data into geometric densities; the spectral Einstein functional is the bilinear form $G(u,w)=\mathrm{Wres}(u\{D,w\}D|D|^{-n})$ whose torsion-free value reproduces the Einstein tensor. Against this, the paper perturbs the Dirac operator by a zero-order Clifford endomorphism $B$ representing antisymmetric torsion, expands symbols in normal coordinates, and computes the residue densities by pseudodifferential calculus. The classification of modified functionals into the families $F_1(k)$ and $F_2(k)$ by power counting is what lets the authors exclude coherent alternatives.
What would settle it
A single explicit spectral functional belonging to neither of the two families $F_1(k),F_2(k)$, built from $D$, $|D|$ and their inverses with polynomial powers, whose density has no derivatives of $w$ and whose torsion-free limit is the Einstein tensor, would overturn the no-go. Concretely, on a flat manifold with constant antisymmetric torsion, one could list all residue densities of monomials in $D$ and $|D|$ sandwiched between forms and check whether any cancellation of the $w_{bc}$ terms occurs outside the classified families.
Extended reading notes
Core claim
On the paper's own terms, the central discovery is that the spectral Einstein functional $G(u,w)=\mathrm{Wres}(u\{D,w\}D|D|^{-n})$ is not a tensor-type functional when the Dirac operator is twisted by torsion. Writing $D=D_0+B$ with $B=-\frac{i}{8}T_{abc}\gamma^a\gamma^b\gamma^c$ for antisymmetric torsion $T$, the residue density acquires a correction $\delta G$ whose first term is proportional to $u_a w_{bc}\,\mathrm{Tr}([\gamma^a,\gamma^b]\{\gamma^c,B_0\})$ plus further derivative terms. For a three-form perturbation this trace cannot vanish unless $T=0$, so the naive functional fails to be tensorial. The authors then classify potential modifications by power counting as finite linear combinations of $F_1(k)=\mathrm{Wres}(u|D|^k w|D|^{-n-k+2})$ and $F_2(k)=\mathrm{Wres}(uD|D|^k wD|D|^{-n-k})$, and show that eliminating the derivative terms in these families forces either an unacceptable vanishing of the Einstein tensor in the torsion-free limit or the same impossible trace condition. The conclusion stated is that for the spectral Einstein functional to be well defined, the torsion $T$ must vanish.
Load-bearing premise
The argument depends on the classification claim that any tensorial spectral functional must be a finite linear combination of the two families $F_1(k)$ and $F_2(k)$; if a functional outside that classification cancels the unwanted derivative terms, the obstruction to torsion could disappear.
Editorial extensions
If this is right
- If the argument is right, gravity models that include antisymmetric torsion cannot be obtained from a spectral Einstein functional; the usual Levi-Civita connection is singled out by spectral geometry.
- The spectral Einstein tensor is geometrically meaningful only in the torsion-free setting, so spectral action programs should not expect to generate torsional corrections of this form.
- Any future attempt to include torsion in a spectral gravity model must either step outside the $F_1,F_2$ classification or change the spectral construction itself.
- The no-go applies to the antisymmetric (3-form) part of torsion; vectorial torsion is already excluded by self-adjointness of the Dirac operator, while Cartan torsion is left transparent by this argument.
Reading between the lines
- If carried further, the same derivative obstruction should reappear for any spectral density built from the same power-counting families, since the problematic term comes from $B_0$ rather than from the Einstein-tensor structure; this predicts a similar no-go for spectral Ricci-type functionals.
- A direct extension would compute the analogue for one-form gauge perturbations, which the paper notes satisfy the vanishing trace condition; those perturbations may yield a genuinely tensorial spectral functional and mark the boundary of admissible torsion-like terms.
- The reported mismatch with the parallel computation in [17] invites an independent recalculation of $\delta G$; whichever coefficient set survives, the paper's classification argument would remain a separate, testable step.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The Letter argues that the spectral Einstein functional, defined through the Wodzicki residue as Wres(u{D,w}D|D|^{-n}), cannot be extended to Dirac operators with torsion. An explicit computation shows that, for a torsionful Dirac operator D = D0 + B, the density acquires terms depending on derivatives of the test form w; these terms vanish only if the torsion vanishes. The authors then consider modified spectral Einstein functionals taken to be finite linear combinations of two families F1(k) and F2(k), and show that the derivative obstructions cannot be removed without also killing the standard Einstein tensor. The Supplemental Material provides detailed symbol computations for F1(k) and F2(k). The central conclusion is that spectral geometry imposes a strong obstruction to torsion: for the spectral Einstein functional to be well defined, the torsion must vanish.
Significance. If the classification of admissible modifications in Eq. (8) of the Letter were established, the result would be a substantial and surprising structural obstruction: spectral geometry would single out torsionless connections in a way that is independent of dynamical considerations. The symbolic computations in the Supplemental Material are detailed and appear internally consistent, and the explicit derivation of the derivative obstruction in the unmodified Einstein functional is a concrete and falsifiable statement. The main weakness is that the no-go conclusion depends on an unproved completeness assertion for the family of modified functionals; the paper is therefore not yet sufficient to support the strong unconditional claim made in the abstract and conclusions.
major comments (2)
- [Modified Einstein functionals (Eq. (8))] The assertion that any potential modification of the spectral Einstein functional is a finite linear combination of F1(k) and F2(k) is load-bearing but is not proved. The Supplemental Material states that the authors verify that no functional depending polynomially on D, |D|, and |D|^{-1} works, but the actual computations treat only the two families F1(k) and F2(k). No lemma shows that every polynomial word in u, w, D, |D|, and |D|^{-1} of total order 2-n reduces to these families using cyclicity of the Wodzicki residue and D^2 = |D|^2. For instance, the word u D |D|^{1-n} w has the same total order as the terms in Eq. (8), is not manifestly a combination of F1(k) and F2(k), and its subprincipal symbol can contribute to the Wodzicki residue with coefficients not controlled by the factors k(n+k) and k(n+k-2) used in the cancellation argument. Unless such one-D terms are shown to vanish or to be linear combinations of F1 and F2, the conclusion that the torsion-dependent trace must vanish does not follow.
- [Conclusions / Supplemental Material Section III] The unconditional conclusion that 'for the spectral Einstein functional to be well defined, the torsion T must vanish' is stronger than what the manuscript actually verifies. The Supplemental Material restricts attention to functionals depending polynomially on D, |D|, and |D|^{-1}, and within that class it computes only the F1/F2 subfamily. Non-polynomial spectral functionals, such as those involving non-integer powers |D|^alpha with alpha + beta = 2 - n, are not addressed, and no argument is given that a well-defined tensorial spectral functional must be polynomial. The claim should be restricted to the F1/F2 family, or a completeness proof for the classification in Eq. (8) should be supplied.
minor comments (5)
- [Eq. (10)] The term 'uawaGab' has an index clash; it should be written, for example, as u^a w^b G_{ab}. The same correction applies to Eq. (63) of the Supplemental Material.
- [Supplemental Material, Proposition II.1] The proof introduces the operator O = \hat v D \hat w D, but the symbol \hat v is not defined; from the context it should clearly be \hat u.
- [Eq. (8) and following paragraph] The argument that the condition sum_k beta_k = 0 is unacceptable assumes that the modified functional must reduce to the standard Einstein tensor in the torsionless case, which presumably requires a normalization such as sum_k alpha_k = sum_k beta_k = 1. This normalization should be stated explicitly before it is used.
- [Eq. (7)] The expansion of w, namely w = (w_a + w_ab x^b + w_abc x^b x^c) gamma^a, is introduced after Eq. (7), although Eq. (7) already uses the coefficients w_ab and w_abc. Moving this definition before the displayed formula would improve readability.
- [Footnote [14]/[17]] The footnote states that the coefficients of a structurally similar result in reference [17] are inconsistent with the present result, but no comparison or detail is provided; either the comparison should be given or the remark should be softened.
Circularity Check
No significant circularity: the torsion obstruction follows from a direct Wodzicki-residue computation; the unproven completeness of Eq. (8) is a correctness gap, not a circular step.
full rationale
The paper's derivation chain is not circular. The central no-go conclusion is obtained by computing the Wodzicki residue of the spectral Einstein functional in the presence of torsion (main-text Eq. (3), Supplemental Proposition II.1) and of the modified family (Eqs. (8)-(10), Supplemental Propositions III.5 and III.7). The torsion-dependent obstruction, namely the first-order derivative term proportional to Tr([\gamma^a,\gamma^b]\{\gamma^c,B_0\}), is derived explicitly from pseudodifferential symbol expansions and does not reduce to a fitted parameter, a renaming, or to the authors' earlier results. Citations to the same group's work ([11] and [1] in the Supplemental Material) supply the torsion-free Einstein functional and standard symbol-expansion lemmas; these are independently published mathematical theorems with stated assumptions, not assumptions that already contain the target no-go statement, so they do not constitute load-bearing circularity. The main caveat is non-circular but serious: Eq. (8) is introduced by a 'power-counting argument' and the Supplemental Material verifies only the F1(k) and F2(k) families, leaving open whether other spectral functionals, such as terms with a single D adjacent to w, could cancel the derivative obstruction. If the classification is incomplete, the no-go conclusion would not follow; however, that is an omitted completeness proof and a correctness risk, not a circular reduction.
Assumptions & free parameters
assumptions (5)
- standard math Existence and uniqueness of the Wodzicki residue as a trace on classical pseudodifferential operators.
- standard math The symbol expansions for powers of Laplace-type operators (Supplemental, Eq. (4)) and for |D|^k (Lemma III.2) are correct.
- domain assumption A well-defined spectral Einstein functional must be of tensor type: its density is an evaluation of tensors on the inserted forms without derivatives of those forms.
- ad hoc to paper Any tensorial modification of the spectral Einstein functional must belong to the family G' = sum_k (alpha_k F1(k) + beta_k F2(k)) with finitely many nonzero coefficients.
- domain assumption The self-adjoint Dirac operator with torsion has the form D = D0 + B with B = -i/8 T_abc gamma^a gamma^b gamma^c, i.e., only the antisymmetric part of the torsion tensor contributes.
Cite this review
Pith. "Pith review of An impediment to torsion from spectral geometry." pith.science (2026). https://pith.science/paper/FT77FYJQ
@misc{pith2026241219626,
author = {Pith},
title = {Pith review of: An impediment to torsion from spectral geometry},
year = {2026},
howpublished = {\url{https://pith.science/paper/FT77FYJQ}},
note = {Machine review of arXiv:2412.19626}
}
read the original abstract
Modifications of standard general relativity that bring torsion into a game have a long-standing history. However, no convincing arguments exist for or against its presence in physically acceptable gravity models. In this Letter, we provide an argument based on spectral geometry (using methods of pseudo-differential calculus) that suggests that the torsion shall be excluded from the consideration. We demonstrate that there is no well-defined functional extending to the torsion-full case of the spectral formulation of the Einstein tensor.
Reference graph
Works this paper leans on
-
[16]
The vectorial one also must be trivial due to the self-adjointness of the Dirac oper- ator
More precisely, this shows that the antisymmetric part of the torsion has to vanish. The vectorial one also must be trivial due to the self-adjointness of the Dirac oper- ator. Cartan torsion can potentially be allowed as being completely transparent in this approach
-
[1]
Kac, Can One Hear the Shape of a Drum?, The American Mathematical Monthly 73, 1 (1966)
M. Kac, Can One Hear the Shape of a Drum?, The American Mathematical Monthly 73, 1 (1966)
work page 1966
-
[2]
Connes, Noncommutative Geometry (Academic Press, 1994)
A. Connes, Noncommutative Geometry (Academic Press, 1994)
work page 1994
-
[3]
A. H. Chamseddine and A. Connes, Universal Formula for Noncommutative Geometry Actions: Unification of Gravity and the Standard Model, Phys. Rev. Lett. 77, 4868 (1996)
work page 1996
-
[4]
Connes, Gravity coupled with matter and the foundation of non-commutative geometry, Commun.Math
A. Connes, Gravity coupled with matter and the foundation of non-commutative geometry, Commun.Math. Phys. 182, 155– (1996)
work page 1996
-
[5]
J. Bellissard, A. van Elst, and H. Schulz- Baldes, The noncommutative geometry of the quantum Hall effect, Journal of Mathematical Physics 35, 5373 (1994)
work page 1994
-
[6]
M. A. Shubin, Pseudodifferential operators and spectral theory , 2nd ed. (Springer-Verlag, Berlin, 2001) pp. xii+288, translated from the 1978 Russian original by Stig I. Andersson
work page 2001
-
[7]
Wodzicki, Noncommutative residue Chapter i
M. Wodzicki, Noncommutative residue Chapter i. Fundamentals, in K-Theory, Arithmetic and Geometry: Seminar, Moscow Univer sity, 1984–1986 , edited by Y. I. Manin (Springer Berlin Heidelberg, Berlin, Heidelberg, 1987) pp. 320–399
work page 1984
Show all 20 references
-
[8]
P. B. Gilkey, Invariance theory, the heat equation and the Atiyah-Singer index theorem (Publish or Perish, Dilming- ton, 1984)
1984
-
[9]
Kalau and M
W. Kalau and M. Walze, Gravity, non- commutative geometry and the Wodzicki residue, Journal of Geometry and Physics 16, 327 (1995)
1995
-
[10]
Kastler, The dirac operator and gravitation, Commun.Math
D. Kastler, The dirac operator and gravitation, Commun.Math. Phys. 166, 633 (1995)
1995
-
[12]
Dąbrowski, A
L. Dąbrowski, A. Sitarz, and P. Zalecki, Spectral Torsi on, Commun. Math. Phys. 405, 130 (2024)
2024
-
[13]
See Supplemental Material at [URL will be inserted by publisher]
-
[14]
We recently became aware of a structurally similar resu lt in [17], however exact values of the coefficients presented therein are inconsistent with our result
-
[15]
We remark that allowed perturbations of the Dirac oper- ator for which the term vanishes identically include gauge perturbations, where B is a one-form
-
[17]
An impediment to torsion from spectral geometry
J. Hong and Y. Wang, The spectral Einstein functional for the Dirac operator wit h torsion (2024), arXiv:2412.08028 [math.DG]. arXiv:2412.19626v2 [gr-qc] 5 Jan 2025 APS/123-QED Supplemental Material for “An impediment to torsion from spectral geometry” Arkadiusz Bochniak ∗ Max...
2024 arXiv
-
[18]
In a flat space, its principal symbol should be just d1 = |ξ|
(36) |D| is a positive first-order pseudodifferential operator with t he principal symbol being a scalar. In a flat space, its principal symbol should be just d1 = |ξ|. In normal coordinates it gives d1 = |ξ| + d11 c xc + d12 cdxcxd + o(x2). Eqn. (32) is obtained by setting (d1)2...
-
[19]
Dąbrowski, A
L. Dąbrowski, A. Sitarz, and P. Zalecki, Spectral metric and Einstein functionals, Advances in Mathematics 427, 109128 (2023)
2023
-
[20]
P. B. Gilkey, Invariance theory, the heat equation and the Atiyah-Singer index theorem (Publish or Perish, Dilmington, 1984)
1984
-
[21]
Taylor, Pseudodifferential operators (Princeton University Press, 1981)
M. Taylor, Pseudodifferential operators (Princeton University Press, 1981). 20
1981
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.