REVIEW 3 major objections 4 minor 1 cited by
Alternative Action for Generalized Unimodular Gravity
T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read Every generalized unimodular gravity model admits an equivalent action in which general relativity couples to an auxiliary cosmological-constant field through a spatially nonlocal operator, with unimodular gravity as the exceptional local…
desk verdict A genuinely new HT-type action for GUMG with a real proof gap: the first-class status of the central constraint is asserted rather than demonstrated. 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 load-bearing object is the delocalization operator $E=I+W^{-1}\tilde I W$, where $I$ is the identity, $I$ projects onto spatially homogeneous (average) functions, and $\tilde I=I-I$ projects onto average-free functions; $W=d\ln F/d\ln\sqrt{\gamma}$ is the model's barotropic parameter. The operator appears when the Hamiltonian constraint of general relativity, which lives only in its spatially averaged part after consistent parameterization, must be merged with GUMG's secondary constraint into one functionally complete constraint. Because $E$ is invertible, the constraint basis can be rearranged as $\pi_0+E F H_\perp=0$ and $\pi_{0,m}=0$, which is what allows the metric momentum dependence to be reduced as in general relativity and the auxiliary fields to assemble into $\partial_\mu\mathcal{V}^\mu$. The consistency of the whole construction rests on the first-class constraint $P_I=\pi_0+F H_\perp+U_0^n H_n$ in the parameterized action. The same operator controls the nonlocal on-shell relation for $\Lambda$ and the quantum measure, whose determinant is $\mathrm{Det}\,E=W\,\overline{W^{-1}}\ge 1$.
What would settle it
Choose a GUMG model with spatially varying $W(\sqrt{\gamma})$ on a compact spatial slice and test the parameterized action's new constraint $P_I=\pi_0+F H_\perp+U_0^n H_n$ by the standard Dirac consistency procedure: if preserving that constraint in time generates new constraints, or if the rank of the constraint matrix changes, the two actions are not equivalent. The exceptional characteristic functions with $\Omega(\sqrt{\gamma})=0$, such as $F\propto|b+\sqrt{\gamma}^{-1}|$, are the natural place to look, since the paper itself identifies them as potentially pathological.
Extended reading notes
Core claim
The central claim is a constructive proof that the generic GUMG action $S[g,\lambda_\perp]=\int dt\,dx\,\sqrt{|g|}\,R-\int dt\,dx\,\lambda_\perp(N^\perp-F(\sqrt{\gamma}))$ is classically equivalent to $S_{\rm alt}[g,\Lambda,\mathcal{V}]=\int dt\,dx\,\sqrt{|g|}(R-\Lambda)+\int dt\,dx\,\partial_\mu\mathcal{V}^\mu\,E F\sqrt{\gamma}\,\Lambda$, where $\Lambda(t,x)$ is an auxiliary cosmological-constant field, $\mathcal{V}^\mu$ is an auxiliary vector field entering only through $\partial_\mu\mathcal{V}^\mu$, and $E$ is an invertible, local-in-time but spatially nonlocal operator built from average and average-free projectors. The equivalence is established by introducing time parametrization into the canonical action, adding an auxiliary canonical pair, and rearranging the constraint basis so that all metric-momentum dependence sits in the usual general-relativity structures. The resulting action reproduces the original GUMG dynamics and gauge structure on both dynamical branches: the GR branch has the same degrees of freedom as general relativity, while the non-GR branch carries one extra degree of freedom that on shell is a cosmological perfect fluid with barotropic parameter $W(\sqrt{\gamma})$. When $W$ is constant the operator $E$ becomes the identity and the action is manifestly local; when $W=-1$ the whole action becomes the standard Henneaux–Teitelboim covariant action. On shell the effective cosmological constant feels the spatial average of $W^{-1}$, namely $\Lambda\sim\sqrt{\gamma}^{-1}F^{-1}\,\overline{W^{-1}}/W^{-1}\,c_0$.
Load-bearing premise
The whole equivalence chain rests on the claim that the reparametrization constraint $P_I=\pi_0+F H_\perp+U_0^n H_n$ is exactly first-class in the parameterized action, and that the nonlocal operator $E$ is invertible everywhere; if $W$ or $\Omega$ vanish on some configuration, the constraint algebra branches and the two actions describe different physics.
Editorial extensions
If this is right
- Every GUMG model can be analyzed in one common representation, with all model dependence isolated in $F$, $W$, and $E$; this should simplify comparisons between different restriction functions and with unimodular gravity.
- The on-shell value of the effective cosmological constant becomes a spatially nonlocal functional of the metric, so cosmological solutions in the same GUMG model can differ purely from the global spatial profile of $W$.
- The gauge symmetry on the non-GR branch is exhausted by homogeneous time reparametrizations and transverse (volume-preserving) spatial diffeomorphisms; local time reparametrizations and longitudinal spatial diffeomorphisms are broken.
- In the w-GUMG subfamily with $W=\mathrm{const}$, the alternative action is fully local while keeping the same mixed-class gauge structure, giving a simpler setting for studying GUMG dynamics.
- The path-integral measure in the $\Lambda$ representation acquires a non-ultralocal factor $\mathrm{Det}\,E=W\,\overline{W^{-1}}\ge 1$, so spatially inhomogeneous $W$ configurations are weighted differently in the quantum theory.
Reading between the lines
- A general criterion suggested by this construction is that a restricted gravity theory admits a Henneaux–Teitelboim-like covariantization exactly when its secondary constraint is, up to a local weight, the average-free part of the Hamiltonian; theories with additional structure would require genuinely new techniques.
- A testable extension is to insert the nonlocal on-shell relation for $\Lambda$ into a homogeneous cosmological model, where $W(\sqrt{\gamma})$ depends on the scale factor, and compare the resulting effective dark-energy equation of state with reconstructions from supernovae and cosmic microwave background data.
- Because the surviving spatial gauge symmetry is exactly volume-preserving diffeomorphisms, the whole family could be reformulated as general relativity on a spacetime with a fixed spatial volume form; the paper does not develop this coordinate-free reading, but the constraint analysis points directly to it.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper claims to construct, by a sequence of canonical transformations, an alternative action (7) for the whole family of generalized unimodular gravity (GUMG) theories introduced in Eq. (1). The construction starts from the canonical extended action (31), implements time reparametrization, rearranges the constraint basis, and introduces a spatially nonlocal operator E defined in Eqs. (95)-(96). For the constant-W subfamily (w-GUMG) the action is claimed to become local and to reproduce the Henneaux-Teitelboim action of unimodular gravity in the special case W = -1. The paper also analyzes the on-shell dynamics, the gauge structure, and the effect of the nonlocality on the quantum measure. The central claim is that the alternative action (7)/(106) is classically equivalent to the original GUMG action for all nonexceptional characteristic functions F(√γ).
Significance. If the central equivalence claim were correct, the paper would be a valuable contribution: it would provide a Henneaux-Teitelboim-like covariant formulation for the entire GUMG family, make explicit the spatial delocalization inherent in the model, and open the way to a systematic study of the gauge structure and quantum measure. The paper is constructive, contains many explicit computations, and the w-GUMG subfamily (Section 3) appears to be handled carefully; the local action (55) for constant W is a credible result that could be useful on its own. However, the general-GUMG equivalence is the main result, and it is not established: the proof contains a load-bearing error in the constraint-basis rearrangement of Section 4.2, and the first-class property of the key constraint P_I in Section 4.1 is only asserted. Because the primary claim of the paper is the equivalence for the general GUMG family, the significance of the paper as it stands is substantially reduced.
major comments (3)
- [Section 4.2, Eq. (98)] The claimed equivalence of the constraint bases in (98) is not correct for nonconstant W(√γ). Using the explicit kernel (95), one has E(FH⊥) = FH⊥ − W^{-1} avg(W F H⊥) + avg(F H⊥). On the secondary surface (W F H⊥),m = 0, i.e. W F H⊥ = A(t), this gives E(FH⊥) = A(t) avg(W^{-1}), whereas F H⊥ = A(t) W^{-1}. Consequently the new first constraint π0 + E F H⊥ = 0 forces π0 = −A(t) avg(W^{-1}), which is spatially constant, while the old first constraint π0 + F H⊥ = 0 forces π0 = −A(t) W^{-1}, which is spatially constant only when W is constant. The two sets in (98) therefore describe different constraint surfaces in the extended phase space. This invalidates the derivation of the alternative action (7)/(106) and the subsequent on-shell relation (108) for general GUMG theories. The subsequent Lagrange-multiplier redefinition (100) does not repair the discrepancy, since it cannot change the constraint surface itself.
- [Section 4.1, Eq. (92)] The first-class property of P_I = π0 + F H⊥ + U_0^n H_n is asserted rather than proved. The text says that the system of involution conditions 'now has a solution given by (92)', but no Poisson-bracket computation is shown for {P_I, ∫f(π0+FH⊥)}, {P_I, ∫ξ^n H_n}, or {P_I, ∫η^n (WFH⊥),n}. This matters because U_0^n is only the on-shell Lagrange-multiplier solution (34); the standard Dirac construction yields an integrated first-class Hamiltonian, and it is not automatic that the local smeared density P_I is in involution with all constraints. The rank and degree-of-freedom equivalence between (91) and (31) depends directly on this point. A concrete bracket computation, or a precise reference to where it is performed, is required.
- [Section 4.2, Eq. (99)] The assertion that (91) and (99) are equivalent because 'the equivalence of representations is guaranteed by the freedom of choosing any equivalent constraint basis' is too quick. A constraint-basis change is permissible only if the two sets have the same constraint surface and the same rank structure. As shown in the first major comment, the two sets in (98) do not define the same surface for nonconstant W, so the equivalence of (91) and (99) is not established. This is not a presentation issue but a load-bearing gap in the constructive proof of the main result.
minor comments (4)
- [Section 4.2 and Appendix A] The notation for the identity operator and the averaging projector is extremely confusing: the same symbol I is used for both in equations such as (96) and (135). This ambiguity appears to be directly connected to the error in the reformulation of the first constraint in (98). The authors should use distinct symbols, e.g. Id for the identity and P_avg or I_avg for the averaging projector, and recheck all equations involving E.
- [Section 4.1, after Eq. (91)] The claim that the consistently parameterized action (91) is 'physically equivalent' to the original GUMG extended action (31) is stated without a full proof of the rank and transversality conditions; please provide the explicit consistency conditions and show that they are satisfied.
- [Section 5, Conclusions] There is a typo in the first paragraph: 'various issues are still are still open' should read 'various issues are still open'.
- [Appendix D] The treatment of noncompact spatial sections is explicitly heuristic and relies on a finite-volume regulator; this is acknowledged by the author, but the main text refers to the results as if they were established. Please state more clearly which results in Sections 4.3-4.4 depend on the compactness assumption and which are expected to survive in the noncompact case.
Circularity Check
No significant circularity: the alternative action is derived by explicit equivalence-preserving canonical transformations from the original GUMG action, and the perfect-fluid equation of state is inherited as a consistency check, not fitted.
full rationale
This paper is a constructive reformulation, not an empirical prediction paper. Section 2 re-derives the canonical extended action (31) from the GUMG action (1) with explicit Poisson brackets (32), so the starting constraint structure is re-derived rather than imported solely from [13]. The alternative actions (52), (55), (104), and (106) are obtained by parameterization, invertible constraint-basis changes, and reduction; the text states that all intermediate transformations preserve classical equivalence, and the equations of motion of the alternative action are therefore equivalent to the original action by construction. That is the stated theorem, not hidden circularity. The recovered perfect-fluid equation of state p = W rho (117) is the same relation already present in the original model (5) and is used as a consistency check, not as a prediction obtained by fitting; the constant c0 is an on-shell integration constant, not a fitted parameter. Self-citations to [13] and [19] provide context and prior constraint analysis, but Section 2 re-derives the needed structures, so they are not load-bearing in a circular way. The point most open to challenge is the asserted involution of P_I in (92) and the basis equivalence (98), but those are unexplained or terse mathematical steps, not reductions of a prediction to its inputs. No circular step can be exhibited from the paper's own equations.
Assumptions & free parameters
assumptions (6)
- domain assumption The canonical extended action and constraint structure of GUMG from Section 2 (taken from [13]) is correct and complete.
- domain assumption The characteristic function F(√γ) is such that W(√γ)=d ln F/d ln √γ and Ω(√γ)=d ln W/d ln √γ + W + 1 are sign-definite and nonzero on the domain.
- domain assumption Spatial sections are compact for the central proof; noncompact sections are handled by a finite-volume regularization whose asymptotic behavior is not fully specified.
- standard math The delocalization operator E defined by (95) is invertible on the relevant function space with inverse given by Lemma A.1.
- standard math Standard results from the theory of constrained systems (Dirac consistency, existence of first-class Hamiltonian, correspondence gauge reduction) as presented in [22] are assumed.
- domain assumption Reduction of gravitational momenta by their own variational equations proceeds as in GR even though the lapse multiplier is associated with a mixed-class constraint on the non-GR branch.
invented entities (3)
-
Auxiliary vector field V^μ (Henneaux-Teitelboim field)
-
Cosmological-constant field Λ0 (or rescaled Λ)
-
Time-parametrization pair τ0, π0 (homogeneous and average-free components)
Cite this review
Pith. "Pith review of Alternative Action for Generalized Unimodular Gravity." pith.science (2026). https://pith.science/paper/4ZCWGSQQ
@misc{pith2026250513548,
author = {Pith},
title = {Pith review of: Alternative Action for Generalized Unimodular Gravity},
year = {2026},
howpublished = {\url{https://pith.science/paper/4ZCWGSQQ}},
note = {Machine review of arXiv:2505.13548}
}
read the original abstract
We present an alternative formulation of generalized unimodular gravity (GUMG), a class of modifications to general relativity characterized by a special partial breaking of general coordinate covariance. The action for this formulation is derived constructively through a sequence of equivalent representations, starting from the original GUMG setup and extending the configuration space and gauge structure by introducing time parametrization. Our approach, based on a canonical formalism, parallels the method used by Henneaux and Teitelboim to covariantize the action of unimodular gravity (UMG), which was generalized to consistently accommodate a parameterization via local fields for the entire GUMG family. For completeness, we explore the dynamical structure of the theory and provide a detailed account of its gauge properties. A notable consequence of the consistent parametrization is the emergence of explicit spatial delocalization in the action, manifestations of which we carefully examine. Within the GUMG family, we identify a subfamily of models described by a local action. While preserving the essential structural features, these models avoid the complications of spatial nonlocality allowing a clearer analysis. The dynamical and constraint structures of GUMG family models are different from those of UMG, however the latter appears as an exceptional special case within the family, providing valuable comparative insights.
Forward citations
Cited by 1 Pith paper
-
Henneaux-Teitelboim Form of the Generalized Unimodular Gravity Action
A generalized Henneaux-Teitelboim action for generalized unimodular gravity is constructed, introducing a spatially nonlocal operator and showing that the theory is not fully diffeomorphism invariant.
Reference graph
Works this paper leans on
-
[1]
A. Einstein, “Do gravitational fields play an essential part in the structure of the elementary particles of matter?,” Sitzungsber. Preuss. Akad. Wiss. Berlin (Math. Phys.) (1919), 349–
work page 1919
-
[2]
The Cosmological Constant Problem,
S. Weinberg, “The Cosmological Constant Problem,” Rev. Mod. Phys. 61 (1989), 1-23 doi:10.1103/RevModPhys.61.1
-
[3]
A Unimodular Theory of Canonical Quantum Gravity,
W. G. Unruh, “A Unimodular Theory of Canonical Quantum Gravity,” Phys. Rev. D40 (1989), 1048 doi:10.1103/PhysRevD.40.1048
-
[4]
Unimodular Theory of Gravity and the Cosmological Constant,
Y. J. Ng and H. van Dam, “Unimodular Theory of Gravity and the Cosmological Constant,” J. Math. Phys.32 (1991), 1337-1340 doi:10.1063/1.529283
doi:10.1063/1.529283 1991
-
[5]
TASI Lectures on the Cosmological Constant,
R. Bousso, “TASI Lectures on the Cosmological Constant,” Gen. Rel. Grav. 40 (2008), 607-637 doi:10.1007/s10714-007-0557-5 [arXiv:0708.4231 [hep-th]]
arXiv 2008
-
[6]
Howunimodulargravitytheoriesdifferfromgeneral relativity at quantum level,
R.Bufalo, M.OksanenandA.Tureanu, “Howunimodulargravitytheoriesdifferfromgeneral relativity at quantum level,” Eur. Phys. J. C75 (2015) no.10, 477 doi:10.1140/epjc/s10052- 015-3683-3 [arXiv:1505.04978 [hep-th]]
arXiv 2015
-
[7]
The Cosmological Constant and General Covariance,
M. Henneaux and C. Teitelboim, “The Cosmological Constant and General Covariance,” Phys. Lett. B222 (1989) 195 doi:10.1016/0370-2693(89)91251-3
-
[8]
Quantum Gravity at a Lifshitz Point,
P. Hoˇ rava, “Quantum Gravity at a Lifshitz Point,” Phys. Rev. D 79 084008 (2009) [arXiv:0901.3775 [hep-th]]
arXiv 2009
Show all 28 references
-
[9]
Consistent Extension of Hoˇ rava Gravity,
D. Blas, O. Pujolas and S. Sibiryakov, “Consistent Extension of Hoˇ rava Gravity,” Phys. Rev. Lett. 104 181302 (2010) [arXiv:0909.3525 [hep-th]]
2010 arXiv
-
[10]
Modified Gravity and Cosmology,
T. Clifton, P. G. Ferreira, A. Padilla and C. Skordis, “Modified Gravity and Cosmology,” Phys. Rept. 513 (2012), 1-189 doi:10.1016/j.physrep.2012.01.001 [arXiv:1106.2476 [astro- ph.CO]]
2012 arXiv
-
[11]
Darkness without dark matter and energy — gen- eralizedunimodulargravity,
A. O. Barvinsky and A. Y. Kamenshchik, “Darkness without dark matter and energy — gen- eralizedunimodulargravity,” Phys.Lett.B 774(2017)59doi:10.1016/j.physletb.2017.09.045 [arXiv:1705.09470 [gr-qc]]
2017 arXiv
-
[12]
Canonical structure and extra mode of generalized uni- modular gravity,
R. Bufalo and M. Oksanen, “Canonical structure and extra mode of generalized uni- modular gravity,” Phys. Rev. D97 (2018) no.4, 044014 doi:10.1103/PhysRevD.97.044014 [arXiv:1712.09535 [hep-th]]
2018 arXiv
-
[13]
Dynamics of the generalized unimodular gravity theory,
A. O. Barvinsky, N. Kolganov, A. Kurov and D. Nesterov, “Dynamics of the generalized unimodular gravity theory,” Phys. Rev. D 100 (2019) no.2, 023542 doi:10.1103/PhysRevD.100.023542 [arXiv:1903.09897 [hep-th]]
2019 arXiv
-
[14]
Inflation in generalized unimodular gravity,
A. O. Barvinsky and N. Kolganov, “Inflation in generalized unimodular gravity,” Phys. Rev. D 100 (2019) no.12, 123510 doi:10.1103/PhysRevD.100.123510 [arXiv:1908.05697 [gr-qc]]. 50
2019 arXiv
-
[15]
Dynamical Structure and Definition of Energy in General Relativity,
R. L. Arnowitt, S. Deser and C. W. Misner, “Dynamical Structure and Definition of Energy in General Relativity,” Phys. Rev.116 (1959), 1322-1330 doi:10.1103/PhysRev.116.1322 ; S. Deser, R. Arnowitt and C. W. Misner, “Consistency of Canonical Reduction of General Relativity,” J...
1959 doi
-
[16]
Gravitation,
C. W. Misner, K. S. Thorne and J. A. Wheeler, “Gravitation,” W. H. Freeman, 1973, ISBN 978-0-7167-0344-0, 978-0-691-17779-3
1973
-
[17]
Evolution of dark energy recon- structed from the latest observations,
Y. Wang, L. Pogosian, G. B. Zhao and A. Zucca, “Evolution of dark energy recon- structed from the latest observations,” Astrophys. J. Lett.869 (2018), L8 doi:10.3847/2041- 8213/aaf238 [arXiv:1807.03772 [astro-ph.CO]]
2018 arXiv
-
[18]
Observa- tional Constraints on Dynamical Dark Energy Models,
O. Avsajanishvili, G. Y. Chitov, T. Kahniashvili, S. Mandal and L. Samushia, “Observa- tional Constraints on Dynamical Dark Energy Models,” [arXiv:2310.16911 [astro-ph.CO]]
-
[19]
Restricted gauge theory formalism and unimod- ular gravity,
A. O. Barvinsky and D. V. Nesterov, “Restricted gauge theory formalism and unimod- ular gravity,” Phys. Rev. D 108 (2023) no.6, 065004 doi:10.1103/PhysRevD.108.065004 [arXiv:2212.13539 [hep-th]]
2023 arXiv
-
[20]
Non-Linear Field Theories,
P. G. Bergmann, “Non-Linear Field Theories,” Phys. Rev. 75 (1949), 680-685, doi:10.1103/PhysRev.75.680 ; P. G. Bergmann and J. H. M. Brunings, “Non–Linear Field Theories II. Canonical Equations and Quantization,” Rev. Mod. Phys., 21:480, 1949, doi: 10.1103/RevModPhys.21.480
1949 doi
-
[21]
Generalized Hamiltonian dynamics,
P. A. M. Dirac, “Generalized Hamiltonian dynamics,” Can. J. Math.2 (1950), 129-148, doi:10.4153/CJM-1950-012-1
1950 doi
-
[22]
Quantization of gauge systems,
M. Henneaux and C. Teitelboim, “Quantization of gauge systems,” Princeton, USA: Univ. Pr. (1992) 520 p
1992
-
[23]
Gauge Algebra and Quantization,
I. A. Batalin and G. A. Vilkovisky, “Gauge Algebra and Quantization,” Phys. Lett. B102 (1981), 27-31 doi:10.1016/0370-2693(81)90205-7
1981 doi
-
[24]
On the reciprocal of the general algebraic matrix,
Moore, E. H. (1920). “On the reciprocal of the general algebraic matrix,” Bulletin of the American Mathematical Society. 26 (9): 394–95. doi:10.1090/S0002-9904-1920-03322-7 ; Penrose, R. (1955). “A Generalized Inverse for Matrices,” Mathematical Proceedings of the Cambridge Ph...
1920 doi
-
[25]
Operational quantization of dynamical systems subject to second class constraints
Batalin I.A. and Fradkin E.S., “Operational quantization of dynamical systems subject to second class constraints”, Nucl. Phys. B279 (1987) 514; Batalin I.A. and Tyutin I.V., “Existence theorem for the effective gauge algebra in the generalized canonical formalism with abelian...
1987
-
[26]
Non-Abelian conversion and quan- tization of non-scalar second-class constraints,
I. Batalin, M. Grigoriev and S. Lyakhovich, “Non-Abelian conversion and quan- tization of non-scalar second-class constraints,” J. Math. Phys. 46 (2005), 072301 doi:10.1063/1.1935430 [arXiv:hep-th/0501097 [hep-th]]
2005 arXiv
-
[27]
General conversion method for constrained systems,
I. A. Batalin and P. M. Lavrov, “General conversion method for constrained systems,” Phys. Lett. B 787 (2018), 89-93 doi:10.1016/j.physletb.2018.10.046 [arXiv:1808.04528 [hep-th]]. 51
2018 arXiv
-
[356]
Lorentz et al
Translated and included in The Principle of Relativity, by H.A. Lorentz et al. (Dover Press, New York, 1923)
1923
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.