REVIEW 3 major objections 4 minor 26 references
Trace-free Einstein gravity as two interacting constrained $BF$ theories
T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash
Pith's one-line read This paper reports two fully diffeomorphism-invariant actions whose nondegenerate solutions are exactly the equations of trace-free Einstein gravity, with the cosmological constant emerging as an integration constant from the Bianchi…
desk verdict Two new complex BF actions for trace-free Einstein gravity: the core derivation is sound, but the degenerate sector and the hand-imposed reality condition are understated and need fixing. 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
BF theory is the class of actions built from the wedge product of a 2-form $B$ with a curvature $F$. The central machinery here is a pair of BF-type actions, Eqs. (15) and (43), in which a 2-form $\Sigma^i$ and connection $A^i$ are supplemented by symmetric-matrix Lagrange multipliers $\Psi_{ij}$ (or $\Phi_{ij}$) and 4-form multipliers $\rho$, $\nu$, $\mu$. The interaction term $\rho(\mathrm{Tr}\Psi - \mathrm{Tr}\bar{\Psi})$ in the first action couples the two copies and prevents the degeneracy that makes the standalone Husain–Kuchař model lack gravity; in the second action the $\mu$-term imposes the reality condition $\Sigma^i\wedge\Sigma^i + \bar{\Sigma}^i\wedge\bar{\Sigma}^i = 0$ explicitly. Combining the equations of motion recovers the trace-free Einstein equations through the self-dual/anti-self-dual splitting of the Riemann tensor, while the Bianchi identity converts the trace condition into $dR=0$.
What would settle it
Perform the Dirac constraint analysis of the first action and check whether the reality condition $\Sigma^i\wedge\Sigma^j=0$ is preserved by the total Hamiltonian and whether the $\rho=0$ sector is dynamically inaccessible; if the phase space has additional propagating degrees of freedom, or if real Lorentzian configurations are not selected by the dynamics, the claimed equivalence to trace-free Einstein gravity fails.
Extended reading notes
Core claim
The central discovery is that the action $S = \int[\Sigma^i\wedge F_i - \tfrac{1}{2}\Psi_{ij}\Sigma^i\wedge\Sigma^j + \text{conjugate} - \rho(\mathrm{Tr}\Psi - \mathrm{Tr}\bar{\Psi})]$ is a fully diffeomorphism-invariant action for trace-free Einstein gravity: its field equations, supplemented by the reality condition $\Sigma^i\wedge\Sigma^j=0$ and the nonvanishing of $\rho$, imply $F=E$, $Q=H$, $E=E^T$, $H=H^T$, and $\mathrm{Tr}H=0$, which are exactly the trace-free Einstein equations. The same is shown for the equivalent second action built from two chiral Plebanski copies. The trace of the equations, combined with the Bianchi identity, yields $dR=0$, so the Ricci scalar is constant and $4\Lambda$ enters purely as an integration constant. The paper stresses that no nondynamical fields and no unimodular condition are used, and that the only gravitational sector is trace-free Einstein gravity.
Load-bearing premise
Both derivations assume, rather than derive, the reality condition $\Sigma^i\wedge\Sigma^j=0$, and they exclude the degenerate sectors where the 4-form $\rho$ (or $\nu$) vanishes; if those sectors are admitted, or if the reality condition cannot be made consistent, the actions describe a complexified or degenerate theory instead of Lorentzian trace-free Einstein gravity.
Editorial extensions
If this is right
- Trace-free Einstein gravity now has two concrete, fully diffeomorphism-invariant BF-type actions with no general-relativity sector, making it a cleaner alternative to unimodular gravity for addressing vacuum energy.
- The cosmological constant emerges as an integration constant via $dR=0$, realizing Einstein's original proposal without imposing a unimodular condition.
- The BF form opens a route to canonical and path-integral quantization of trace-free Einstein gravity along the lines already developed for Plebanski-type actions.
- The first action needs only one externally imposed reality condition because the other emerges from the equations, reducing the hand-imposed assumptions relative to the full Plebanski formulation.
- The second action shows that the reality condition can double as the interaction that excludes general relativity from the theory.
Reading between the lines
- The paper leaves open whether the reality condition $\Sigma^i\wedge\Sigma^j=0$ and nonvanishing $\rho,\nu$ are dynamically enforced or must be imposed on the phase space; if a Hamiltonian analysis shows they are not preserved by evolution, any quantization must treat them as external constraints, which would weaken the claim that the action alone defines the theory.
- Because $\rho$ is tied to the 4-volume form, the constraints lock a volume element; a natural extension is to couple matter and check whether energy-momentum conservation is needed to keep $dR=0$, clarifying how robust the integration-constant mechanism is.
- Both actions are non-chiral, so a chiral BF formulation of trace-free Einstein gravity may or may not exist; finding one would test whether complex self-dual variables are essential to keeping the general-relativity sector out.
- A testable extension is to restrict the path integral to nondegenerate sectors and see whether summing over the complexified solutions destabilizes the Lorentzian reality conditions, which would matter for quantum applications.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes two fully diffeomorphism-invariant actions for trace-free Einstein gravity, both using complex variables. The first action (Eq. 15) couples two copies of the constrained BF action for the Husain-Kuchař model through a term -ρ(TrΨ - TrΨ̄) with a 4-form multiplier ρ. The second action (Eq. 43) couples two copies of the chiral Plebanski action with an additional term imposing the reality condition (25). Assuming ρ ≠ 0 and the additional reality condition (26), the authors show that the equations of motion reduce to the trace-free Einstein equations (11), and that the Bianchi identity implies dR = 0, so the cosmological constant emerges as an integration constant (Eqs. 40-42 and 57-58).
Significance. The construction is attractive: it gives simple, explicitly diffeomorphism-invariant actions for trace-free Einstein gravity without unimodular fields, with the cosmological constant arising dynamically. The first action is an elegant combination of two Husain-Kuchař sectors, and the derivations are mostly standard. However, the advertised exclusion of degenerate sectors is not established, and the 'only gravitational sector' claim in the abstract needs qualification. The paper explicitly builds on the authors' earlier work and offers a concrete step toward spin-foam formulations of trace-free gravity, provided the nondegeneracy and reality issues are properly handled.
major comments (3)
- [Section 3, Introduction paragraph 'Remarkably, it is the interaction term...' and Eqs. (16)-(22)] The claim that the interaction term prevents the theory from being degenerate is contradicted by explicit solutions with ρ = 0. For example, set Σ^i = 0, Σ̄^i = 0, A^i and Ā^i flat, and take any symmetric Ψ, Ψ̄ with TrΨ = TrΨ̄. These satisfy all equations of motion (16)-(22), the interaction term vanishes, and the solutions are not trace-free Einstein. The derivation of (23)-(37) and of dTrΨ = 0 in (40) uses ρ ≠ 0; for ρ = 0 those steps fail. The paper should either prove ρ ≠ 0 from the dynamics, or explicitly restrict to the open sector ρ ≠ 0 and state that the actions as written admit additional degenerate solutions.
- [Abstract and Conclusions] The phrase 'their only gravitational sector is trace-free Einstein gravity' is too strong as stated. The equivalence proof requires two external inputs not enforced by the actions: the nonvanishing of ρ (respectively ν in the second action) and the reality condition (26), which for the first action is imposed by hand. As written, the actions also admit degenerate, non-metric solutions. The abstract and conclusions should be amended to say that the actions describe trace-free Einstein gravity on the nondegenerate sector that satisfies the reality conditions, and that additional degenerate solutions exist unless those conditions are imposed by hand.
- [Section 4, Eqs. (43)-(58)] The same degeneracy issue affects the second action. Configurations with ν = 0, Σ^i = Σ̄^i = 0, flat connections, and traceless Φ, Φ̄ satisfy (44)-(52). The interaction term -µ/2(Σ^i∧Σ_i + Σ̄^i∧Σ̄_i) imposes the reality condition (25) but does not enforce ν ≠ 0. The discussion of the 'double role' of the interaction term in preventing the description of general relativity and in introducing a reality condition should be revised to acknowledge that the nondegeneracy of ν is an additional assumption rather than a consequence of the action.
minor comments (4)
- [Eqs. (39)-(40) and (56)-(57)] The reduction from Eq. (39) to Eq. (40) and from Eq. (56) to Eq. (57) is sketched rather than fully shown. Please display the intermediate steps, especially how the second term in (39) is handled via the contracted Bianchi identity, to make the derivation transparent.
- [Section 4, meaning of Φ] In the sentence 'For κ = -1, the roles of Ψ and Ψ are exchanged,' the matrices should be Φ and Φ̄, not Ψ and Ψ̄, to match the discussion of the second action.
- [Section 3 and 4, definition of ρ and ν] The phrases 'ρ is a nonvanishing 4-form' and 'ν is a nonvanishing 4-form Lagrange multiplier' should be clarified: if nonvanishing is a condition on the configuration space, it should be stated explicitly in the definition of the action and accounted for in the variational principle; otherwise the actions admit the zero-mode solutions discussed above.
- [Eq. (24)] The sentence 'Substituting this solution in (16), the 4-form ρ acquires the following form' is slightly ambiguous because (16) was already used to obtain (23). Clarify that (24) follows by taking the trace of (16) after substituting (23).
Circularity Check
No significant circularity: the equivalence of actions (15) and (43) to trace-free Einstein gravity is derived from the equations of motion, not assumed, and the cosmological constant emerges as an integration constant.
full rationale
The paper's derivation chain is self-contained with respect to its central claims. Starting from the action (15), the equations of motion (16)-(22) are written out; the solution (23) of the algebraic simplicity constraints is a standard result cited to Plebanski and Capovilla-Dell-Jacobson-Mason, not to the authors' own conclusion, and its use is a mathematical lemma rather than a circular import. The derivation then proceeds by substituting (27) into (18), obtaining (31)-(34), and showing that (34) is equivalent to F=E and Q=H, with (22) giving Tr H=0, which is exactly the trace-free Einstein condition (11). No fitted parameter is renamed as a prediction; the cosmological constant is derived from dR=0 via the Bianchi identity, equations (38)-(42) and (57)-(58), so it is an integration constant, not an input. The reality condition (26) is explicitly imposed by hand and acknowledged as such, and the nonvanishing of rho is also stated in the definition of the action; these are external domain restrictions (needed for Lorentzian signature and nondegenerate coframes, analogous to requiring a nondegenerate tetrad in standard gravity), not hidden assumptions that make the result equivalent to its inputs. The paper does cite prior work by the same authors, notably [6] for the first-order trace-free Einstein formulation and [12] for Plebanski reality conditions, but the relevant equivalences and algebraic steps are re-derived in the text or are standard results with independent sources, so these self-citations are not load-bearing in a circular way. The degenerate rho=0 sector noted by a skeptic is excluded by the paper's explicit stipulation that rho is nonvanishing; if that stipulation is relaxed, the variational principle admits trivial solutions, but that is a stated assumption/correctness caveat rather than a circularity in the derivation.
Assumptions & free parameters
free parameters (2)
- τ =
±1
- κ =
±1
assumptions (4)
- domain assumption Torsion-free condition De^I=0 (Eq. 3)
- domain assumption Nonvanishing of the 4-form ρ (nondegenerate sector)
- domain assumption Reality condition (26), Σ^i ∧ Σ^j = 0, imposed by hand
- standard math Plebanski decomposition of 2-forms (Eq. 23)
Cite this review
Pith. "Pith review of Trace-free Einstein gravity as two interacting constrained $BF$ theories." pith.science (2026). https://pith.science/paper/4IHYC6KE
@misc{pith2026250111948,
author = {Pith},
title = {Pith review of: Trace-free Einstein gravity as two interacting constrained $BF$ theories},
year = {2026},
howpublished = {\url{https://pith.science/paper/4IHYC6KE}},
note = {Machine review of arXiv:2501.11948}
}
abstract
A theory of gravity alternative to general relativity is trace-free Einstein gravity, which has the remarkable property that the cosmological constant emerges as an integration constant. In this paper, we report two fully diffeomorphism-invariant actions for trace-free Einstein gravity. They describe the theory as two $BF$ theories supplemented with some constraints. The first action comprises two copies of the constrained $BF$ theory for the Husain-Kucha\v{r} model plus an interaction term involving the fields that impose the constraints on the $B$ fields. The second action employs two copies of the chiral Plebanski action for general relativity plus an additional constraint. Both actions use complex variables, and naturally include one of the reality conditions imposed in the Plebanski formulation of general relativity. The new actions have the advantage of not involving any nondynamical fields or unimodular condition, and their only gravitational sector is trace-free Einstein gravity.
Reference graph
Works this paper leans on
-
[1]
Einstein A 1919 Sitzungsberichte der Preussischen Akad. d. Wissen-schaft en 349–356
work page 1919
-
[2]
Einstein A, Lorentz H A, W eyl H and Minkowski H 1952 The Principle of Relativity. Notes by A. Sommerfeld. Pages 189-198 (Mineola, NY: Dover Publications)
work page 1952
-
[3]
Einstein A 1927 Mathematische Annalen 97 99–103
work page 1927
-
[4]
Ellis G F R, van Elst H, Murugan J and Uzan J P 2011 Classical and Quantum Gravity 28 225007
work page 2011
-
[5]
Ellis G F R 2014 General Relativity and Gravitation 46 1619
work page 2014
-
[6]
Montesinos M and Gonzalez D 2023 Phys. Rev. D 108 124013
work page 2023
-
[7]
Anderson J L and Finkelstein D 1971 American Journal of Physics 39 901–904
work page 1971
-
[8]
W einberg S 1989 Rev. Mod. Phys. 61 1–23
work page 1989
Show all 26 references
-
[9]
Carballo-Rubio R, Garay L J and Garc ´ ıa-Moreno G 2022 Classical and Quantum Gravity 39 243001
2022
-
[10]
Bengochea G R, Le´ on G, Perez A and Sudarsky D 2023 Journal of Cosmology and Astroparticle Physics 2023 011
2023
-
[11]
Pleba´ nski J F 1977 J. Math. Phys. 18 2511–2520
1977
-
[12]
Celada M, Gonz´ alez D and Montesinos M 2016 Classical and Quantum Gravity 33 213001
2016
-
[13]
Montesinos M and Vel´ azquez M 2009 AIP Conf. Proc. 1196 201
2009
-
[14]
Husain V and Kuchar K V 1990 Phys. Rev. D 42 4070
1990
-
[15]
Rovelli C 2004 Quantum Gravity (Cambridge University Press, Cambridge)
2004
-
[16]
Thiemann T 2007 Modern Canonical Quantum General Relativity (Cambridge University Press, Cambridge)
2007
-
[17]
Perez A 2013 Living Rev. Relativ. 16 1205
2013
-
[18]
Livine E R 2024 Spinfoam models for quantum gravity: Ove rview ( Preprint 2403.09364)
2024 arXiv
-
[19]
Capovilla R, Dell J, Jacobson T and Mason L 1991 Classical and Quantum Gravity 8 41
1991
-
[20]
Montesinos M and Gonzalez D 2024 Phys. Rev. D 110 064057
2024
-
[21]
Gielen S and Nash E 2024 Classical and Quantum Gravity 41 085009
2024
-
[22]
Montesinos M and Celada M 2020 Phys. Rev. D 101 084043
2020
-
[23]
Montesinos M, Escobedo R and Celada M 2021 General Relativity and Gravitation 53 52
2021
-
[24]
Ashtekar A and Bianchi E 2021 Reports on Progress in Physics 84 042001 Trace-free Einstein gravity as two interacting constraine d BF theories 10
2021
-
[25]
Smolin L 2009 Phys. Rev. D 80 084003
2009
-
[26]
Smolin L 2011 Phys. Rev. D 84 044047
2011
Reviewed August 10, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.