Generalization of the 2-form interactions

Georg Trenkler; Lavinia Heisenberg

REVIEW 3 major objections 4 minor 1 cited by

Generalization of the 2-form interactions

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read In four dimensions a massless two-form has no Galileon-like self-interactions or non-minimal gravitational couplings, while a massive two-form admits only $L_2$ and $L_4$ self-interactions and one non-minimal coupling, to the double dual…

desk verdict A useful parity-even classification of massive 2-form Galileon-like interactions, but the negative claims are conditional on a two-Levi-Civita ansatz and miss a concrete parity-odd counterexample. read the letter →

arxiv 1908.09328 v1 pith:C5XMTTOM submitted 2019-08-25 hep-th gr-qc

Lavinia Heisenberg , Georg Trenkler This is my paper · ORCID

classification hep-thgr-qc

keywords two-formfieldsGalileoninteractionsmassive2-formmasslessLevi-Civitaconstructionnon-minimalcouplingdoubledualRiemanntensorsecond-orderequationsofmotion

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

The paper asks how far derivative self-interactions of antisymmetric two-form fields can go in four spacetime dimensions before the equations of motion stop being second order. It claims that a massless two-form admits no Galileon-like self-interactions and no non-minimal coupling to gravity, while a massive two-form admits exactly two Galileon-like interactions, $L_2$ and $L_4$, and exactly one non-minimal coupling to gravity, through the double dual Riemann tensor $\mathcal{L}^{\mu\nu\alpha\beta}B_{\mu\nu}B_{\alpha\beta}$. The classification comes from a systematic ansatz $f(B^2)\,\epsilon\epsilon(\partial B)^m B^n$ in which all derivative indices are contracted through two Levi-Civita tensors, together with the requirement that non-minimal couplings involve only divergenceless tensors. If the classification is right, it constrains the building blocks available for modified-gravity models built from massive two-forms and sharpens the analogy with massive vector theories.

What carries the argument

Two pieces of machinery carry the argument. The first is the Levi-Civita construction: a systematic ansatz $f(B^2)\,\epsilon\epsilon(\partial B)^m B^n$ in which every derivative index is contracted through two totally antisymmetric Levi-Civita tensors; index counting then decides which orders survive, giving $m=2$ with $n=0,1,2$ as the only possibilities and forcing $L_3=0$ and $L_{i\ge5}=0$. The second is the divergenceless-tensor condition for non-minimal couplings: to keep equations of motion second order, the two-form can couple only to tensors whose divergence vanishes, namely the metric, the Einstein tensor, and the double dual Riemann tensor, and the paper checks that no product of these tensors preserves that property. The double dual Riemann tensor $\mathcal{L}^{\mu\nu\alpha\beta}$ emerges as the unique surviving gravitational object for the massive case.

What would settle it

Exhibit an explicit Lorentz-invariant four-dimensional Lagrangian for a massive two-form containing a term cubic in $\partial B$ with second-order equations of motion that cannot be rewritten in the $\epsilon\epsilon$ form, or exhibit a massless-two-form non-minimal coupling to gravity built from the Einstein and double dual Riemann tensors whose equations of motion remain second order. Either would break the claimed classification.

Watch

Extended reading notes

Core claim

The core discovery is a classification. In four dimensions the Galileon-like derivative self-interactions of a massive antisymmetric two-form terminate at fourth order: the cubic interaction vanishes identically ($L_3=0$) because one Levi-Civita tensor has four indices while $\partial_\alpha B_{\mu\nu}$ has three, and all interactions of order five and higher vanish because two Levi-Civita tensors carry only eight indices while $(\partial B)^3$ carries nine. The genuinely new fourth-order interaction is a modified kinetic term multiplied by a function of $B^2$, together with variants in which additional powers of $B$ are contracted among themselves. For the massless gauge-invariant field the construction re-establishes a no-go: only the standard kinetic term survives. In the gravitational sector, the paper argues that no non-minimal coupling exists for the massless field, because the Einstein tensor and the double dual Riemann tensor are divergenceless only individually, not in the products that index contractions require, while the massive field has a unique non-minimal coupling $\sqrt{-g}\,\mathcal{L}^{\mu\nu\alpha\beta}B_{\mu\nu}B_{\alpha\beta}$.

Load-bearing premise

The load-bearing premise is that the ansatz $f(B^2)\,\epsilon\epsilon(\partial B)^m B^n$, with all derivative indices contracted through the two Levi-Civita tensors, exhausts the possible Galileon-like interactions; the paper proves this series terminates, but it does not prove that other index contractions or covariant-derivative couplings are impossible.

Editorial extensions

If this is right

Massless antisymmetric two-forms in four dimensions keep only their standard kinetic term; any gauge-invariant Galileon-like extension would require at least seven spacetime dimensions.
Massive two-forms have no Galileon-like self-interactions beyond $L_2$ and the modified kinetic term $L_4$; no cubic or higher-order derivative self-interactions exist within the Levi-Civita construction.
The unique non-minimal coupling of a massive two-form to gravity is $\sqrt{-g}\,\mathcal{L}^{\mu\nu\alpha\beta}B_{\mu\nu}B_{\alpha\beta}$, and promoting $L_4$ to curved spacetime requires compensating non-minimal terms such as $f_4(B^2)R$ on symmetric backgrounds.
Dressing the unique non-minimal coupling with a function of $B^2$ requires simultaneous $B^2(\partial B)^2$ interactions, mirroring the compensation structure found in massive vector theories.

Reading between the lines

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

Because the massive two-form is dual to a massive vector field, the unique non-minimal coupling may be the two-form avatar of the known non-minimal vector-curvature couplings; tracing the duality explicitly could predict the corresponding vector-side coupling.
The termination at $L_4$ is a statement about the $\epsilon\epsilon$ ansatz, not about all Lorentz-invariant constructions; interactions with derivative indices contracted among themselves, or with covariant derivatives in curved spacetime, remain open possibilities.
The massless no-go suggests that any apparent non-minimal coupling obtained through duality must be an infinite resummation of curvature terms rather than a finite polynomial, which is testable by expanding such a candidate interaction order by order.
Testing the unique massive coupling on backgrounds less symmetric than the maximally symmetric ones considered here could reveal whether the expected three two-form plus two graviton degrees of freedom survive.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

The paper gives a clean, readable classification of massive 2-form derivative self-interactions, but only inside a parity-even two-Levi-Civita ansatz. The central negative claims — L3=0, Li=0 for i≥5, and Eq. (37) as the only surviving m=2,n=1 interaction — do not survive contact with a parity-odd single-epsilon candidate. The abstract's "only L2 and L4" wording is too strong.

What is genuinely new and good: the L4 modified kinetic term f4(B^2)(∂μBμν ∂αBν α + ∂νBμα∂αBμν) in Eq. (36), and the unique non-minimal coupling √−g L^{μναβ}B_{μν}B_{αβ} in Eq. (40) for the massive 2-form. The index-counting argument for L3=0 is clean, and the termination at L4 is sound within the stated f(B^2)εε(∂B)^m B^n series. The decoupling-limit consistency check for Eq. (40) is a nice supporting argument. The paper also honestly flags its own limits, including that terms with (∂B)^m indices contracted among themselves are outside the construction.

The soft spots are real but addressable. First, the exhaustiveness gap is not merely formal. Take L_cand = ε^{μνρσ}∂μB_{να}∂ρB_{σβ}B^{αβ}. It has one Levi-Civita tensor, the same m=2,n=1 index content, is not a total derivative (e.g. B12=x0, B23=x1 gives a nonzero density), and its Euler-Lagrange equations contain at most second derivatives. It is not a linear combination of Eqs. (30)–(38), which all use two Levi-Civita tensors. The paper neither considers single-epsilon contractions nor states a parity-even restriction in the massive section, so the claim that Eq. (37) is the only surviving L4 interaction is incomplete. Second, the massless non-minimal no-go is supported by examples and the phrase "we were not able," not by a general proof; plausible, but not airtight. Third, the curved-spacetime tuning in Eq. (42) is asserted without derivation, and the authors explicitly defer the Hamiltonian analysis. These are gaps, not fatal contradictions.

Citation pattern is fine: self-citations point to the generalized Proca work the construction builds on. No fitted parameters or data, so circularity is not an issue.

Bottom line: this is a worthwhile paper for people working on massive 2-form theories, but the classification should be labelled parity-even or else extended to the single-epsilon sector. I would send it to a serious referee with a request to tighten the exhaustiveness statement, not desk-reject it.

Referee Report

3 major / 4 minor

Summary. The paper studies derivative self-interactions and non-minimal gravitational couplings for massless and massive 2-forms in four dimensions. It argues that, within a systematic construction based on contractions with two Levi-Civita tensors, the massless 2-form admits no Galileon-like self-interactions and no non-minimal couplings to gravity, while the massive 2-form admits only L2 and L4 interactions (with L3=0 and Li=0 for i>=5), together with a unique non-minimal coupling to the double dual Riemann tensor given in Eq. (40). The paper also presents a curved-spacetime promotion of L4 in Eq. (42) with a specific relative tuning coefficient, and supports the L4 and non-minimal-coupling identifications with a Stueckelberg/decoupling-limit argument.

Significance. If taken as a statement within the explicit two-epsilon ansatz, the paper is a useful and largely correct contribution: the index-counting arguments for L3=0 and Li=0 for i>=5 are clean, the reduction of the L4 contractions in Eqs. (29)-(36) is explicit, and the identification of the unique non-minimal coupling (40) is interesting and receives plausible decoupling-limit support. A notable strength is the paper's honesty about what it cannot construct, which is rare in this literature. However, the absence of a completeness proof for the ansatz, the unstated parity-even restriction in the massive section, and the unsupported 'impossible' language in the massless no-go section mean that the global classification claims are not established. The value of the paper lies in the explicit partial construction and the concrete coupling (40), not in a definitive no-go theorem.

major comments (3)

[III.B, Eqs. (28), (37), (39)] The negative claims L3=0, Li=0 for i>=5, and the uniqueness of Eq. (37) are proved only within the ansatz f(B^2) ǫǫ(∂B)^m B^n in which all derivative indices are contracted through two Levi-Civita tensors. The paper itself states in Section III.B that it 'cannot grasp terms where the indices of (∂B)^m are contracted among themselves', and the massive section does not state a parity-even restriction. The concern is not purely formal: the single-epsilon, parity-odd term ε^{μνρσ}∂_μ B_{να}∂_ρ B_{σβ}(B^2)^{αβ} lies outside the ansatz, has at most second-order Euler-Lagrange equations, and is not a linear combination of Eqs. (30)-(38). (The simpler candidate ε^{μνρσ}∂_μ B_{να}∂_ρ B_{σβ}B^{αβ} actually vanishes identically by a block-swap antisymmetry argument, but the (B^2)^{αβ} version does not.) The classification should be explicitly labelled as valid within the parity-even two-epsilon ansatz, or the analysis should be extended to cover single-epsilon contractions; as written, the abstract and conclusion overstate the result.
[II.C] The massless no-go for non-minimal couplings is phrased as 'it is impossible to contract the gauge invariant field strength of the 2-form with a divergenceless tensor', but the evidence is a set of examples with contractions involving G and L and the phrase 'we were not able'. This does not prove exhaustiveness. The paper itself cites [12] as a construction of a non-minimal coupling for a gauge-invariant 2-form and reconciles it only through an infinite-series/inverse-Einstein-tensor structure, which shows that the example-based argument is not a general proof. The 'impossible' statement should be weakened to 'no such coupling was found within the contractions considered', or a systematic argument covering all possible index structures should be provided.
[III.C, Eq. (42)] The curved-spacetime action for L4 is introduced with 'the analysis for cosmological backgrounds yields the following relative tuning', but no derivation is shown and no reference is given for the coefficient 3 in front of the kinetic composite. This equation is the only explicit proposal for promoting L4 to curved spacetime, and the relative tuning is essential for preserving the claimed 2+3 degrees of freedom. A derivation, or at least a reference to a companion paper, should be supplied; as written, Eq. (42) is an unsupported quantitative claim.

minor comments (4)

[Abstract and Introduction] The field is spelled 'Kalb-Rammond' in the abstract and introduction; the standard spelling is 'Kalb-Ramond'.
[II.C and III.B] The word 'divergeceless' appears repeatedly; it should be 'divergenceless'.
[III.B, Eqs. (29)-(38)] The notation L^{(0B)}_4, L^{(1B)}_4, L^{(2B)}_4 is not defined; the superscript indicates the power n of B in the ansatz and should be explained.
[References] References [11] and [13] contain corrupted author names ('Gmrkolu' and 'Beltrn Jimnez'), and the 'PACS numbers:' line in the header is empty; a final reference and metadata cleanup is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the classification is explicitly conditional on the stated Levi-Civita ansatz, and the negative conclusions are honestly phrased as 'unable to construct' statements.

full rationale

The derivation is self-contained relative to its stated scope. Section III B declares the ansatz explicitly: 'we can perform a series expansion order by order in powers of the fundamental object ∂αBµν contracted with the Levi-Civita tensors, in the schematic form f (B2)ǫǫ(∂B )mBn'. The termination claim (39), 'Li = 0 for i ≧ 5', follows from the index count inside that ansatz (two Levi-Civita tensors carry eight indices while (∂B)^3 has nine), so it is a counting result for the class defined, not an output secretly fed back as an input. The paper also concedes the limitation: 'we cannot grasp terms where the indices of (∂B )m are contracted among themselves', which is an incompleteness caveat, not a circular step. The non-minimal coupling claims are supported by external no-go theorems for the massless spin-1 and massless 2-form sectors (Refs. [4] and [11]), and the self-citations to generalized Proca work are motivational rather than load-bearing; no fitted parameter or measured datum is used, and no result is renamed as a prediction. The central weakness is the unproven exhaustiveness of the two-epsilon ansatz, but that is a correctness/scope risk, not circularity.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The paper's central claims rest on the divergence-free tensor criterion, the assumed exhaustiveness of the Levi-Civita ansatz, and cited no-go theorems for massless fields. The only hand-tuned number is the coefficient 3 in the curved-spacetime promotion of L4; no data are fitted and no new entities are introduced.

free parameters (1)

Relative tuning coefficient 3 in Eq. (42) = 3
Chosen by hand so that the Ricci-scalar term f4(X)R cancels contributions from the 3 f4,X(...) term and keeps gauge modes non-dynamical on cosmological backgrounds; no derivation from first principles is given.

assumptions (5)

standard math The Einstein tensor and the double dual Riemann tensor are divergence-free on their own.
Follows from the Bianchi identities in Riemannian geometry; used in Section II.C as the criterion that allows integration by parts without producing higher-order equations of motion.
domain assumption A healthy non-minimal coupling must be built only from divergence-free tensors so that integration by parts does not put extra derivatives on the metric.
This is the standard Horndeski/generalized Proca criterion, stated in Section II.C; the paper does not prove it from first principles but relies on it to exclude couplings.
ad hoc to paper The Levi-Civita series f(B^2) epsilon epsilon (dB)^m B^n, with derivative indices contracted through the Levi-Civita tensors, exhausts all possible Galileon-like self-interactions.
This exhaustiveness is the load-bearing assumption for L3=0 and Li=0 for i>=5 (Eqs. 28, 39). It is asserted, not proved, and the paper does not discuss alternative contractions.
domain assumption The massless 2-form and massless spin-1 fields have no Galileon-like self-interactions in D=4 when gauge invariance is preserved.
Imported from Refs. [4] and [11] and used in the decoupling-limit arguments in Sections III.B and III.C to rule out certain L4 and non-minimal structures.
standard math The dualities: massless 2-form dual to a scalar, massive 2-form dual to a massive vector, hold in four dimensions.
Standard Hodge duality in D=4, used in Sections II and III.A to count degrees of freedom and to justify the decoupling limit.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Generalization of the 2-form interactions." pith.science (2026). https://pith.science/paper/C5XMTTOM

@misc{pith2026190809328,
  author       = {Pith},
  title        = {Pith review of: Generalization of the 2-form interactions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C5XMTTOM}},
  note         = {Machine review of arXiv:1908.09328}
}

abstract

We systematically construct derivative self-interactions for massless and massive 2-forms. There exists a no-go theorem in the literature for constructing Galileon-like Lagrangians in four dimensions for the 2-form with gauge invariance, the Kalb-Rammond field. The presence of non-minimal couplings strongly relies on the contraction with divergenceless tensors. In four dimensions these are the Einstein tensor and the double dual Riemann tensor. Even though they are divergenceless on their own, their combination ceases to be. In the case of massless 2-forms we are not able to establish non-minimal couplings of the 2-form to the gravity sector with second order equations of motion due to the impossibility of building consistent combinations of divergenceless tensors. Using the systematical construction in terms of the Levi-Civita tensor, we aim at constructing Galileon-like derivative self-interactions for the massive 2-form. Apart from $L_2$ and $L_4$ we are not able to construct further Galileon-like Lagrangians. For the massive case, an important non-minimal coupling between the 2-form and the double dual Riemann tensor arises, which receives additional support from the decoupling limit. Promoting the interactions in $L_4$ requires the presence of appropriate non-minimal couplings and we give concrete examples for this.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Degenerate higher-order Maxwell-Einstein theories
gr-qc 2025-02 conditional novelty 7.0 of 10

A complete classification of quadratic degenerate Maxwell-Einstein theories is given, including a new theory that generalizes Horndeski's non-minimal coupling to gauge fields.

Reference graph

Works this paper leans on

14 extracted references · 1 canonical work pages · cited by 1 Pith paper

[12]

Yoshida, arXiv:1906.02462 [hep-th]

D. Yoshida, arXiv:1906.02462 [hep-th]

arXiv 1906
[1]

G. R. Dvali, G. Gabadadze and M. Porrati, Phys. Lett. B 485, 208 (2000) [hep-th/0005016]

arXiv 2000
[2]

Nicolis, R

A. Nicolis, R. Rattazzi and E. Trincherini, Phys. Rev. D 79 (2009) 064036 [arXiv:0811.2197 [hep-th]]

arXiv 2009
[3]

G. W. Horndeski, Int. J. Theor. Phys. 10, 363 (1974); C. Deﬀayet, S. Deser and G. Esposito-Farese, Phys. Rev. D 80 (2009) 064015 [arXiv:0906.1967 [gr-qc]]; C. Def- fayet, X. Gao, D. A. Steer and G. Zahariade, Phys. Rev. D 84, 064039 (2011) [arXiv:1103.3260 [hep-th]]

arXiv 1974
[4]

Deﬀayet, S

C. Deﬀayet, S. Deser and G. Esposito-Farese, Phys. Rev. D 82, 061501 (2010) [arXiv:1007.5278 [gr-qc]]

arXiv 2010
[5]

Deﬀayet, S

C. Deﬀayet, S. Mukohyama and V. Sivanesan, arXiv:1601.01287 [hep-th]

arXiv
[6]

Heisenberg, JCAP 1405, 015 (2014) [arXiv:1402.7026 [hep-th]]

L. Heisenberg, JCAP 1405, 015 (2014) [arXiv:1402.7026 [hep-th]]

arXiv 2014
[7]

Allys, P

E. Allys, P. Peter and Y. Rodriguez, JCAP 1602, no. 02, 004 (2016) [arXiv:1511.03101 [hep-th]]

arXiv 2016

Show all 14 references

[8]

Beltran Jimenez and L

J. Beltran Jimenez and L. Heisenberg, Phys. Lett. B 757, 405 (2016) [arXiv:1602.03410 [hep-th]]

2016 arXiv
[9]

Tasinato, JHEP 1404, 067 (2014) [arXiv:1402.6450 [hep-th]]; A

G. Tasinato, JHEP 1404, 067 (2014) [arXiv:1402.6450 [hep-th]]; A. De Felice, L. Heisenberg, R. Kase, S. Muko- hyama, S. Tsujikawa and Y. l. Zhang, JCAP 1606, no. 06, 048 (2016) [arXiv:1603.05806 [gr-qc]]; A. De Fe- lice, L. Heisenberg, R. Kase, S. Mukohyama, S. Tsu- jikawa and...

2014 arXiv
[10]

Heisenberg, Phys

L. Heisenberg, Phys. Rept. 796, 1 (2019) [arXiv:1807.01725 [gr-qc]]

2019 arXiv
[11]

Deﬀayet, A

C. Deﬀayet, A. E. Gmrkolu, S. Mukohyama and Y. Wang, JHEP 1404 (2014) 082 [arXiv:1312.6690 [hep- th]]

2014 arXiv
[13]

Beltrn Jimnez, C

J. Beltrn Jimnez, C. de Rham and L. Heisenberg, arXiv:1906.04805 [hep-th]

1906 arXiv
[14]

Gabadadze, K

G. Gabadadze, K. Hinterbichler, J. Khoury, D. Pirt- skhalava and M. Trodden, Phys. Rev. D 86, 124004 (2012) [arXiv:1208.5773 [hep-th]]

2012 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [12]

Yoshida, arXiv:1906.02462 [hep-th]

D. Yoshida, arXiv:1906.02462 [hep-th]

arXiv 1906

[2] [1]

G. R. Dvali, G. Gabadadze and M. Porrati, Phys. Lett. B 485, 208 (2000) [hep-th/0005016]

arXiv 2000

[3] [2]

Nicolis, R

A. Nicolis, R. Rattazzi and E. Trincherini, Phys. Rev. D 79 (2009) 064036 [arXiv:0811.2197 [hep-th]]

arXiv 2009

[4] [3]

G. W. Horndeski, Int. J. Theor. Phys. 10, 363 (1974); C. Deﬀayet, S. Deser and G. Esposito-Farese, Phys. Rev. D 80 (2009) 064015 [arXiv:0906.1967 [gr-qc]]; C. Def- fayet, X. Gao, D. A. Steer and G. Zahariade, Phys. Rev. D 84, 064039 (2011) [arXiv:1103.3260 [hep-th]]

arXiv 1974

[5] [4]

Deﬀayet, S

C. Deﬀayet, S. Deser and G. Esposito-Farese, Phys. Rev. D 82, 061501 (2010) [arXiv:1007.5278 [gr-qc]]

arXiv 2010

[6] [5]

Deﬀayet, S

C. Deﬀayet, S. Mukohyama and V. Sivanesan, arXiv:1601.01287 [hep-th]

arXiv

[7] [6]

Heisenberg, JCAP 1405, 015 (2014) [arXiv:1402.7026 [hep-th]]

L. Heisenberg, JCAP 1405, 015 (2014) [arXiv:1402.7026 [hep-th]]

arXiv 2014

[8] [7]

Allys, P

E. Allys, P. Peter and Y. Rodriguez, JCAP 1602, no. 02, 004 (2016) [arXiv:1511.03101 [hep-th]]

arXiv 2016