Covariant scalar-tensor theories beyond second derivatives

arXiv:2604.09170 · hep-th, astro-ph.CO, gr-qc · Submitted 2026-04-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.

Kai: We're looking at the paper: "Covariant scalar-tensor theories beyond second derivatives".

Mira: Here's the gist of the paper. We propose a covariant, gauge-independent construction of foliation-based scalartensor theories, yielding diffeomorphism-invariant operators involving only gradients on the hypersurfaces where the scalar field is constant, assumed to be spacelike Construction of Invariants The strategy involves developing a covariant operator basis adapted to the foliation defined by the scalar field ϕ (assuming ϕµ ≡ ∇µϕ).

Kai: This is done order by order in derivatives of ϕ, identifying diffeomorphism scalars that encode only intrinsic derivative data on the hypersurfaces where ϕ = const.

Mira: The construction proceeds as follows: 1.

Kai: At zeroth order in ϕµν, the independent scalar invariants are given by ϕ and X ≡ gµνϕµϕν (Page 4).

Title and authors: Kai: So, what the paper actually does is show how to construct these operators systematically, order by order in the derivatives of phi mu, making sure they are diffeomorphism invariant and only depend on the intrinsic data on those constant scalar field hypersurfaces.

Mira: They then build a minimally coupled model using an arbitrary function of these invariants that can depend on both the parity-even building blocks and some parity-odd ones.

Lev: The key insight here is how they compare this construction to existing frameworks, specifically DHOST theories and U-DHOST theories, which are usually defined by certain degeneracy conditions on the coefficients of the Lagrangian terms.

Kai: What they do is show that their framework goes beyond those degenerate DHOST theories and provides a nonlinear extension of U-DHOST directly in a covariant form, without needing to start with unitary gauge or imposing degeneracy right away.

Mira: They use the Hamiltonian constraint structure to analyze the dynamics, and they find that when you look at linear cosmological perturbations around a flat Friedmann-Lemaître-Robertson-Walker background, all those higher derivative invariants vanish on that homogeneous background.

Lev: That vanishing on the homogeneous background is important because it means these higher derivative terms don't affect the overall smooth expansion of the universe; they only show up when you look at fluctuations.

Kai: And for those fluctuations, they find that the theory propagates three degrees of freedom: two tensor modes and one scalar mode.

Mira: The paper shows that this propagation is consistent with what we expect from general relativity, but it allows for extra scalar modes that arise from the foliation structure itself.

The paper's summary: Lev: One big improvement they suggest is how they handle the higher derivative terms in perturbation theory. They track the parametric orders very carefully to show exactly which invariants contribute at which order.

Kai: For example, they point out that the invariant B starts contributing at quadratic order because it’s built from second derivatives of phi mu, while C one two up to third order because they involve more derivatives.

Mira: This level of detail is crucial for anyone trying to predict the actual behavior of fluctuations in a specific scenario; you need to know exactly when the corrections kick in.

Lev: And another thing they do is provide a way to check the algebraic independence of these building blocks using field configuration analysis, which helps determine if any parity-odd pseudoscalars beyond D odd are actually independent or just redundant.

Kai: That’s helpful because it means we don't have to assume everything is independent; we can verify that the basis they built is as compact as possible.

Mira: They also developed a function for calculating a parameter, let's call it beta, which encodes all the effects of those higher-derivative terms on linear cosmological perturbations.

Lev: So, this beta parameter is useful because in the limit where beta equals zero, they recover the standard k-essence result that we already know from other theories.

The paper's improvements: Kai: So to wrap up on "Covariant scalar-tensor theories beyond second derivatives," they've given us a covariant and gauge independent construction for foliation based scalar tensor theories using a compact basis of invariants up to fourth order in the derivatives of phi mu.

Mira: The main implication is that this framework provides a systematic way to build models that go beyond degenerate higher-order scalar-tensor theories and U-DHOST theories without needing to start from unitary gauge.

Lev: From an engineering standpoint, it means we can predict exactly how many degrees of freedom you have—three modes total—and how those higher derivative corrections show up when you run them on a real system.

Kai: And they confirmed that in linear perturbations around a flat FLRW background, the higher derivative invariants vanish there, so we just deal with three degrees of freedom: two tensor modes and one scalar mode.

Mira: This construction is a convenient starting point for model building because it organizes all the possible interactions into this well-defined basis up to fourth order.

Lev: I just want to stress that they explicitly state their limitation: the method stops at fourth order in derivatives of phi mu, so if you need higher derivative physics, you'd need to go back and build more terms.

Kai: Yeah, exactly. This paper gives us a solid starting point for building these theories systematically and looking at phenomenology in less symmetric backgrounds.

Conclusion: Kai: So we looked at "Covariant scalar-tensor theories beyond second derivatives" and they gave us this really systematic way to build these operators, order by order in the derivatives of phi mu.

Mira: That's right, it’s a compact basis of invariants up to fourth order in the derivatives of phi mu that are diffeomorphism invariant. It’s like they mapped out all the possible terms you could have without repeating yourself.

Lev: From an engineering side, I see what they did with the Hamiltonian analysis in unitary gauge—it shows that for these theories, you're at most three degrees of freedom in your phase space for those tensor modes and one scalar mode.

Kai: And they confirmed that when you look at linear perturbations around a flat Friedmann-Lemaître-Robertson-Walker background, all those higher derivative invariants just vanish on the homogeneous background.

Mira: That’s significant because it means these higher derivative terms don't affect the smooth expansion of the universe; they only show up when you look at fluctuations.

Lev: So for someone building real hardware, that vanishing on the background is good news because it keeps your initial conditions manageable, and you know where the effects are going to land—specifically in those fluctuations.

Kai: The quadratic action for those scalar perturbations has two tensor modes with the standard dispersion relation and one scalar mode that scales like a single field.

Mira: And they showed this result is consistent with what we expect from general relativity but allows for extra scalar modes that come from the foliation structure itself.

Lev: It’s a convenient starting point for model building, I agree, because you have this whole basis to start with when you want to incorporate non-minimal curvature couplings or look at less symmetric backgrounds.

Kai: Exactly. The construction provides a systematic way to organize these operators up to fourth order in the derivatives of phi mu.

Mira: It’s a solid foundation for systematic model building, and they proved the algebraic independence of their basis elements by showing that the parity-even set is independent on generic branches when you include those odd terms.

Lev: That proof about algebraic independence is important because it tells us that we don't have to worry about redundant terms if we stick to this structure.

Kai: So, "Covariant scalar-tensor theories beyond second derivatives" gives us a clear roadmap for building these theories systematically and understanding their propagation.

Mira: It’s a good piece of work that clarifies the relationship between these new covariant constructions and existing degenerate frameworks like DHOST theories.

Lev: They showed how the higher derivative invariants vanish on the cosmological background, so they only contribute to fluctuations, which is a key distinction from other setups.

Kai: We're done with this paper, but there’s still a lot of stuff out there—we need to keep building up these systematic tools for testing them.

Mira: Definitely. Next up, we have that paper on "Casimir multiplets suffice," which gets into strong generation in N equals four super Yang-Mills theory and Schur algebra invariants.

Mohammad Ali Gorji, Pavel Petrov, Karim Noui

Center for Theoretical Physics of the Universe, Institute for Basic Science (IBS) · Université Paris-Saclay, CNRS/IN2P3

hep-th, astro-ph.CO, gr-qc

Submitted: 2026-04-10

Updated: 2026-04-10

DOI: 10.1088/1475-7516/2026/10/089

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: We propose a covariant, gauge-independent construction of foliation-based scalartensor theories, yielding diffeomorphism-invariant operators involving only gradients on the hypersurfaces where the…

Key concepts

Foliation-based Scalar Invariants
This method uses the scalar field's gradient ($ abla_ ext{µ} ilde{\phi}$) to define surfaces (hypersurfaces) where the scalar field is constant. The invariants constructed are those that only depend on derivatives intrinsic to these surfaces, ensuring diffeomorphism invariance.
Covariant Operator Basis
The construction develops a set of operators organized by the order of derivatives in the scalar field ($\phi$). This basis systematically includes terms like $\phi$, $X$, $B$, and higher-order combinations, allowing for the systematic building of Lagrangians up to fourth derivative order.
Propagation Analysis
By examining linear perturbations around a flat cosmological background (FLRW), the theory's dynamics are tested. The results show that all higher-derivative invariants vanish on the smooth background, meaning their physical effects only appear in fluctuations, leading to three propagating degrees of freedom.
Parameter $\beta$
This parameter quantifies how much higher-derivative terms modify linear cosmological perturbations. When $\beta$ is zero, the theory recovers the well-known results from standard k-essence theories, providing a direct link between this new construction and established models.

Terminology

Summary

We propose a covariant, gauge-independent construction of foliation-based scalartensor theories, yielding diffeomorphism-invariant operators involving only gradients on the hypersurfaces where the scalar field is constant, assumed to be spacelike

Construction of Invariants

The strategy involves developing a covariant operator basis adapted to the foliation defined by the scalar field ϕ (assuming ϕµ ≡ ∇µϕ). This is done order by order in derivatives of ϕ, identifying diffeomorphism scalars that encode only intrinsic derivative data on the hypersurfaces where ϕ = const. p. 2

The construction proceeds as follows:

  1. At zeroth order in ϕµν, the independent scalar invariants are given by ϕ and X ≡ gµνϕµϕν (Page 4). p. 3

  2. At first order in ϕµν, a natural scalar built without introducing explicit connection coefficients is Y ≡ ϕµXµ (Page 4). p. 3

  3. At second order in ϕµν, the foliation-intrinsic information is packaged in the two-form Fµν ≡ ϕµXν − ϕνXµ (Page 4). p. 3

  4. The associated scalar invariant at this derivative order is uniquely obtained by taking the squared norm of this two-form, which yields the distinguished combination B ≡ (1/2)FµνFµν = XZ − Y2 (Page 4). p. 3

  5. The construction extends up to fourth order, yielding a basis of independent building blocks including ϕ, X, B, C[1,2], D[1,2,3], E[1,2,3] and E[4], which form a basis for the algebra of elementary Lagrangians up to fourth order in derivatives (Page 6). p. 5

Model Coupling and Dynamics

The minimally coupled action is constructed by allowing an arbitrary function of the parity-even building blocks in (2.18) to depend on the parity-odd invariant D[odd] (Page 7). p. 6

The resulting Lagrangian is given by S = Z d4x√−g [M2Pl2R + Pϕ, X, B, C[1,2], D[1,2,3], E[1,2,3] + P˜ϕ, X, B, C[1,2], D[1,2,3], E[10]D [odd]] (Page 7). p. 6

The Hamiltonian analysis in unitary gauge shows that the theory admits (20 − 2 · 6 − 2) = 3 degrees of freedom in the phase space corresponding to the tensor modes and one scalar mode (Page 9). p. 8

Propagation Analysis

Analyzing linear cosmological perturbations around a spatially flat FLRW background shows that all higher-derivative invariants vanish on the homogeneous background, so their first effects arise in perturbations (Page 10).

The theory propagates three degrees of freedom, two tensor modes and one scalar mode (Page 12). p. 2

The quadratic action for scalar perturbations is given by S(2) = Z d3k(2π)3 dt N a¯3 h L(2) E. p. 9H. + L(2) P i (Page 11).

The parameter β encodes all effects of the higher-derivative terms on linear cosmological perturbations such that in the limit β = 0, one recovers the well-known k-essence result [28,30] (Page 12). p. 8

Comparison with Existing Theories

The construction clarifies the relation to existing degenerate frameworks, where DHOST theories are defined by degeneracy conditions on the coefficients of L(2)i [21] (Page 7).

U-DHOST theories enlarge this landscape by requiring degeneracy only in unitary gauge formulation, meaning the extra mode is non-propagating and instead obeys an elliptic equation on each hypersurface when ∇µϕ is supposed to be timelike (Page 2). p. 1

The authors show that the higher-derivative invariants vanish on the cosmological background and contribute only to fluctuations, so that the theory propagates three degrees of freedom, two tensor modes and one scalar mode (Page 12). p. 2

Summary and Conclusions

In this work we introduced a covariant and gauge-independent construction of scalar-tensor theories endowed with a preferred foliation defined by the timelike gradient of the scalar field, ∇µϕ (Page 13). Our approach provides a systematic way to build diffeomorphism-invariant operators that are intrinsically spatial on the hypersurfaces ϕ = const, organized as a compact basis of independent invariants up to four derivatives of ϕ (Page 6). p. 0 The Hamiltonian analysis in unitary gauge shows that the theory propagates at most three degrees of freedom (Page 9). p. 0 We then explicitly confirmed this expectation by analyzing linear perturbations around a spatially flat FLRW background: all higher-derivative invariants vanish on the homogeneous background, so their first effects arise in perturbations and the quadratic action exhibits two tensor modes with the standard dispersion relation together with a single scalar mode (Page 13). p. 2 The construction provides a convenient starting point for systematic model building and for future extensions, including the incorporation of non-minimal curvature couplings and a broader investigation of phenomenology and stability in less symmetric backgrounds (Page 13).

Appendix: Independence of the basis

The proof proceeds by contradiction, assuming there exists a nontrivial algebraic relation among the basis elements that is identically satisfied for arbitrary field configurations (Page 15). This leads to the conclusion that on the generic nondegenerate branch where this 3 × 3 minor is non-vanishing, the parity-even basis (2. p. 518) is algebraically independent, whereas D[odd] supplies the additional parity-odd information (Page 15). p. 5 This completes the proof (Page 15). p. 5

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

1

Improve mathematical derivation robustness by explicitly defining all coordinate transformations used when moving between covariant and unitary gauge formulations to ensure that gauge dependence is clearly isolated from physical degrees of freedom. This will allow AI systems to perform more reliable analysis across different theoretical frameworks.

2

Enhance the ability of the system to handle higher-order derivative terms in perturbation theory by rigorously tracking parametric orders and their contributions at each order, as demonstrated in Section 4.2. The AI can then accurately predict which higher-derivative invariants contribute only at specific perturbation orders, such as how B starts at quadratic order while C[1] is cubic.

3

Implement a mechanism for checking algebraic independence of the basis elements defined in Section 2 based on the field configuration analysis in Appendix A. This will allow the AI to verify if any parity-odd pseudoscalars beyond D[odd] are genuinely independent or redundant on generic branches.

4

Develop a function for calculating the parameter beta and its relation to the standard k-essence result, as described in Section 4.2. This enables the AI to quickly assess whether a given theory recovers standard kinetic coefficients in specific limits, such as when beta equals zero or when certain field derivatives vanish.

Sources

Related papers