Deser–Woodard Nonlocal Gravity Theory

• Deser–Woodard Nonlocal Gravity Theory is a modified gravitational model that adds nonlocal terms via inverse d'Alembertian operators on curvature, explaining cosmic acceleration without dark energy.
• The theory employs auxiliary fields to localize the nonlocal action, enabling analytic solutions for black holes, wormholes, and nonsingular bouncing cosmologies.
• Its predictions for gravitational dynamics, cosmic structure formation, and screening effects yield observable signatures in gravitational waves while satisfying solar-system tests.

The Deser–Woodard Nonlocal Gravity Theory is a class of modifications to Einstein gravity in which inverse d'Alembertian operators act on curvature invariants, producing nonlocal terms in the gravitational action. Designed to address late-time cosmic acceleration without introducing explicit dark energy or new mass scales, the theory replaces the cosmological constant with a dynamically generated term driven by the entire spacetime curvature history. Its revised versions—particularly those incorporating multiple auxiliary fields—lead to new phenomena such as nontrivial ^^^^1^^^^ on small scales, characteristic modifications of gravitational dynamics, and analytic black-hole and wormhole solutions that deviate in quantified ways from general relativity.

1. Formal Structure: Action, Localization, and Auxiliary Fields

The revised Deser–Woodard (DW) model is constructed from the nonlocal action

$$S = \frac{1}{16\pi}\int d^4x\,\sqrt{-g}\;R\bigl[1+f(Y)\bigr],$$

where $f(Y)$ is a distortion function and the nonlocal scalar $Y$ is defined as

$$Y = \Box^{-1}(g^{\mu\nu}\partial_\mu X\,\partial_\nu X),\qquad X = \Box^{-1}R.$$

Here, $\Box \equiv g^{\mu\nu} \nabla_\mu \nabla_\nu$ is the covariant d'Alembertian.

The action is localized by introducing the four auxiliary fields $X, Y, U, V$ obeying

$$\begin{aligned} &\Box X = R,\ &\Box Y = g^{\mu\nu}\partial_\mu X\,\partial_\nu X, \ &\Box U = -2\,\nabla_\mu(V\,\nabla^\mu X),\ &\Box V = R f'(Y)/Y, \end{aligned}$$

which ensure equivalence with the original nonlocal formulation when appropriate retarded boundary conditions are imposed. The resultant local action, after integration by parts, takes the form

$$S_{\text{local}} = \frac{1}{16\pi}\int d^4x\sqrt{-g}\bigl\{ R[1+U+f(Y)] + g^{\mu\nu}\bigl(\partial_\mu X\partial_\nu U + \partial_\mu Y\partial_\nu V + V\,\partial_\mu X\partial_\nu X\bigr)\bigr\}.$$

This extended field content enables analytic handling of the theory and underpins its cosmological and astrophysical predictions (D'Agostino et al., 21 Feb 2025, D'Agostino et al., 29 Jan 2025, Ding et al., 2019).

2. Field Equations, Tetrad Simplification, and Solution Space

Variation of the localized action with respect to $g_{\mu\nu}$ yields the modified Einstein equations

$$(G_{\mu\nu}+g_{\mu\nu}\Box-\nabla_\mu\nabla_\nu)[1 + U + f(Y)] + \mathcal{K}_{(\mu\nu)} - \tfrac{1}{2}g_{\mu\nu}g^{\rho\sigma}\mathcal{K}_{\rho\sigma} = 0,$$

with $$\mathcal{K}_{\mu\nu} = \partial_\mu X\,\partial_\nu U + \partial_\mu Y\,\partial_\nu V + V\,\partial_\mu X\,\partial_\nu X$$.

Static, spherically symmetric solutions benefit from reformulation in a "proper tetrad frame," where the field equations reduce to two ordinary differential equations in the radial variable. This allows a direct construction of solutions by first specifying the $g_{tt}$ component (redshift function), then deducing $g_{rr}$ and the associated auxiliary fields. Such a procedure enables full reconstruction of $f(Y)$ given an ansatz for the metric, leading to both analytic and numerical families of solutions (D'Agostino et al., 21 Feb 2025, D'Agostino et al., 29 Jan 2025).

3. Black Holes, Wormholes, and Nontrivial Geometries

The revised DW model admits a rich spectrum of asymptotically flat static, spherically symmetric solutions. The primary black-hole metrics at first order in the nonlocal perturbation parameter $\alpha$ and integer $n>1$ are

$$A(r) = 1 - \frac{2}{r} - \frac{\alpha}{r^n},\qquad B(r) = 1 - \frac{2}{r} + \frac{\alpha}{3^n r^{n+1}(r-3)^2}\Bigl\{3^n r[n(r-3)(r-2)+4r-9]-3(r-2)(2r-3)r^n\Bigr\}.$$

Horizon, photon sphere, shadow, and ISCO radii acquire well-defined corrections, with

$$r_{\rm H} = 2 + \frac{\alpha}{2^{n-1}},\qquad r_{\rm ps} = 3\left[1+\alpha\frac{n+2}{2\cdot3^n}\right],\qquad b_c = 3\sqrt{3}\left[1+\alpha\frac{3^{1-n}}{2}\right],\qquad r_{\rm ISCO} = 6\left[1-\alpha\frac{3^{1-n}}{2^{n+2}}\right],$$

and all curvature invariants remain finite for $r > r_{\rm H}$, smoothly reducing to Schwarzschild for $\alpha \rightarrow 0$. The new black hole family is characterized by an inverse power-law correction to $g_{tt}$ and a first-order perturbation in $g_{rr}$ (D'Agostino et al., 21 Feb 2025).

The theory also possesses traversable wormhole and regular black-hole solutions. For example, a Morris–Thorne wormhole is realized with $A(r)=1$, $B(r) = (1+\mathcal{C}/r^2)^{-1}$, where $\mathcal{C} < 0$ sets the throat radius $r_{\text{thr}} = \sqrt{-\mathcal{C}}$. In the case of extremal charged black holes ($Q = M$), it allows solutions with regular curvature but divergent auxiliary fields at the outer horizon (Neves, 1 Jul 2025, D'Agostino et al., 29 Jan 2025).

4. Cosmological Phenomenology and Screening

In the cosmological sector, nonlocality is configured via $f(Y)$ to reproduce the entire $\Lambda$CDM background expansion history without a cosmological constant. $Y$, constructed as $\Box^{-1}(g^{\mu\nu}\partial_\mu X\partial_\nu X)$, encodes both cosmic and local (e.g., static) curvature gradients. This enables scale-dependent "screening": for cosmic backgrounds $Y>0$ and $f(Y)$ activates, while for bound (static) systems $Y<0$ and $f(Y)$ can be chosen to vanish, restoring GR locally (Ding et al., 2019). The sign-changing behavior of $Y$ underlies the theory’s compatibility with stringent Solar-System constraints while allowing large effects on cosmic scales.

The linear structure growth rate in the DW model is generally suppressed relative to GR and $\Lambda$CDM when the distortion function $f(Y)$ is fit to cosmic expansion data. Numerical solutions predict lower $f\sigma_8$ and consequently lower $\sigma_8$ than $\Lambda$CDM, which better matches certain weak lensing and RSD datasets. However, some variants—in particular, those with strong nonlocal effects—exhibit distinctive features in $f\sigma_8(z)$ such as sharp drops at particular redshifts, serving as observational discriminants (Ding et al., 2019, Nersisyan et al., 2017).

5. Gravitational Perturbations and Quasinormal Modes

For black holes in the revised DW theory, axial and polar gravitational perturbations each satisfy modified Regge–Wheeler– and Zerilli-type master equations. The effective potentials and their corrections are explicitly computable, and quasinormal ringdown frequencies show deviations from GR up to $\sim 12\%$ for $\alpha \sim 0.1$ and exponent $k \sim 1$ (D'Agostino et al., 2 Jul 2025, D'Agostino et al., 16 Jan 2026). Perturbative analysis reveals axial modes remain unaltered at leading order in the absence of scalar excitation, but polar modes generally differ, sourcing additional massless scalar excitations and breaking axial–polar isospectrality (Chen et al., 2021).

The detailed functional dependence of ringdown spectrum deviations on the nonlocal parameters implies that upcoming GW ringdown measurements can impose bounds in the $(\alpha,k)$ parameter space at the percent level. A deviation measure $\Gamma$ averaging over fundamental scalar, electromagnetic, and gravitational modes provides a sharp diagnostic for observational constraints (D'Agostino et al., 2 Jul 2025).

6. Solar-System Constraints and Theoretical Consistency

Solar-System experiments—light deflection, Shapiro delay, perihelion advance, geodetic precession—place stringent bounds on the magnitude and radial falloff parameter $b$ in nonlocal corrections parameterized as $f(X) = \zeta X^b$. Combined analysis yields constraints $\zeta \lesssim 2 \times 10^{-5}$ for $b = 1$, rapidly weakening with increasing $b$ (Liu et al., 11 Nov 2025). To be compatible with both cosmological predictions and local measurements, the model must invoke scale-dependent screening or function forms such that nonlocal effects are exponentially suppressed in the Solar System.

At the theoretical level, second-order perturbative analyses show that the DW nonlocal sector generically leads to divergent gravitational wave energy-momentum fluxes ($t^{0i} \sim 1/r$ at large $r$), violating asymptotic flatness and the well-tested $1/r^2$ flux law, unless the distortion function is tuned such that $f'(0) = f''(0) = 0$ (Chu et al., 2024, Chu et al., 2018). This restricts viable forms and can introduce tension between strong cosmological effects and weak local modifications.

7. Early-Universe Applications: Nonsingular Bounces and Beyond

The DW model can realize nonsingular bouncing cosmologies by reconstructing $f(Y)$ to drive a reversal of contraction to expansion. For an FLRW bounce, the distortion function is chosen such that all nonlocal terms vanish after the bounce, smoothly recovering GR. Anisotropies—measured via the shear density—grow more mildly near the bounce, controlled by nonlocal contributions; this avoids the isotropy problem. In all viable cases, $f(Y)$ is a single-valued, monotonic function ensuring ghost freedom and cosmic viability (Jackson et al., 2021, Chen et al., 2019).

Broader classes of bounces (symmetric, oscillatory, matter, finite-time singularities) have been demonstrated to be consistent with the theory, provided $H(t)$ and $\dot{H}(t)$ remain finite throughout the bounce (Jackson et al., 2021).

---

References:

• "Black hole solutions in the revised Deser-Woodard nonlocal theory of gravity" (D'Agostino et al., 21 Feb 2025)
• "Primordial bouncing cosmology in the Deser-Woodard nonlocal gravity" (Chen et al., 2019)
• "Structure formation in the new Deser-Woodard nonlocal gravity model" (Ding et al., 2019)
• "Nonlocal gravity in a proper tetrad frame: traversable wormholes" (D'Agostino et al., 29 Jan 2025)
• "Black holes and wormholes in Deser-Woodard gravity" (Neves, 1 Jul 2025)
• "Solar-system experimental constraints on nonlocal gravity" (Liu et al., 11 Nov 2025)
• "Divergent Energy-Momentum Fluxes In Nonlocal Gravity Models" (Chu et al., 2024)
• "Gravitational perturbations of nonlocal black holes" (D'Agostino et al., 16 Jan 2026)
• "Quasinormal modes of nonlocal gravity black holes" (D'Agostino et al., 2 Jul 2025)
• "Black hole quasinormal modes and isospectrality in Deser-Woodard nonlocal gravity" (Chen et al., 2021)
• "Non-local gravity in bouncing cosmology scenarios" (Jackson et al., 2021)