⚛️ general relativity

Covariant variation for point-particle Lagrangians

This paper establishes a rigorous variational framework using covariant variations and parallel transport to derive simple Lagrangians for point-particle models, including the Mathisson-Papapetrou-Dixon equations for spinning bodies and a null-particle model for light propagation, where standard variational methods require careful handling of worldline-coupled tensor fields.

Original authors: Finnian Gray, Sebastian Murk, Daniel R. Terno

Published 2026-06-25

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Finnian Gray, Sebastian Murk, Daniel R. Terno

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Walking a Tightrope in a Curved World

Imagine you are trying to describe how a tiny object moves through space. In the simplest version of physics (General Relativity), if an object has no size and no spin, it just follows the "straightest" possible path through the universe, called a geodesic. Think of this like a marble rolling perfectly down a smooth, curved hill. It doesn't steer; the shape of the hill does the steering for it.

However, real things aren't perfect points. They have size, and they can spin.

Spinning objects: If you throw a spinning basketball, it doesn't just follow the hill; the spin interacts with the curve of the hill, causing it to wobble or drift slightly off the "perfect" path.
Light waves: Even light, which we usually think of as a straight beam, can act like a spinning particle when we look at it very closely. If the light is polarized (like sunglasses filtering light), its path can bend slightly differently depending on how it's spinning.

The paper tackles a specific problem: How do we write the mathematical rules (Lagrangians) that describe these wobbly, spinning paths without getting lost in a mess of confusing math?

The Problem: The "Moving Target" Mess

When physicists try to calculate how these objects move, they have to use a tool called "calculus of variations." This is like asking, "If I nudge the path of the object just a tiny bit, does the total energy change?"

The problem is that space itself is curved. If you nudge a path in curved space, you aren't just moving the object; you are moving it into a completely different "direction" relative to the ground.

The Analogy: Imagine you are walking on a globe. If you take a step forward, you are moving North. If you take a step "forward" but start from a slightly different spot, you might end up moving Northeast. To compare your two paths, you have to "transport" your direction from one spot to the other. If you don't do this carefully, your math breaks, and the results look different depending on how you wrote the equations.

The authors say that previous attempts to write these rules were often messy, inconsistent, or hid the fact that the laws of physics should look the same no matter how you look at them (covariance).

The Solution: The "Parallel Transport" Toolkit

The authors introduce a new, cleaner way to do these calculations. They use a concept called parallel transport.

The Metaphor: Imagine you have a compass needle. You walk from point A to point B on a curved surface. To compare the needle at point A with the needle at point B, you have to slide the needle along the surface while keeping it pointing in the same direction relative to the surface (this is parallel transport).
The Innovation: The authors distinguish between different types of "nudges" (variations):
1. Moving the path: Nudging the object's trajectory.
2. Moving the object's internal spin: Nudging how the object is spinning, without moving its path.
3. Comparing them: Using parallel transport to make sure we are comparing apples to apples, even though they are in different places.

By separating these steps clearly, they create a "covariant" (consistent) framework. This means the math stays clean and the physical laws remain obvious, regardless of the coordinate system used.

What They Did: Three Examples

The paper applies this new toolkit to three specific scenarios:

1. The Simple Marble (Spinless Particles)
They showed that for a simple object with no spin, their method quickly reproduces the standard rule: "Just follow the curve of space." This was a warm-up to prove their math works.

2. The Spinning Top (Massive Spinning Bodies)
They applied their method to heavy, spinning objects (like a spinning black hole or a neutron star).

The Result: They derived the famous MPD equations (Mathisson–Papapetrou–Dixon). These equations describe how a spinning object is pushed off its path by the curvature of space (the "spin-curvature force").
The Benefit: Their method makes it very easy to see why the object spins the way it does and how to find the "conserved quantities" (things that stay the same, like total energy or momentum) without getting bogged down in messy algebra.

3. The Spinning Light Beam (Post-Eikonal Light Rays)
This is the main highlight. They looked at light waves that are so high-frequency they act like particles, but with a twist: the light is polarized (spinning).

The Discovery: They built a simple "recipe" (a Lagrangian) that predicts how this spinning light bends.
The Physics: The light doesn't just follow the curve of space; it gets a tiny "kick" based on its polarization (helicity). If the light spins clockwise, it bends one way; if counter-clockwise, it bends the other. This is known as the gravitational spin Hall effect.
The Breakthrough: They showed that this complex behavior can be described by a surprisingly simple set of rules, provided you treat the light's polarization as a constraint (a rule that must be followed) rather than a free-floating variable.

The Takeaway

The paper doesn't invent new physics; it invents a better way to do the math for existing physics.

Think of it like this: Before, trying to calculate the path of a spinning object in curved space was like trying to solve a puzzle while wearing blindfolds and gloves. You could get the answer, but it was slow, clumsy, and easy to make mistakes.

The authors have taken off the blindfolds and gloves. They provided a clear, step-by-step manual (the covariant variation framework) that lets physicists:

Derive the equations of motion for spinning objects much faster.
Clearly see which parts of the motion come from the shape of space and which come from the object's own spin.
Create simple models for how polarized light travels near massive objects (like black holes), which is crucial for understanding modern astrophysical observations.

In short, they gave physicists a cleaner, more reliable ruler to measure how spin and gravity interact.

Technical Summary: Covariant Variation for Point-Particle Lagrangians

Problem Statement
Structureless test particles in general relativity follow geodesics, but extended bodies with internal structure (such as spin) deviate from geodesic motion. Massive spinning bodies are described by the Mathisson–Papapetrou–Dixon (MPD) equations, derived from a multipole expansion of the stress-energy tensor. Similarly, the leading-order correction to the eikonal approximation for electromagnetic wave propagation can be modeled as the nongeodesic motion of spinning null particles. While effective point-particle Lagrangians can reproduce these equations of motion, their variational formulation presents significant technical challenges. The relevant dynamical variables are defined on curved spacetime and involve constrained internal degrees of freedom (e.g., tetrads or spin tensors). Standard variational techniques often become cumbersome and obscure covariance, leading to inconsistencies in the literature regarding the variational structure of these models.

Methodology
The authors adopt and extend DeWitt's covariant-variation formalism to address these difficulties. The core methodological innovation is the explicit distinction between several types of variations and the construction of their corresponding covariant extensions. The framework distinguishes between:

Spacetime fields ( $\phi(x)$ ), defined on the manifold.
Worldline tensors ( $\psi(x(\tau), \tau)$ ), defined only along the trajectory (e.g., tangent vectors, tetrads).

The authors define four elementary variations:

Same-point variation ( $\delta$ ): The difference between fields at the same coordinate point.
Shift variation ( $\delta_1$ ): The change due to the displacement of the coordinate point.
Total variation ( $\delta_2$ ): The combined effect of displacement and tensor transformation.
Parallel-transport variation ( $\delta_\parallel$ ): The change required to compare tensors at different points via parallel transport.

From these, the authors construct covariant variations ( $\Delta_1$ and $\Delta_2$ ) by comparing elementary variations against parallel transport. Crucially, they introduce an extended shift variation ( $\hat{\delta}_1$ ) and an extended covariant variation ( $\hat{\Delta}$ ). This extension allows for the independent variation of worldline tensors (like the internal frame of a spinning particle) while keeping the worldline trajectory fixed, separating the variation induced by spacetime diffeomorphisms from the variation of internal degrees of freedom.

The formalism is applied to:

Spinless point particles.
Massive spinning particles (deriving MPD equations).
Massless spinning particles (post-eikonal light rays).

Key Contributions and Results

Clarification of Variational Types: The paper systematically categorizes variations, demonstrating that $\Delta_1$ is the natural covariant shift for spacetime fields (keeping the metric inert), while $\Delta_2$ is tied to the total diffeomorphism action and is appropriate for identifying Killing symmetries. The extended variation $\hat{\Delta}$ is shown to be necessary for handling independent worldline tensors.
Derivation of MPD Equations: Applying this framework to massive spinning particles, the authors derive the MPD equations and the associated conservation laws in a manifestly covariant form.
- They model spin using an orthonormal frame $\{e^\mu_A\}$ and construct a Lagrangian dependent on scalar invariants of the tangent vector $l^\mu$ and the frame's angular velocity $\Omega_{\mu\nu}$ .
- By varying the Lagrangian with respect to the worldline displacement and the independent frame rotation, they recover the momentum equation $\nabla_\tau p_\mu = -\frac{1}{2}R_{\mu\nu\rho\sigma}l^\nu S^{\rho\sigma}$ and the spin evolution equation $\nabla_\tau S_{\mu\nu} = p_\mu l_\nu - p_\nu l_\mu$ .
- They derive the conserved charge associated with a Killing vector $\xi$ as $Q_\xi = p_\mu \xi^\mu + \frac{1}{2}S_{\mu\nu}\nabla_\mu \xi^\nu$ .
Lagrangian for Post-Eikonal Light Rays: The authors construct a simple local worldline Lagrangian for massless particles whose trajectories describe light rays in the post-eikonal approximation (including polarization-dependent corrections).
- The Lagrangian consists of a kinetic term $L_{kin} = \omega(k_\mu l^\mu - \frac{e}{2}k_\mu k^\mu)$ and a spin-orbit coupling term $L_{spin} = Ca$ , where $a$ represents the angular velocity of the transverse polarization plane.
- By setting the coefficient $C = -\sigma$ (where $\sigma = \pm 1$ is the helicity) and imposing constraints to enforce the Newman–Penrose tetrad propagation conditions, the Euler–Lagrange equations reproduce the trajectory equation $\nabla_\tau l^\mu = -\frac{\sigma}{\omega} R^\mu_{\nu\alpha\beta} l^\nu e^\alpha_{(2)} e^\beta_{(3)}$ .
- The resulting spin tensor is purely transverse and fixed by the helicity, distinguishing this model from generic MPD systems with independent spin degrees of freedom.

Significance
The paper claims that its primary significance lies in providing a systematic and covariant variational calculus for effective point-particle models in curved spacetime. By distinguishing between variations of spacetime fields, variations induced by spacetime symmetries, and independent variations of internal worldline degrees of freedom, the framework:

Simplifies the derivation of equations of motion and conservation laws for complex systems like spinning bodies and polarized light.
Clarifies the geometric origin of curvature forces and spin transport equations.
Offers a unified language for treating geodesic motion, MPD dynamics, and polarization-corrected light propagation as worldline variational problems.

The authors note that this approach is particularly useful for deriving presymplectic potentials and conserved quantities (including those related to hidden symmetries like Killing–Yano tensors) for effective ray models arising from high-frequency expansions of field equations. They suggest this formalism can be extended to higher orders in the $1/\omega$ expansion and to other fields, such as gravitational and Dirac wave packets, to further investigate the universality of spin-curvature coupling effects.