Canonical Vielbeins for General Relativity: D + 1 Decomposition and Constraint Analysis

This paper presents a self-contained derivation of the Hamiltonian formulation of General Relativity in vielbein variables across $D+1$ dimensions, establishing the constraint algebra, relating it to the metric formulation, and constructing the boost generator to recover full local Lorentz symmetry within an $\mathrm{SO}(D)$-covariant framework.

The Gist

Imagine the universe as a giant, flexible fabric. For a long time, physicists have described this fabric using a single, smooth map called a metric. This map tells you the distance between any two points. However, sometimes, especially when dealing with tiny particles like electrons (spinors), this smooth map is too rigid. Physicists prefer to describe the fabric using a set of local "rulers" and "compasses" placed at every point. These are called vielbeins (or frame fields). Think of them not as a single map, but as a grid of tiny, movable coordinate systems that can rotate and tilt independently at every spot in space.

This paper is a detailed instruction manual on how to take the laws of gravity (General Relativity) and rewrite them entirely in terms of these local rulers and compasses, specifically breaking the universe down into space and time (a "D+1" split).

Here is a breakdown of what the authors did, using simple analogies:

1. The Setup: Cutting the Cake

To study how gravity evolves over time, you have to slice the 4D spacetime cake into 3D layers (like slicing a loaf of bread).

• The Metric Approach: Traditionally, physicists slice the cake and measure the shape of each slice.
• The Vielbein Approach: The authors slice the cake but also track the orientation of the local rulers on each slice. They show how to translate the "shape" of the slice into the language of these rulers.

2. The Two Ways to Slice the Rulers

The authors explore two different ways to organize these local rulers, which is like looking at a spinning top from two different angles:

• Approach A: The "Full Spin" View (Lorentz-Covariant)
  Imagine the rulers can spin and tilt in any direction in 4D space (including time). The authors derive the rules for how these rulers move while keeping the ability to rotate them in any direction intact. They identify "rules of the game" (constraints) that say: "You can't just spin the rulers arbitrarily; their movement is tied to the shape of space."

  • The Result: They found a set of equations that describe the energy and momentum of the universe, ensuring that if you rotate your rulers, the physics stays the same.

• Approach B: The "Flat Floor" View (SO(D)-Covariant)
  Imagine you force the rulers to stand straight up on the floor of each time slice, only allowing them to spin around the vertical axis (like a spinning top that can't lean). This is called the "time gauge."

  • The Problem: By forcing them to stand straight, you lose the ability to describe tilting (boosts) naturally. It's like describing a car only by how it drives forward, ignoring that it can also tilt on a banked turn.
  • The Fix: The authors show that even if you start with this "flat floor" view, you can mathematically reconstruct the "tilting" ability. They built a special "boost generator"—a mathematical tool that acts like a lever to tip the rulers back into a 4D tilt, recovering the full symmetry of the universe.

3. The "Ghost" Rules (Constraints)

In this system, not every part of the ruler is free to move. Some parts are "ghosts"—they don't have their own independent energy but are tied to the others.

• The authors identified these "ghost" rules (primary constraints). They showed that these rules are like the gears in a clock: if one gear (a rotation) moves, the others must move in a specific way to keep the clock working.
• They proved that all these rules fit together perfectly in a "first-class algebra." In plain English, this means the rules are consistent. If you follow one rule, you don't accidentally break another. The system is stable and self-consistent.

4. The "Translation" Problem

One of the paper's key insights is about translation.

• If you try to move the whole universe to the left (a spatial shift), the "flat floor" rulers don't just move; they also have to rotate slightly to stay aligned with the new position.
• The authors showed that the standard "move" button in the math was missing a "rotate" instruction. They fixed this by adding a term that says: "When you move space, also rotate the local rulers." This ensures that the math correctly describes how the universe looks from a moving perspective.

5. The Big Picture

The paper is essentially a rigorous proof that:

1. You can describe gravity using local rulers (vielbeins) just as well as using the smooth map (metric).
2. You can break time and space apart to study how the universe evolves.
3. Even if you start with a simplified view where rulers only spin (not tilt), you can mathematically "unfreeze" them to recover the full, complex ability to tilt and spin in 4D space.
4. All the mathematical rules governing these movements fit together without contradictions.

In summary: The authors took a complex, abstract way of describing gravity (using local frames instead of a global map), sliced it into time and space, and wrote a complete, self-consistent rulebook for how these local frames move, spin, and tilt. They fixed a few missing "instructions" in the math to ensure that moving through space automatically includes the necessary rotations, and they proved that you can recover the full 4D symmetry even if you start with a simplified 3D view.

Technical Summary

Problem Statement
The paper addresses the need for a self-contained, detailed derivation of the Hamiltonian formulation of general relativity using vielbein (frame field) variables in $d = D + 1$ dimensions. While the metric-based Arnowitt-Deser-Misner (ADM) decomposition is well-established, the vielbein formulation is essential for coupling to spinors and offers distinct geometric insights. The authors note that existing literature on the vielbein $D+1$ decomposition (e.g., works by Schwinger, Kibble, Isham, and Deser) is often terse, incomplete, or contains intermediate steps that lack detail or are incorrect. Furthermore, technical nuances regarding constraint algebras and symmetry generators have been largely overlooked. The specific challenge lies in correctly identifying the primary constraints associated with the non-dynamical Lorentz degrees of freedom and constructing generators that act correctly on the full phase space, including the internal Lorentz indices.

Methodology
The authors adopt a straightforward field-theory perspective, avoiding the language of differential forms and frame bundles to maintain pedagogical clarity. They work in $d = D + 1$ dimensions with units $c = \hbar = 1$ and the mostly-plus signature convention. The methodology proceeds in two parallel tracks:

1. Lorentz-Covariant Formulation: Starting from the Einstein-Hilbert action, the authors perform a $D+1$ foliation of spacetime. They decompose the vielbein $e^A_\mu$ into a lapse $N$, shift $N^i$, a spatial vielbein $E^A_i$, and a unit timelike Lorentz vector $X^A$ orthogonal to the spatial hypersurfaces. They derive the conjugate momenta $\pi^i_A$ and identify the primary constraints arising from the local Lorentz invariance. The Hamiltonian is constructed, and the constraint algebra is computed using the Dirac-Bergmann algorithm.
2. SO(D)-Covariant Formulation: The authors introduce a "time gauge" decomposition where the internal Lorentz space is split into a temporal direction and a spatial $SO(D)$ sector. This involves parametrizing the general vielbein via a Lorentz boost acting on a lower-triangular (ADM) vielbein. They derive the phase-space action manifestly invariant under $SO(D)$ rotations. Crucially, they extend this phase space by reintroducing the boost parameters as canonical variables to recover the full local Lorentz symmetry ($SO(1, D)$).

Throughout the derivation, the authors utilize the Moore-Penrose pseudoinverse to handle the inversion of the velocity-momentum relation, which is singular due to the constraints. They also explicitly construct the Castellani generators to ensure the gauge transformations act correctly on the lapse and shift variables, not just the spatial fields.

Key Contributions and Results

• Derivation of Phase-Space Actions: The paper provides explicit derivations of both the Lorentz-covariant and $SO(D)$-covariant phase-space actions. The authors demonstrate that while these actions are equivalent to the standard metric phase-space action on the constraint surface, they differ in their off-shell structure and their action on the internal Lorentz degrees of freedom.
• Constraint Identification: The primary constraints $J^{AB} = \pi^i_{[A}E^B_{i]} = 0$ (Lorentz-covariant) and $J^{ab} = \pi^i_{[a}E^b_{i]} = 0$ ($SO(D)$-covariant) are identified as the momenta conjugate to the non-dynamical Lorentz/rotational fields. The secondary constraints are identified as the Hamiltonian constraint $R \approx 0$ and the momentum constraint $R_i \approx 0$.
• Constraint Algebra: The authors compute the full first-class constraint algebra. They show that the algebra closes weakly, reproducing the local Lorentz-covariant diffeomorphism algebra. A key result is the explicit construction of the spatial diffeomorphism generator $R_i$. The authors show that the naive generator $\tilde{R}_i$ derived directly from the action fails to generate pure diffeomorphisms on the vielbein; it must be modified by a term proportional to the Lorentz generator, $R_i = \tilde{R}_i + {}^{(D)}\omega_{iAB}J^{AB}$, to correctly transform the internal indices.
• Boost Generator Construction: In the $SO(D)$ formulation, the authors explicitly construct the boost generator $K[\vec{\zeta}]$ by extending the phase space. They demonstrate that recovering full $SO(1, D)$ symmetry requires the boost generator to include a term dependent on the Thomas-Wigner rotation, ensuring the correct transformation of the spatial vielbein under boosts.
• Algebraic Closure: The paper verifies that the algebra of the rotation and boost generators closes weakly, with the commutator of two boosts yielding a rotation only on the primary constraint surface.

Significance and Claims
The authors claim that their work fills gaps in the existing literature by providing a detailed, step-by-step derivation that clarifies common sources of confusion regarding the vielbein Hamiltonian formulation. Specifically, they highlight:

• The necessity of modifying the spatial diffeomorphism generator to include Lorentz rotation terms to act correctly on the full vielbein phase space.
• The explicit relationship between the metric and vielbein formulations, clarifying why different constraint representatives are necessary for the correct transformation of internal degrees of freedom.
• The recovery of full local Lorentz symmetry in the $SO(D)$ formulation through the construction of the boost generator, a step often omitted or treated incompletely in other works.

The paper concludes that the derived first-class algebra corresponds to the expected $d(d-3)$ dimensions of the physical phase space for a massless spin-2 field, confirming the consistency of the vielbein formulation with general relativity. The work is presented as a pedagogical and technical reference intended to bring clarity to the formalism without introducing new physical applications or experimental proposals.