On Time-Evolution in Quantum Gravity

Technical Summary: On Time-Evolution in Quantum Gravity

Problem Statement
The paper addresses a foundational challenge in the quantum theory of gravity: the definition of time evolution. In the canonical quantization of General Relativity (GR), the classical Hamiltonian vanishes (modulo boundary terms) upon satisfying the constraint equations. This "frozen formalism" suggests that physical states do not evolve in time, creating a tension with the observed time-dependence of classical gravitational backgrounds and the standard path-integral formulation of scattering amplitudes. While previous works (e.g., [1, 2]) highlighted this discrepancy, a unified framework that retains a well-defined notion of time as the Hamiltonian flow while respecting the gauge constraints of GR has remained elusive. Specifically, there is a prevailing "lore" that the BRST-exactness of the bulk Hamiltonian implies a trivialization of time evolution, rendering the recovery of classical dynamics and non-trivial scattering amplitudes problematic.

Methodology
The authors work within a BRST-invariant quantization of GR treated as a low-energy effective field theory. Their approach relies on the gauge-fixed BRST Lagrangian formulation (incorporating Einstein-Hilbert, gauge-fixing, and Faddeev-Popov ghost terms) rather than the Batalin-Fradkin-Vilkovisky (BFV) formalism for ungauge-fixed systems.

Key methodological steps include:

1. Canonical Quantization: The authors identify canonical variables for the gauge-fixed theory, including spatial metric components ($\gamma_{ij}$), temporal metric components parameterized by $A_\mu$, and the associated ghost/anti-ghost fields ($c_\mu, \bar{c}_\mu$) and auxiliary fields ($b_\mu$).
2. Operator Derivation: They promote these fields to operators and establish canonical (anti-)commutation relations.
3. Hamiltonian Reconstruction: The total Hamiltonian is derived by combining the Einstein-Hilbert part and the ghost/gauge-fixing sector. Crucially, the authors utilize the equations of motion for the auxiliary fields to rewrite the Hamiltonian in a concise form.
4. BRST Analysis: They demonstrate that the bulk Hamiltonian can be expressed as the anticommutator of the BRST charge ($\hat{Q}$) and the temporal component of the ghost momentum ($\hat{\Pi}^c_0$).
5. Case Studies: The implications of this structure are analyzed in two scenarios:
   • No-Boundary (Compact Space): Where the total Hamiltonian is purely the bulk term.
   • Asymptotic Minkowski (With Boundary): Where the total Hamiltonian includes the necessary ADM boundary term.

Key Contributions and Results

• Explicit BRST-Exact Identity: The primary result is the derivation of an explicit operator identity for the bulk Hamiltonian:
  $$\hat{H}_{\text{bulk}} = M_{pl} \int d^3x \{ \hat{Q}, i\hat{\Pi}^c_0 \}$$
  This confirms that the bulk Hamiltonian is BRST-exact, a structural feature expected for systems with first-class constraints, but here derived explicitly within the gauge-fixed Lagrangian framework.

• Time-Evolution as Reparameterization: The authors show that the Hamiltonian flow acts as a time-reparameterization on the correlation functions of physical degrees of freedom. Using the Jacobi identity and the properties of the BRST transformation, they prove that for any operator $\hat{O}$ commuting with $\hat{\Pi}^c_0$:
  $$\partial_t \hat{O} = i[\hat{H}, \hat{O}] = -i M_{pl} \int d^3z \{ [\hat{Q}, \hat{O}], i\hat{\Pi}^c_0(z) \}$$
  This demonstrates that the time evolution of operators is generated by the BRST variation projected onto the time direction.

• Resolution of the "Triviality" Paradox: The paper refutes the claim that BRST-exactness implies trivial time evolution.

  • Physical States: While the BRST cohomology (the space of physical states) is annihilated by the bulk Hamiltonian (up to zero-norm states), the representatives of these cohomology classes are not time-independent.
  • Correlation Functions: The time evolution of correlation functions for gauge-variant operators (such as the metric or curvature) is non-trivial. The "image" of the state under time evolution (which is BRST-exact) contributes non-vanishingly to the expectation values of these operators, thereby recovering the expected semi-classical time dependence.

• Role of Boundary Terms in Scattering: In asymptotically Minkowski spacetimes, the total Hamiltonian includes a boundary term corresponding to the ADM energy ($\hat{E}_{ADM}$).

  • The matrix elements of the time-evolution operator between physical states are governed by this boundary term.
  • While the bulk Hamiltonian generates time-reparameterization (which is trivial on cohomology), the boundary term ensures that physical states (specifically those with non-zero energy) evolve in time.
  • This mechanism allows for the definition of asymptotic particle states and a non-trivial S-matrix, resolving the apparent conflict between BRST-exactness and scattering amplitudes.

Significance and Claims
The paper claims to provide a consistent unification of the path-integral description of gravitational scattering and the Hamiltonian description of time evolution within a single BRST-invariant framework.

1. Clarification of Time: It establishes that the "absence of time" in quantum gravity is a misinterpretation of the BRST-exact bulk Hamiltonian. Time evolution is preserved as a reparameterization of the physical degrees of freedom, consistent with classical GR dynamics.
2. Reconciliation of Formalisms: The work bridges the gap between the "over-constrained" canonical quantization (where physical states seem static) and the path-integral approach (which yields dynamic scattering). It shows that the bulk Hamiltonian's BRST-exactness is compatible with non-trivial real-time evolution of bulk correlators.
3. S-Matrix Consistency: The authors argue that the S-matrix emerges naturally. While strictly BRST-invariant asymptotic states (infrared-dressed) would have trivial bulk contributions, standard perturbative calculations (tree-level) utilize states that are not fully BRST-invariant, allowing the bulk Hamiltonian to contribute to scattering amplitudes. The framework thus restores the usual S-matrix and classical equations of GR by consistency.
4. DeWitt's Puzzle: The paper suggests it answers the puzzle posed by DeWitt regarding the mismatch between canonical and path-integral quantizations by identifying a class of BRST-invariant states (analogues to DeWitt's states) and demonstrating how time-dependence arises for physical observables despite the vanishing of the Hamiltonian on the cohomology.

In summary, the authors do not claim to have solved the non-perturbative definition of quantum gravity but have provided an explicit operator realization showing that the BRST-exact nature of the bulk Hamiltonian does not preclude the existence of time evolution or scattering amplitudes, provided boundary contributions and the distinction between cohomology classes and their representatives are properly accounted for.