Operational Observables in Quantum Gravity

In a diffeomorphism-invariant theory, a field evaluated “at the coordinate point $x$” is not by itself an observable: a gauge transformation moves both the coordinate label and the fields. At the 2026-08-20 evidence cutoff, asymptotic charges, perturbatively dressed operators, and relational observables provide controlled answers in important domains. Exact nonperturbative localization and subsystem factorization remain open, and operational measurability is a separate test from formal gauge invariance.

Helpful background. Gauge Symmetry, Relational Observables, and Locality develops gravitational constraints, dressings, and relational clocks.

Gauge transformations, charges, and observables

Canonical gravity is constrained. A smearing of the Hamiltonian and momentum constraints with gauge parameters that vanish suitably at the boundary generates redundancy. If the parameter approaches a nonzero asymptotic symmetry, differentiability of the generator requires a surface term, and that term is a physical charge. Regge and Teitelboim establish this boundary distinction in the Hamiltonian formulation Regge and Teitelboim 1974, §§2–4.

Assessment at 2026-08-20	Meaning
Question status	Partially resolved
Evidence status	Supported overall: established perturbatively and asymptotically; speculative nonperturbatively
Confidence	High
Domain	Fixed boundary conditions, dressing/reference choice, perturbative order, and measurement protocol
What could change the conclusion	An exact relational algebra or factorization theorem

An observable must commute with pure-gauge constraints, but it can transform under a physical asymptotic symmetry and carry its charge. This prevents a common mistake: “diffeomorphism invariant” does not mean “invariant under every transformation called a diffeomorphism.” Boundary conditions decide which transformations are gauge.

A starting project: follow a charge to infinity

Choose this direction if you can perform linearized diffeomorphisms and distinguish constraints from boundary charges. Read the linked observables background, then Donnelly and Giddings 2016, §III.1, the dressing theorem and its proof. Extract the perturbative order, asymptotic falloffs, allowed gauge transformations, and nonzero Poincaré-charge hypothesis. The theorem concerns a specific asymptotic setting; the later compact-dressing proposals on this page change consequential assumptions.

Reconstruct the proof for one charge-carrying scalar excitation around Minkowski space. Translate its gravitational coupling convention, calculate the action of the asymptotic charge, and identify the step that forces metric dependence outside a compact region. Check that transformations vanishing at infinity are treated differently from asymptotic symmetries. As a second control, ask which proof step becomes silent for a Poincaré-singlet operator. This is not a construction of a compactly supported physical operator; it is a precise test of the theorem’s hypothesis.

Deliver an annotated derivation and a boundary-term calculation, with the operator, state, falloffs, and order of approximation written explicitly. A computer algebra system can check variations, but this is mainly an analytic project. Access to a supervisor comfortable with canonical gravity and perturbative QFT matters more than computing resources.

A candidate extension is a comparison of two specified dressings or a relational construction in one fixed background, with an operational preparation/readout question. Check existing constructions before claiming novelty. Stop if the comparison changes charges, background symmetries, or gauge falloffs without acknowledging it. If a spacelike commutator is nonzero, separately analyze whether the proposed operations can be localized: the commutator alone is not a faster-than-light communication protocol.

Dressing a local excitation

In weak-field gravity, one can combine a matter field with a functional of the metric that fixes where the insertion sits relative to infinity. Schematically,

$$\Phi(x)=\phi\!\left(x+V_h(x)\right),$$

where $V_h$ transforms so that the coordinate shift of $\phi$ is canceled to the working order. A line dressing and a Coulomb-like dressing are different gauge-invariant completions of the same underlying excitation; their difference is physical radiation or soft-field data.

Donnelly and Giddings construct such observables and show that their gravitational fields extend to infinity and that spacelike commutators need not vanish exactly Donnelly and Giddings 2016, §§II–V. The result is perturbative and dressing dependent. It demonstrates an obstruction to strictly local QFT-like subsystems, not an operational superluminal signalling channel.

The obstruction depends on the charge and the background. Antonini, Chen, Maxfield, and Penington argue that, when the background has no isometries, an arbitrary compactly supported matter operator can instead be accompanied by compactly supported gravitational data that solve the constraints order by order in $G$ Antonini et al. 2025, §§4.1–4.6, especially Eqs. (4.7)–(4.13). The resulting relational operator is perturbatively gauge invariant and leaves the asymptotic charges fixed. It is not a compact operator carrying nonzero ADM energy, and the construction does not establish exact nonperturbative compact support; approximate symmetries and charged sectors require separate treatment.

Results and open questions at 2026-08-20

Evidence	Secure result	Domain	Boundary of the result
Hamiltonian surface charges	Nonvanishing boundary transformations are physical symmetries	Declared falloffs and topology	Different boundaries define different charge algebras
Perturbative gravitational dressing	Explicit gauge-invariant operators with long-range fields	Weak field through a stated order	No exact nonperturbative local algebra follows
Compact relational dressing, Antonini et al. 2025, §§4.1–4.6	Compactly supported matter operators can receive compact gravitational dressings order by order	Background with no isometries and fixed asymptotic charges	Exact nonperturbative support, approximate symmetries, and charged operators remain open
Relational constructions	Fields can be localized relative to dynamical clocks or rods	Models with controlled reference variables	Reference fluctuations and global existence limit precision
Soft-graviton theorem	Universal leading soft factor for asymptotic amplitudes	S-matrix setting	Does not encode all hard-state information
Recent algebra proposals	Approximate splittings and type-II structures in special settings	Semiclassical sectors and specified algebras	Exact global structure remains unsettled

Weinberg’s soft theorem fixes the leading infrared response of an amplitude to an added soft graviton Weinberg 1965, pp. B516–B524. This supports universal asymptotic constraints, but a soft cloud does not by itself record every hard-state detail. Infrared universality should not be inflated into a complete information-recovery mechanism.

Relational localization

A relational observable asks for the value of one dynamical field when other fields take specified values. Clocks, material reference systems, geodesic anchoring, or asymptotic frames can provide the reference. Tambornino reviews explicit canonical constructions and their limited domains Tambornino 2012, §§2–5. The reference choice is part of the observable, much as a coordinate chart is part of a calculation but with dynamical uncertainty included.

In a semiclassical laboratory, relational observables can be sharply localized compared with experimental resolution even though they are not compactly supported exact operators. This is why formal nonlocality and operational nonlocality must be separated. One must compute the commutator and then construct an allowed preparation-and-readout protocol before claiming signalling.

Algebra and subsystem questions

Gauss constraints tie a region to its boundary data. An excitation that changes a total asymptotic charge cannot be both strictly compactly supported and invisible to the corresponding boundary field. Perturbatively compact zero-charge relational dressings on nonsymmetric backgrounds do not evade that statement; they alter local matter and gravitational data together while preserving the asymptotic charges. Recent work also organizes approximate gravitational splitting and candidate operator algebras, including type-II structures in selected semiclassical situations Giddings 2025, §§3–7. These proposals sharpen the question beyond a naive tensor product.

The unresolved issue is exactness: how large a state space admits the algebra, at what perturbative order, with what center or edge data, and under which boundary conditions? Approximate factorization useful for laboratories may coexist with a failure of exact Hilbert-space factorization.

Counterexamples, null results, and failed routes

The direct adverse test asks for a compactly supported gauge-invariant excitation with nonzero total energy in asymptotically flat gravity. The constraint forces a long-range gravitational field, so the requested object does not exist in the same sense as a neutral local operator in nongauge QFT. This rules out one exact localization ideal.

That adverse test is not a universal no-go for every relational observable. The Antonini et al. construction keeps total asymptotic charges fixed and uses symmetry-breaking background features as relational anchors, thereby producing compact support order by order in perturbation theory. Its scope boundary is equally important: a background isometry obstructs the construction for operators carrying the associated charge, and no exact nonperturbative compact-support result follows.

The converse route also fails. A nonzero spacelike commutator between two chosen dressings does not automatically yield a controllable signal, because the shared constraint field, measurement disturbance, and allowed apparatus matter. Coordinate-invariant curvature scalars can also fail as global labels when several points share the same values.

What would change the assessment

An exact nonperturbative relational algebra with a specified physical Hilbert space, regulator-independent commutators, and an operational measurement protocol would move the question toward resolution. A theorem establishing or excluding factorization for a precise class of asymptotics and states would do the same.

The perturbative verdict would change if higher-order constraints obstructed the known dressings or if two dressings claimed equivalent produced inequivalent measurable predictions after radiation data were matched. Such a test must hold boundary conditions and state space fixed.

Research checklist

• Declare asymptotics, falloffs, topology, and which transformations are gauge.
• Record the background isometries and whether the operator changes an asymptotic charge.
• Specify the dressing or relational reference and its perturbative order.
• Solve the constraints and verify gauge invariance explicitly.
• Compute dressing dependence and spacelike commutators in the same state.
• Distinguish an algebraic commutator from an operational signalling protocol.
• State the state space, center or edge data, and accuracy of any subsystem split.
• Separate universal soft information from hard-state reconstruction.

Source and data map

Canonical surface terms fix the gauge/charge distinction; perturbative asymptotic dressings establish the long-range obstruction for charged excitations; the 2025 compact-dressing proposal supplies the no-isometry, fixed-charge counterbranch; relational constructions supply dynamical localization; the soft theorem fixes asymptotic universality; and the Giddings synthesis maps the current algebraic frontier. No algebraic computation was rerun for this assessment. Reproduction requires fixed boundary conditions, a declared background and its isometries, constraint solutions to a declared order, gauge-variation checks, asymptotic-charge checks, commutators, and an explicit preparation/readout protocol.

Return to the seminar

Record this page’s worksheet, return to Case 6 in the Research Seminar, then continue to Select exactly one technical reproduction. Read those seminar instructions before choosing the reproduction candidate.

References

• Antonini, S., C.-H. Chen, H. Maxfield, and G. Penington. “An Apologia for Islands.” arXiv:2506.04311v1 (2025). arXiv.
• Donnelly, W., and S. B. Giddings. “Observables, Gravitational Dressing, and Obstructions to Locality and Subsystems.” Physical Review D 94, 104038 (2016). doi:10.1103/PhysRevD.94.104038.
• Giddings, S. B. “Quantum Gravity Observables: Observation, Algebras, and Mathematical Structure.” arXiv:2505.22708v2 (2025). arXiv.
• Regge, T., and C. Teitelboim. “Role of Surface Integrals in the Hamiltonian Formulation of General Relativity.” Annals of Physics 88, 286–318 (1974). doi:10.1016/0003-4916(74)90404-7.
• Tambornino, J. “Relational Observables in Gravity: A Review.” SIGMA 8, 017 (2012). doi:10.3842/SIGMA.2012.017.
• Weinberg, S. “Infrared Photons and Gravitons.” Physical Review 140, B516–B524 (1965). doi:10.1103/PhysRev.140.B516.