Introduction and scope

Quantum mechanics and general relativity are mutually incompatible in the regime where both matter, yet each is among the best-tested theories in physics. The Quantum-Geometric Correspondence (QGC) takes a conservative stance on this incompatibility: it neither quantises the metric nor postulates a discrete substrate, but treats the two theories as the matter-superposition reading and the geometry-entanglement reading of one underlying quantum state , related by a change of observational framework rather than by a dynamical approximation.

This review states what the programme has established, what it has not, and what would decide it. Its central energy scale is on firm ground; its headline mechanism is a well-posed conjecture reduced to a single open construction; its sharpest prediction is falsifiable but not yet tested. On the programme’s own honest accounting this places it at a B+. Throughout, claims are graded as established (rigorous proof or exact numerical identity), derived (following from the stated axioms under explicitly named assumptions), physically motivated (argued and consistent with bounds, but not derived), or conjectured (supported in special cases, not established in general).

The two faces of the correspondence

Three primitive axioms (Section 2)—generalized-entropy conservation, an entropic action principle whose variation returns the Einstein equations and thermal equilibrium together, and a smooth Planck-scale interpolation between the quantum and geometric descriptions of an observable—recover the Bekenstein–Hawking entropy, the holographic bound, and the Unruh temperature, and yield one new structural claim: placing matter in a spatial superposition and entangling spacetime geometry are dual descriptions of one process. The direct physical consequence is gravitational decoherence—a spatial superposition of mass loses coherence because the dual geometric description is an entangling channel.

The central quantitative claim, and its honest status

For a mass \(M\) in a superposition of two locations separated by \(d\), the programme’s decoherence time is \[\begin{equation} \tdec \;=\; \frac{\hbar d}{G M^2} \;=\; \frac{\hbar}{\EG}, \qquad \EG = \frac{G M^2}{d}, \label{eq:tau-dec} \end{equation}\] the Diósi–Penrose form , with \(\tdec(1\,\mu\mathrm{g}, 1\,\mathrm{mm}) \approx 1.6\) ns. The energy scale \(\EG = G M^2/d\) is established: it is the classical gravitational self-energy difference between the branches, and nothing in the programme is needed to obtain it. We conjecture that the decoherence rate equals this energy scale divided by \(\hbar\), \[\begin{equation} \Gdec \;=\; C\,\frac{\EG}{\hbar}, \qquad C\in [\,2/\pi,\, 1\,] \approx [0.637, 1.000], \label{eq:rate} \end{equation}\] the coefficient \(C\) bounded but not uniquely pinned (Section 3). This is the programme’s load-bearing prediction: perturbative quantum field theory gives a rate suppressed by an additional power of \(G\)—a decoherence time of order \(10^{18}\) years for the same parameters, some \(35\) orders of magnitude longer. The two predictions are operationally distinct—one says a microgram superposition decoheres in nanoseconds, the other says it never decoheres in any laboratory—and an experiment in the right regime (Section 4) decides between them.

The structural development underlying the rate

The operator identity underlying \(\eqref{eq:rate}\)—that the region’s modular Hamiltonian equals \(2\pi\) times the constrained physical Hamiltonian, \(K= 2\pi H_{\mathrm{phys}}\)—cannot be established in flat space, where the relevant local algebra is of Type III\(_1\) and admits no trace. Relocating the construction to a finite causal diamond gives an algebra of Type II, to which the von Neumann machinery of Chandrasekaran–Longo–Penington–Witten  applies. A sequence of results (Section 3) assembles that finite-diamond construction into a single object whose one remaining free parameter is the coefficient \(C\) of \(\eqref{eq:rate}\). Whether that construction can be closed at full rigour is the question on which both the grade and the operator identity turn.

Plan of the review

Section 2 states the three axioms and the theorems they generate. Section 3 develops the gravitational-decoherence core: the mechanism, the rate, the \(G^1\)-versus-\(G^2\) distinction, and the finite-diamond (“R1”) construction. Section 4 collects the testable predictions and the experimental falsifiability map. Section 5 turns to the cosmological sector—the MOND acceleration scale, holographic dark energy, and cosmic decoherence—which shares the programme’s de Sitter thermodynamics. Section 6 positions QGC against spontaneous-collapse models, the Diósi–Penrose proposal, and stochastic gravity. Section 7 states the open problems and a roadmap, and Section 8 concludes. The underlying computations are carried in the cited papers.

Axiomatic foundation

The canonical formulation (v2.0) rests on three primitive axioms; the broader structure that earlier formulations carried as six axioms plus a duality correspondence is recovered here as derived theorems.

The three primitive axioms

Derived theorems

Axioms I–III recover the following standard results, each established or derived as marked, with dimensional consistency and limiting behaviour checked.

• Bekenstein–Hawking entropy \(S_{\mathrm{BH}} = A/(4\lP^2)\)  and the holographic bound: established from Axiom I.

• Unruh temperature \(T_U = \hbar a/(2\pi c \kB)\) : established from Axiom I via the horizon thermodynamics.

• Duality correspondence: a matter superposition and the corresponding geometric entanglement are dual descriptions, derived from Axioms I and II. This is the structural claim that drives gravitational decoherence (Section 3).

• Dark energy \(\rhoDE = \alpha\, c^2 H^2/G\) with \(\alpha \approx 0.082\), derived from Axioms I and II  (Section 5).

• Observer-dependent horizons: physically motivated from Axiom I—a dimensional-analysis relation between acceleration and horizon scale, not a derivation (the boxed coefficient absorbs an \(\mathcal{O}(1)\) factor).

The status of the saturation input

One further ingredient—the Gravitational Information Saturation Principle, which states that the information a region can exchange with the gravitational field saturates the holographic bound—supplies the \(G^1\) scaling of the decoherence rate (Section 3). It is carried throughout as a physically motivated hypothesis, not elevated to an axiom. The several heuristic arguments once advanced for \(G^1\) scaling reduce, in four-dimensional linearized gravity, to a single physical input—the Wheeler–DeWitt Hamiltonian constraint—rather than to several independent derivations: the programme has one physical reason for \(G^1\), expressed in several formal ways.

What the axioms do and do not fix

The axioms fix the kinematics (what the dual descriptions are) and the equilibrium dynamics (the Einstein equations and the Gibbs state). They do not, by themselves, fix the operator-level identity between a region’s modular Hamiltonian and its constrained physical Hamiltonian that the decoherence rate requires; that identity is an additional claim, treated in the next section. The axiomatic layer is internally consistent and recovers known physics, while the one genuinely new dynamical prediction—the \(G^1\) rate—rests on an input that is motivated but not yet derived from the axioms alone.

Gravitational decoherence: mechanism, rate, and the finite-diamond construction

Gravitational decoherence is the core of QGC and the source of its sharpest prediction. This section separates what is established from what is conjectured.

The energy scale

For a mass \(M\) in a superposition of two configurations differing by a separation \(d\), the gravitational self-energy difference between the branches is \[\begin{equation} \EG \;=\; \frac{G M^2}{d} \;=\; 6.7\times10^{-26}\ \mathrm{J}\quad(1\,\mu\mathrm{g},\,1\,\mathrm{mm}). \label{eq:EG} \end{equation}\] For an extended source the point-mass form is replaced by the exact Diósi–Penrose self-energy integral, which saturates at \(1.2\,GM^2/R\) for separations beyond the object diameter. The energy scale \(\eqref{eq:EG}\) is established: it is a statement of classical gravity, independent of any QGC-specific machinery, and it sets the only scale that can appear in a gravitationally induced rate.

The two scalings, \(G^1\) versus \(G^2\)

The physical question is whether the decoherence rate is \(\Gdec \sim \EG/\hbar \propto G^1\) or the perturbative-QFT result \(\propto G^2\). The distinction has a clean interpretation, confirmed by explicit computation:

• the \(G^2\) rate is the rate of entanglement generation between matter and the graviton field, computed perturbatively from an unphysical (un-constrained) initial state;

• the \(G^1\) rate is the rate of entanglement manifestation in the constrained theory, where the Wheeler–DeWitt Hamiltonian constraint ties the gravitational field to its source.

Both are correct answers to different questions. For \((1\,\mu\mathrm{g}, 1\,\mathrm{mm})\) they differ by a factor \(\sim 3\times10^{34}\): \(\tdec \approx 1.6\) ns (\(G^1\)) versus \(\tau_{G^2} \sim 10^{18}\) yr. The \(G^1\) scaling is derived in linearized gravity by a coherent-state argument; the full spin-2 computation confirms the scalar-proxy result , and soft- graviton infrared logarithms do not promote \(G^2\) to \(G^1\). What remains conjectured is the operator-level identity that converts the energy scale into the rate.

The operator identity and the flat-space obstruction

In the modular-theory formulation , the decoherence rate follows from the identity \[\begin{equation} K\;=\; 2\pi\,H_{\mathrm{phys}}+ \order{G^2}, \label{eq:operator} \end{equation}\] where \(K\) is the modular Hamiltonian of the region containing the superposition and \(H_{\mathrm{phys}}\) is the constrained physical Hamiltonian.

Two obstructions prevent a flat-space proof. First, the local algebra of a region in the vacuum is a Type III\(_1\) von Neumann factor: it has no trace, no density matrices, and no canonical “rate per unit time” normalization. Second, the Bisognano–Wichmann  transport of the vacuum modular Hamiltonian to the matter-loaded, gravitationally dressed state fails because the Newtonian field has global support, lying outside the region’s algebra. Identity \(\eqref{eq:operator}\) therefore remains conjectural in flat space.

The finite-diamond construction (R1)

The route around the obstruction is to give the region a cosmological constant—equivalently, to work in a finite causal diamond—where the algebra becomes Type II and carries a trace, and the crossed-product machinery of Chandrasekaran–Longo–Penington–Witten  and Jensen–Sorce–Speranza  applies. The construction (labelled “R1”) assembles six results into a single object.

1. Type II trace, unconditionally. By Takesaki’s structure theorem , the crossed product of the diamond algebra by its modular automorphism group is Type II\(_\infty\) with a canonical semifinite trace, for any modular flow. Established.

2. Geometric flow for conformal matter. The Casini–Huerta–Myers conformal map  sends the diamond to \(\mathbb{R}\times\mathbb{H}^3\), turning the conformal-Killing modular flow into an exact, constant-norm Killing flow. Derived for the conformal proxy.

3. Geometric flow for the graviton. The physical graviton is not a conformal field, but its gauge-invariant ball algebra is two scalar towers with \(\ell\geq2\) (Benedetti–Casini ), each inheriting the local weight \(f(r)= (R^2-r^2)/(2R)\) (Huerta–van der Velde ); the flat-space degeneration of the Regge–Wheeler  and Zerilli  master potentials to the scalar centrifugal potential anchors the reduction. The flow is geometric for the physical field. Derived from published inputs.

4. Intrinsic observer and clock. The crossed-product observer is the diamond-center geodesic; its modular time is \(t(\sigma) = R\tanh(\pi\sigma/2)\), so the diamond tips form a horizon reached only at infinite modular time, while the observer carries a uniform clock at the canonical Kubo–Martin–Schwinger temperature \(\beta_{\mathrm{mod}} = 2\pi\). Established exactly.

5. Residues decoupled from the rate. Three remaining technical residues—the bounded-below observer Hamiltonian (Type II\(_1\) refinement), the entangling-surface edge modes, and the four-dimensional trace anomaly—are all additive, entropy-constant effects. The modular flow that sets the rate is invariant under \(K\to K+ c\,\mathbb{1}\), so none of them can move the coefficient \(C\); they shift only the entropy’s additive constant. The shift-invariance is exact.

6. Infrared safety. The extracted physical rate is independent of the arbitrary diamond size \(R\): the modular weight at the source (\(\propto R\)) cancels the observer clock (\(\propto R\)), giving \(\Gamma_{\mathrm{phys}} \propto \EG/\hbar\) with \(\partial_R \Gamma_{\mathrm{phys}} = 0\). Exact.

One weight \(f(r)\), one Bisognano–Wichmann factor \(2\pi\), one modular flow, and one graviton-to-matter degeneration thread run consistently through all six. The assembled construction has a single remaining free parameter—the clock-model normalization that fixes the coefficient \(C\)—with everything else proven, derived, or scoped.

One formalism for both the scale and the rate

In earlier formulations the energy scale and the rate were computed in different formalisms (linearized gravity for \(\EG\), the full constrained theory for the mechanism) and stitched together. The finite-diamond modular Hamiltonian removes the stitch: \(K= 2\pi H_{\mathrm{phys}}\) is one object with two observables. Its gap between the which-path branches is the energy scale, \[\begin{equation} \frac{\hbar}{2\pi}\,\Delta K\;=\; \EG \qquad (\text{independent of the clock normalization}), \label{eq:scale-from-gap} \end{equation}\] and its flow on the observer clock is the rate, \(\Gdec = C\,\EG/\hbar\). The energy scale read from the gap is established and independent of the open coefficient; only the rate carries it. The Hamiltonian constraint that the linearized formalism could not make manifest is here the defining identity of the single object.

The rate coefficient

What the finite-diamond trace fixes is the modular temperature (\(\beta_{\mathrm{mod}} = 2\pi\)), the variance of the modular Hamiltonian, and the minimum-uncertainty clock width; these remove the normalization freedom that once gave a factor-four window. The residual freedom is a genuine distinction between two operationally different observables built from the same trace-fixed metric: the continuous fringe-visibility (Bures) rate gives \(C= 1\) (the Diósi–Penrose value), while the Margolus–Levitin orthogonalization speed gives the floor \(C= 2/\pi\). An interferometer measures fringe-visibility decay, so the trace-fixed metric selects \(C= 1\) for the falsifiable prediction; but the exactness of that selection still rests on a spectral input that the free-field analysis does not supply (the bath is super-Ohmic, with no zero-frequency weight to whiten). The honest statement is therefore \[\begin{equation} C\in [\,2/\pi,\,1\,], \qquad \text{natural value } C= 1,\ \text{not uniquely pinned}. \label{eq:C-window} \end{equation}\] Pinning \(C\) to a single value, and proving \(\eqref{eq:operator}\) as an operator equation, are the same open problem—the closure of the R1 construction at full rigour—and it is the problem on which the programme’s grade turns.

Predictions and experimental falsifiability

The programme’s value rests on whether it can be tested. The genuinely discriminating prediction is separated here from those that merely accompany it.

The discriminating prediction

The one prediction that distinguishes QGC from standard physics is the \(G^1\) decoherence rate \(\eqref{eq:rate}\): a spatial superposition of a sufficiently massive object decoheres in \(\eqref{eq:tau-dec}\), whereas perturbative quantum field theory predicts effectively no decoherence (\(\tau_{G^2}\sim10^{18}\) yr). The observable is the interference visibility after a hold time \(t\), \[\begin{equation} V(t) \;=\; e^{-t/\tdec}, \qquad \tdec = \hbar/\EG, \label{eq:visibility} \end{equation}\] with the standard prediction \(V \approx 1\) (Figure 2). A measured visibility decay at the rate \(\eqref{eq:visibility}\) would confirm \(G^1\); a null result (sustained coherence well past \(\tdec\)) would falsify it .

The falsifiability map

For a uniform sphere of density \(\rho\) and radius \(R = (3M/4\pi\rho)^{1/3}\) in a superposition of separation \(\Delta x\), the exact self-energy \(\EG(\Delta x)\) saturates at \(1.2\,GM^2/R\) for \(\Delta x \gtrsim 2R\), giving a decoherence time that scales as \(\tdec \propto M^{-5/3}\) in the saturated regime. This produces a sharp threshold: for silica, \(\tdec = 1\) s requires \[\begin{equation} M_\star \;\approx\; 7.6\times10^{-16}\ \mathrm{kg}\quad(R\approx0.44\,\mu\mathrm{m}). \label{eq:threshold} \end{equation}\] Below \(M_\star\) the superposition decoheres too slowly to test on laboratory timescales; above it, the rate turns on rapidly. The decision is therefore a narrow window in mass and separation, not “the larger the better.”

To distinguish \(V(t)\) from the standard null \(V=1\) at the \(5\sigma\) level requires, at the shot-noise floor, \(N_{5\sigma} = 25/(1-V)^2\) interference runs. Figure 1 maps the decoherence time over the \((M,\Delta x)\) plane and locates the platforms, and Table 1 reads off the discriminant for each.

The reading is sharp. The femtogram envelope target—\(1.5\) fg of silica, a \(1\,\mu\)m superposition held for \(1\) s under \(\sim7\times10^{-15}\) mbar cryogenic vacuum—is decided in \(\sim37\) runs. The mass and hold time are within reach of demonstrated levitated-optomechanics capability ; the micrometre superposition is the goal of MAQRO-class proposals . The MAQRO-class mass fully decoheres if the micrometre superposition is achieved. Current levitated spheres are sub-threshold and can only set bounds. Large-separation platforms of the Bose–Marletto–Vedral type carry enough mass but, with separations of hundreds of micrometres, sit in the saturated regime and self-decohere in milliseconds—faster than the entanglement protocol they are designed to run, so for them the \(G^1\) signature is a null entanglement result rather than a measured visibility.

The accompanying predictions, and their honest status

Other consequences are real but less discriminating:

• Bose–Marletto–Vedral visibility suppression : an orientation-dependent, approximately mass-independent law, physically motivated. It currently carries an unresolved inconsistency between two candidate self-decoherence rates (the arm-set rate \(\propto 1/\Delta\) versus the radius-saturated rate), which differ by \(\sim10^2\) at experimental parameters; the prediction is not firm until that rate is resolved.

• \(N\)-particle scaling \(\Gdec \propto M_{\mathrm{total}}^2\): real, but generic to any mass-density-sourced decoherence, so not unique to QGC.

• Short-time Zeno regime: a zero-initial-slope onset \(V(t)\approx 1 - \tfrac12 M_2 t^2\) for \(t < d/c\), distinguishing QGC from a Markovian collapse ansatz but living in a femtosecond window that is not near-term testable.

• A locked structural ratio \(\azero/(c\,\Gtot) = 2/g(1)\) linking the MOND scale to the cosmic decoherence rate (Section 5): unique to QGC as an internal architecture constraint, but untestable, since the cosmic rate is unmeasurably small.

The cosmological sector

The same de Sitter thermodynamics that underlies the decoherence programme has consequences on cosmological scales, developed in dedicated papers. This section summarizes the results and their status and gives the single structural relation that ties the sector to the laboratory one.

The MOND acceleration scale

Treating the entanglement of the de Sitter horizon as an elastic medium yields a covariant field equation \(G_{\mu\nu} + \Lambda g_{\mu\nu} = 8\pi G(T_{\mu\nu}^{\mathrm{matter}} + T_{\mu\nu}^{\mathrm{elastic}})\) with an elastic modulus \(\kappa = c^4/(8\pi G \azero)\) set by a single acceleration scale. That scale is fixed, with no free parameters, by the de Sitter temperature: \[\begin{equation} \azero \;=\; \frac{c \Hz}{2\pi} \;\approx\; 1.08\times10^{-10}\ \mathrm{m/s^2}, \label{eq:a0} \end{equation}\] which reproduces the empirical MOND scale . The recovery of MOND phenomenology from de Sitter thermodynamics is derived; the interpolation function \(\mu(x)\) that governs the transition between the Newtonian and deep-MOND regimes is not uniquely fixed by the framework—only its limiting behaviours (\(\mu\to1\) at high acceleration, \(\mu\to x\) at low) are derived, with the transition region observationally determined. A residual factor-\(\sim2\) tension in galaxy-cluster masses, shared with MOND generally, remains open.

Holographic dark energy

The same horizon thermodynamics gives a dark-energy density \[\begin{equation} \rhoDE \;=\; \alpha\,\frac{c^2 H^2}{G}, \qquad \alpha \approx 0.082 \pm 0.001, \label{eq:rhoDE} \end{equation}\] derived from the holographic framework , with the coefficient \(\alpha\) understood as the product of a horizon-information ceiling and an epoch factor. The acoustic-peak structure of the cosmic microwave background constrains the early-time behaviour: in the saturated (\(w=-1\)) realization the early dark-energy fraction is negligible and the peaks coincide with \(\Lambda\)CDM (a consistency check), while a naive constant-fraction realization is excluded by current data.

Cosmic decoherence

Applying the laboratory decoherence rate to cosmological mass scales naively gives an absurd result—a rate of order \(10^{103}\) Hz—which is a Minkowski-formula artifact. The de Sitter recomputation , together with a quantum-cosmology selection of the relevant mass granularity and an additivity bound, gives instead \[\begin{equation} \Gtot \;=\; g(1)\,\frac{H}{4\pi} \;\approx\; \frac{H}{10\pi} \;\approx\; 7\times10^{-20}\ \mathrm{Hz}, \qquad g(1) = 1 - \tfrac{2}{\pi}\Si(1) \approx 0.398, \label{eq:cosmic-rate} \end{equation}\] which is sub-causal and physically harmless. The engine of the resolution is the exact identity \(\varepsilon_{\dS} = G\hbar H^2/(2\pi c^5) = 1/(2\SdS)\) relating the de Sitter post-Newtonian parameter to the horizon entropy. The result holds as a controlled approximation, with the \(\order{G^2}\) residues disclosed. The one premise the additivity bound once carried—that the decoherent-histories readout occurs at horizon crossing rather than at a fine-tuned sub-horizon time—is now derived from standard de Sitter classicalization : a mode registers its which-path phase only when it freezes and squeezes at horizon crossing, while sub-horizon modes oscillate and average away.

The structural link to the laboratory

The cosmological and laboratory sectors are not independent: they share a single de Sitter frequency, the “horizon clock” . The MOND scale and the cosmic decoherence rate are locked by \[\begin{equation} \frac{\azero}{c\,\Gtot} \;=\; \frac{2}{g(1)}, \label{eq:locked-ratio} \end{equation}\] a relation with no free parameters: the same \(2\pi\) of de Sitter thermality appears in both. This is a genuine internal-consistency constraint—the programme has one cosmic input where a phenomenological model would have two independent knobs—but it is not a falsifiable prediction, because \(\Gtot\) is unmeasurably small. It belongs among the programme’s structural results, not among its tests.

Relation to other approaches

Gravitational decoherence is a crowded field. QGC is positioned here against the three families of proposals it most resembles—where it agrees, where it differs, and what would tell them apart.

Spontaneous-collapse models (GRW, CSL)

Continuous spontaneous localization and its relatives  modify the Schrödinger equation by adding a stochastic, nonlinear localization term with a postulated noise strength and correlation length. They predict decoherence of macroscopic superpositions and are falsifiable, and current experiments already constrain their parameters. QGC differs in kind: it adds no stochastic term and no new fundamental constant. The decoherence is a consequence of the gravitational constraint acting on an otherwise unitary theory, and its rate is fixed by \(G\), \(\hbar\), and the configuration through \(\EG\), with the only freedom the \(\order{1}\) coefficient \(C\) of \(\eqref{eq:C-window}\). Operationally, collapse models predict a noise spectrum (heating, spontaneous radiation) that QGC does not; conversely, QGC’s mass scaling \(\Gdec \propto M^2/d\) is sharper than the collapse models’ free-parameter freedom. The cleanest discriminator from CSL is that QGC’s rate has no adjustable strength: it is the Diósi–Penrose value or it is wrong.

The Diósi–Penrose proposal

QGC is closest to the Diósi–Penrose (DP) gravitational-decoherence proposal , and recovers its rate as the laboratory limit: the decoherence time \(\eqref{eq:tau-dec}\) is the DP form, and the natural coefficient \(C= 1\) is the DP value. The difference is one of foundation rather than of phenomenology. DP posits the rate from a gravitational self-energy uncertainty argument; QGC attempts to derive it from the constrained modular dynamics (Section 3), and in doing so commits to the object-size (continuous-density) cutoff that lets the rate evade the nucleon-scale spontaneous-radiation bound that constrains the original DP prescription. The two are therefore experimentally near-degenerate in the laboratory regime—both predict the visibility decay of Table 1—but QGC carries a falsifiable structural commitment (the cutoff, the \(G^1\) scaling from the constraint) that DP, as a phenomenological rate, does not.

Stochastic and semiclassical gravity

Stochastic- and semiclassical-gravity approaches  treat the metric as a classical field driven by the quantum stress-energy fluctuations, yielding decoherence from the induced metric noise. This is a \(G^2\) effect—the rate is quadratic in the coupling—and so coincides, at the level of scaling, with perturbative quantum field theory rather than with QGC’s \(G^1\) claim. The disagreement is exactly the one the experiments of Section 4 are designed to resolve: stochastic gravity and perturbative QFT predict no laboratory decoherence of a microgram superposition, while QGC predicts nanosecond-scale decoherence. This is the sharpest available test of whether gravity must be quantized: QGC’s \(G^1\) rate is a signature of a genuinely quantum (constraint-bearing) gravitational sector, whereas the \(G^2\) rate is what a classical-channel or perturbative-graviton treatment gives.

Programmes of quantum gravity

QGC is not a candidate fundamental theory of quantum gravity in the sense of loop quantum gravity, string theory, or asymptotic safety; it is a correspondence principle with a near-term prediction. This distinction sets what QGC is and is not accountable for.

Loop quantum gravity and related canonical programmes quantize the gravitational field directly, building a Hilbert space of spin-network states with discrete geometric spectra. QGC takes the opposite stance: it does not quantize the metric, treating geometry instead as an emergent, entropic description dual to the matter sector (Axioms I–II). Where loop quantum gravity predicts Planck-scale discreteness of areas and volumes, QGC’s Axiom III supplies only a smooth interpolation with a minimum length, and makes no claim about a fundamental spacetime microstructure. The programmes are not competitors so much as answers to different questions—“what are the quantum degrees of freedom of geometry?” versus “what does treating geometry as emergent predict for a laboratory superposition?”—and QGC’s content is the latter.

String theory embeds gravity in a unitary quantum theory with a definite ultraviolet completion, and its holographic incarnation (AdS/CFT) is the deepest existing realization of the geometry-as-entanglement idea that QGC takes as a working principle. QGC borrows the modular and crossed-product technology that AdS/CFT and the de Sitter algebra programme have developed (Section 3), and is consistent with the holographic dictionary, but it does not require a string-scale completion or extra dimensions: its predictions live at the linearized, low-energy level where the gravitational constraint, not the ultraviolet completion, is what matters. Given AdS/CFT, QGC is an attempt to extract a flat-space, laboratory-scale consequence of the same entanglement–geometry correspondence.

Asymptotic safety posits a non-trivial ultraviolet fixed point that renders quantized gravity predictive, and shares with QGC the use of thermodynamic and renormalization-group arguments. It differs in retaining a quantized metric with a running gravitational coupling; QGC keeps \(G\) fixed and locates the new physics in the constraint structure rather than in the flow. Of the three, asymptotic safety is the one whose low-energy phenomenology could in principle overlap QGC’s, but the two make their sharpest statements in different regimes (the ultraviolet fixed point versus the laboratory decoherence rate), and no direct conflict has been identified.

The honest position is that QGC neither subsumes nor is subsumed by these programmes. It is a correspondence that, if its \(G^1\) rate is confirmed, would constrain any fundamental theory to reproduce that rate; if falsified, it would leave the fundamental programmes untouched, since none of them predicts \(G^1\) decoherence. Its value is the prediction, not a claim to fundamentality.

Summary of the discriminators

Open problems and roadmap

The problems that remain are ordered by how much their resolution would move the programme.

The grade-moving problem: closing R1

The single problem whose resolution would change the programme’s standing is the closure of the finite-diamond construction at full rigour—equivalently, proving the operator identity \(\eqref{eq:operator}\) and pinning the coefficient \(\eqref{eq:C-window}\) to a single value. By Section 3, the construction reduces to one well-posed question: the clock-model normalization that fixes \(C\). The trace fixes the modular temperature and the metric; what remains is to show that the fringe-visibility (Bures) diffusion coefficient is exactly \(\EG/\hbar\) rather than an \(\order{1}\) multiple, which requires controlling the spectral shape of the constraint-borne noise. A proof here would simultaneously legitimize the operator identity, the rate extraction, and the \(C= 1\) selection, and would raise the grade. A demonstration that the construction is ill-posed, or an experiment showing \(G^2\), would lower it. This is the fulcrum of the whole programme.

Theory problems of secondary impact

• The Bose–Marletto–Vedral rate inconsistency remains to be resolved before that prediction is firm: the arm-set and radius-saturated self-decoherence rates disagree by \(\sim10^2\) at experimental parameters, and the visibility law is contingent on the choice.

• Rigorous \(\order{G^2}\) remainder bounds for the coherent-state argument exist only at the level of power counting; a genuine operator bound is open.

• The MOND interpolation function \(\mu(x)\) is not derivable from the current entropy-equilibrium argument; only its limits are fixed. The galaxy-cluster factor-\(\sim2\) tension is the associated phenomenological gap.

• The edge-mode and trace-anomaly bookkeeping of the finite-diamond construction, while shown not to affect the rate coefficient, remains to be computed for a complete entropy ledger.

The experimental roadmap

The programme’s fate is ultimately experimental. The decisive measurement is a femtogram-to-picogram solid-object superposition, held for of order a second, in cryogenic extreme-high vacuum, with environmental decoherence budgeted below the \(G^1\) rate (Section 4). This is not a far-future programme in principle—the required mass and hold time are within an order of magnitude of demonstrated levitated-optomechanics capabilities—but the central challenge, a micrometre-scale spatial superposition of a sub-micrometre sphere, is at the frontier of the field and is the goal of MAQRO-class proposals on a roughly \(2028\)–\(2035\) horizon. A positive result would establish \(G^1\) and with it the quantum character of the gravitational sector; a clean null below the predicted timescale would falsify the programme’s central claim. Either outcome is informative, which is the property a research programme should want of its predictions. In parallel, wide-binary statistics (Gaia DR4 and beyond) and high-redshift rotation curves test the cosmological MOND scale \(\eqref{eq:a0}\) and its possible redshift dependence \(\azero(z) = c H(z)/(2\pi)\).

What would falsify the programme

The programme is falsified by: (i) an experiment in the decidable window (Table 1) that maintains coherence well past \(\tdec\), contradicting \(G^1\); (ii) a demonstration that the R1 construction is ill-posed, removing the basis for the operator identity; or (iii) a cosmological measurement of \(\azero(z)\) inconsistent with \(cH(z)/(2\pi)\). None of these is currently realized, and the first is within experimental reach this decade.

Conclusion

The Quantum-Geometric Correspondence treats quantum mechanics and general relativity as complementary projections of a single quantum-informational reality, and asks what that stance predicts. The answer divides cleanly. The kinematic and equilibrium layer—three primitive axioms recovering the Bekenstein–Hawking entropy, the holographic bound, the Unruh temperature, the Einstein equations, and a thermodynamic dark-energy density—is on standard ground. The single new dynamical prediction, gravitational decoherence at the \(G^1\) rate \(\Gdec = C \EG/\hbar\), separates into an established energy scale \(\EG = GM^2/d\) and a conjectured operator identity \(K= 2\pi H_{\mathrm{phys}}\) that converts the scale into a rate. The latter has been reduced, through the finite-diamond construction, to a single well-posed open question—the clock-model normalization that fixes the coefficient \(C\in[2/\pi,1]\)—with the energy scale, the geometric flow (now including the physical graviton), the intrinsic observer clock, the decoupling of the entropy residues, and the infrared safety all established along the way.

The programme’s grade, B+, reflects this honestly: the physics is intact and the falsifiable claim is unchanged, while the central identity stands as a strong conjecture rather than a theorem. The decisive virtue of the prediction is that it is testable: a femtogram superposition held for a second in cryogenic vacuum distinguishes the \(G^1\) rate from the \(G^2\) rate of perturbative quantum field theory and stochastic gravity by some thirty-five orders of magnitude, in a few tens of interferometric runs, with an experiment at the frontier of—but not beyond—current levitated-optomechanics capability. Whether spacetime must be quantized, and whether it decoheres superpositions on the Diósi–Penrose timescale, are questions this programme renders sharp and an experiment this decade can answer. That is the outlook: a conservative correspondence with one open construction and one falsifiable number, waiting on a measurement.

Appendices

The rate coefficient: derivation and the residual freedom

This appendix recapitulates how the coefficient \(C\) in \(\Gdec = C\,\EG/\hbar\) is bounded, why it is not yet uniquely pinned, and what the open question is, distinguishing throughout a bound from a derivation.

The free field saturates: an exponent, not a rate

The free-field decoherence exponent for the two-branch coherent-state overlap is \[\begin{equation} \Gamma(t) \;=\; \int\!\frac{d^3k}{(2\pi)^3}\,|\delta\alpha(k)|^2\,2\bigl(1 - \cos\omega_k t\bigr), \qquad \omega_k = c|k|, \label{eq:free-exponent} \end{equation}\] with the equilibrium amplitude \(\delta\alpha(k) \propto (GM/k^2)\) set by the Newtonian fields of the two branches. By the Riemann–Lebesgue lemma the oscillatory term averages away, and the exponent saturates at a static value \(\Gamma_{\mathrm{sat}} = \xi\ln(d/\lP)\) with \(\xi = GM^2/(\hbar c)\). This is a dimensionless coherence-loss exponent—a one-time overlap reduction \(|\langle\Phi_L|\Phi_R\rangle|^2 = e^{-\Gamma_{\mathrm{sat}}}\)—not a rate. The free graviton bath is super-Ohmic, with spectral density \(J(\omega)\propto\omega^3\) and \(J(0^+) = 0\), so it carries no zero-frequency weight and produces no sustained (Markovian) decoherence on its own. The energy scale is present in \(\eqref{eq:free-exponent}\); the rate is not.

The constraint supplies the rate

The linear-in-time rate is not a free-field effect: it is enforced by the Wheeler–DeWitt Hamiltonian constraint, which ties the gravitational field to its source and prevents the branches from sharing a single un-sourced field configuration. In the constrained influence functional the static overlap is converted into a sustained dephasing, and the natural Markovian rate is \[\begin{equation} \frac{d\rho_{\mathrm{matter}}}{dt} \;=\; -\frac{i}{\hbar}[H_{\mathrm{eff}},\rho] - \frac{\Gamma}{2}\,[X,[X,\rho]], \qquad \Gamma = \frac{\EG}{\hbar}, \label{eq:lindblad} \end{equation}\] i.e. \(C= 1\), “up to an \(\mathcal{O}(1)\) coefficient that depends on the detailed clock model.” That residual \(\mathcal{O}(1)\) is the whole of the open question.

What the trace fixes, and the two-observable residue

In the finite-diamond (Type II) formulation the canonical trace removes the normalization freedom that earlier gave a factor-four window: it fixes the modular temperature \(\beta_{\mathrm{mod}} = 2\pi\), the Kubo–Martin–Schwinger variance of the modular Hamiltonian, and hence the Bures–Fisher metric on the two-branch state space. What survives is not a normalization ambiguity but a genuine distinction between two operationally different observables built from the same trace-fixed metric: \[\begin{align} \text{continuous fringe-visibility (Bures) rate:}\quad & \Gamma = \EG/\hbar \;\Rightarrow\; C= 1, \\ \text{Margolus--Levitin orthogonalization speed:}\quad & \Gamma_\perp = (2/\pi)\,\EG/\hbar \;\Rightarrow\; C= 2/\pi. \end{align}\] Their ratio \(C_{\mathrm{vis}}/C_\perp = \pi/2\) is exactly the width of the window \(\eqref{eq:C-window}\); the window is the ratio of two different observables, not the spread of one. An interferometer measures fringe-visibility decay, so the trace-fixed metric selects \(C= 1\) for the falsifiable prediction, and the \(2/\pi\) anchor is a distinct quantity (a speed-limit ceiling), not a competing value.

The honest residue

What the trace does not supply is the spectral shape that would make the Bures-diffusion coefficient exactly \(\EG/\hbar\) rather than \(\kappa\,\EG/\hbar\) with \(\kappa = \mathcal{O}(1)\). The Kubo–Martin–Schwinger analyticity strip gives a correlation time \(\tau_c \sim \beta_{\mathrm{mod}}\) and hence \(\kappa = \mathcal{O}(1)\), Markovian to leading order; but the super-Ohmic spectrum offers no zero-frequency knob to set \(\kappa = 1\) on the nose, so the exactness of \(C= 1\) is constraint-borne and is precisely the content of the operator equation \(K= 2\pi H_{\mathrm{phys}}\). In summary: the bound \(C\in [2/\pi, 1]\) is derived; \(C= 1\) is selected for the measured observable by the trace-fixed metric; and the uniqueness of \(C= 1\) is conjectured, equivalent to closing the finite-diamond construction at full rigour (Appendix 10).

Status of the finite-diamond (R1) construction

Section 3 summarizes the finite-diamond construction in prose; the table below sets out what is established and what remains open, so that the single open lever is visible at a glance. Entries marked as symbolic or lattice rest on the corresponding computation.

The structural content is that the diamond modular Hamiltonian is a single object whose gap is the energy scale, \[\begin{equation} \frac{\hbar}{2\pi}\,\Delta K= \EG \quad(\text{normalization-independent}), \end{equation}\] and whose flow on the intrinsic clock is the rate \(\Gdec = C\,\EG/\hbar\), with the infrared-safe form \(\Gdec = 2N(1-a^2)\,\EG/\hbar\) for a source at fractional radius \(a\) inside a diamond of any size, \(N\) the clock normalization. The energy scale is fixed; the coefficient \(C= 2N\) (for a centered source) is the one quantity the construction does not yet pin, and pinning it is equivalent to proving the operator equation.

The exact sphere self-energy and the falsification numerics

The decoherence times of Section 4 use the exact Diósi–Penrose self-energy of a uniform sphere, not the point-mass form. For a sphere of mass \(M\) and radius \(R = (3M/4\pi\rho)^{1/3}\) in a superposition of separation \(\Delta x\), the gravitational self-energy difference between the branches is \[\begin{equation} \EG(\Delta x) \;=\; \frac{GM^2}{R}\,\times \begin{cases} \tfrac12 s^2 - \tfrac{3}{16}s^3 + \tfrac{1}{160}s^5, & s = \Delta x/R \le 2,\\[1mm] \tfrac65 - R/\Delta x, & \Delta x \ge 2R, \end{cases} \label{eq:exact-sphere} \end{equation}\] with the two limiting behaviours \[\begin{equation} \EG \to \tfrac12\,\frac{GM^2}{R^3}\,\Delta x^2 \quad(\Delta x \ll R, \text{ tidal}), \qquad \EG \to 1.2\,\frac{GM^2}{R} \quad(\Delta x \gtrsim 2R, \text{ saturated}). \label{eq:sphere-limits} \end{equation}\] The decoherence time is \(\tdec = \hbar/\EG\), the visibility \(V(t) = e^{-t/\tdec}\).

The threshold.

In the saturated regime \(\EG = 1.2\,GM^2/R\) with \(R \propto M^{1/3}\), so \(\tdec \propto M^{-5/3}\): there is a sharp threshold mass below which the superposition is undecidable on laboratory timescales. Setting \(\tdec = 1\) s gives, for silica (\(\rho = 2200\) kg/m\(^3\)), \[\begin{equation} M_\star \;=\; \left[\frac{\hbar}{1.2\,G}\left(\frac{3}{4\pi\rho}\right)^{1/3}\right]^{3/5} \;\approx\; 7.6\times10^{-16}\ \mathrm{kg} \quad (R \approx 0.44\,\mu\mathrm{m}). \label{eq:Mstar} \end{equation}\]

The \(5\sigma\) run count.

A visibility deficit \(1-V\) is resolved against the standard null \(V=1\) with a per-run shot-noise error \(\sigma_V \approx 1/\sqrt{N}\), so \(5\sigma\) discrimination requires \((1-V)\sqrt{N} \ge 5\), i.e. \[\begin{equation} N_{5\sigma} \;=\; \frac{25}{(1-V)^2}. \label{eq:N5} \end{equation}\] This is an idealized floor for unit per-run fringe contrast; a real per-run contrast \(c<1\) multiplies it by \(1/c^2\).

The environmental budget.

The test is clean only if residual-gas decoherence \(\Gamma_{\mathrm{gas}} = n\sigma\bar{v}\) (with \(\sigma = \pi R^2\), \(n = P/\kB T\)) is slower than \(\Gdec = 1/\tdec\), requiring \[\begin{equation} P \;\lesssim\; \frac{\kB T}{\pi R^2 \bar{v}\,\tdec}, \qquad \bar{v} = \sqrt{\frac{8\kB T}{\pi m_{\mathrm{gas}}}}, \label{eq:Pmax} \end{equation}\] which for the envelope target (\(1.5\) fg, \(\Delta x = 1\,\mu\)m, \(T = 4\) K, helium) gives \(P \lesssim 7\times10^{-15}\) mbar—cryogenic extreme-high vacuum. Every value in Table 1 and Figures 1–2 follows from \(\eqref{eq:exact-sphere}\)–\(\eqref{eq:Pmax}\) and the physical constants.

← Back to Research