GR vs. QFT

Here’s the gist!


The problem of quantum gravity is usually described as technical. Quantum field theory’s equations blow up at small scales. The math produces infinities that no known procedure can clean up. We need, the story goes, a clever new technique — string theory, loop quantum gravity, something else — that tames the divergences and finally unifies our two best theories of nature.
That story is incomplete. The problem is not technical. It is topological.
Start with a black hole. A black hole is one of the most extreme objects in nature. Anything that falls in — a star, a planet, a library, a person — is gone. All the complexity, all the information, all the history: gone. But not everything is lost. Three things survive, no matter what fell in: the total mass, the total electric charge, and the total angular momentum. That is it. Three numbers completely describe any black hole in the universe. This is the no-hair theorem, and it has been established rigorously.
Why three? This is the question that unlocks everything.
The answer is not arbitrary. The event horizon of a black hole — the boundary of no return — is topologically a sphere. A sphere has a specific symmetry structure: it admits exactly three independent continuous symmetries in the relevant physical setting. Time translation. Axial rotation. Electromagnetic gauge transformation. By Noether’s theorem — one of the most important results in all of physics — each continuous symmetry produces exactly one conserved quantity. Three symmetries. Three conserved quantities. Three numbers. The sphere cannot hold more. This is its topological capacity. The gate preserves only what the topology of its boundary can encode.
Now look at classical mechanics. The same count appears, for the same reason. In the Lagrangian formulation, a physical system is described by generalized coordinates and their time derivatives, and the dynamics are selected by the Principle of Stationary Action — the requirement that the physically realized trajectory extremizes the action integral. In the Hamiltonian formulation, the same dynamics are expressed on phase space: position coordinates paired with their conjugate momenta, evolving under Hamilton’s equations. Both formulations operate on a three-axis structure: position, momentum, and time. Not two axes, not four. Three.
This is not a coincidence of convention. The Lagrangian formulation is symmetric under spatial translation, rotational symmetry, and time translation. By Noether’s theorem, these three symmetries yield conservation of momentum, conservation of angular momentum, and conservation of energy respectively. The Hamiltonian formulation makes the same structure explicit: the symplectic geometry of phase space is organized by exactly these three independent directions. The Lagrangian and Hamiltonian formulations are related by the Legendre transform, which maps between the velocity space of the Lagrangian and the momentum space of the Hamiltonian — and both live on the same three-axis foundation. Classical mechanics saturates the topological capacity of the spherical boundary, arriving at the same count as the black hole by a completely independent route.
Now look at quantum field theory. QFT is our most precise description of the subatomic world, tested to extraordinary accuracy. But look at what it treats as dynamical. It promotes two quantities to dynamical variables — the field value and its conjugate momentum — and fixes spacetime geometry as a prior commitment. The metric of spacetime is not a dynamical participant in QFT. It is a stage, set in advance, on which the quantum fields perform.
This is not a gap in QFT’s observable algebra. QFT conserves mass, charge, and angular momentum correctly. What is missing is not a conserved quantity but a symmetry — diffeomorphism invariance, the freedom to treat smooth relabelings of spacetime points as a genuine dynamical freedom rather than a background assumption. In general relativity, no coordinate system is preferred; the metric itself evolves in response to matter. In QFT, the metric is fixed before the theory begins. The Noether charge of diffeomorphism invariance — essentially the dynamical content of the gravitational field — is absent not because QFT forgot to conserve something, but because the symmetry that would generate it was eliminated as a precondition. You cannot derive the assumption from within the system that was built on it.
Now look at general relativity. GR treats spacetime not as a fixed background but as a dynamical participant: matter tells spacetime how to curve, and spacetime tells matter how to move. The metric evolves. It is a genuine dynamical variable. Three dynamical axes — matter, momentum, and spacetime geometry — all coupled. No fixed background. GR saturates the topological capacity of the sphere. QFT, operating with two dynamical axes, does not.
The incompatibility is this. When you try to merge QFT with GR by adding perturbative corrections, you are attempting to introduce a new dynamical axis from within a framework that fixed that axis as a precondition. The corrections are written on a background metric. They inherit the two-axis structure of QFT. At every loop order, the missing dynamical axis asserts itself as a divergence — not a technical nuisance awaiting a clever regularization scheme, but the absent symmetry making itself felt at the boundary of the theory’s foundational commitment.
The known paradoxes are the same diagnosis. The black hole information paradox arises because Hawking’s original calculation fixes the black hole background. It is a two-axis calculation applied to a three-axis situation. The information is carried by the dynamical degree of freedom — spacetime geometry itself — that the calculation excludes by assumption. The firewall paradox forces a choice between unitarity, the equivalence principle, and effective field theory at the horizon — three requirements that a complete three-axis theory would satisfy simultaneously. The impossibility of satisfying all three within QFT is the signature of a two-axis framework encountering a three-axis reality.
Any successful theory of quantum gravity must treat all three dynamical axes as genuine participants — field value, conjugate momentum, and spacetime geometry itself. No fixed background at any scale. This is a necessary condition. It does not specify which theory will succeed. It specifies the topological category within which the successful theory must live. Perturbative approaches reintroduce a fixed background at the level of the perturbative metric and remain two-axis theories. Non-perturbative approaches — string theory’s non-perturbative formulations, loop quantum gravity, others — are each attempting, in their own way, to restore the third axis. Their difficulties arise precisely where they slip back into fixed-background assumptions.
The search is not for a new equation. It is for a theory that treats spacetime topology itself — not the metric, not the connection, but the topology — as the fundamental dynamical variable.


The full mathematical treatment is in the papers below, also available on the Compendium page!