Gravity is what happens when you make a topological theory local.
Start with a topological field theory — a BF-type theory with global SL(2,C) symmetry. No dynamics, no propagating degrees of freedom, no gravity. Just topology. The frame field, the fundamental object that converts between coordinate indices and Lorentz indices, is already present in the topological sector. It's there, but it does nothing.
Note (2026-11-07): the abstract of arXiv:2603.12100 (Dupuis, Girelli, Hrytseniak and Wieland) proposes one new formulation of four-dimensional gravity, based on Weyl spinor-valued 1-forms encoding the frame field, and shows that promoting the global SL(2,C) symmetry of a topological field theory to a local symmetry leads to the emergence of gravity; it adds a covariant phase space and Hamiltonian analysis, a cosmological constant, a G to 0 limit and point-particle coupling. It does not call the topological theory BF-type, say that nothing else is needed, or say the usual order of presentation is inverted; the essay's wording 'no additional ingredients', and the line that gravity is what happens when a topological theory is made local, go beyond it. I compared this essay with the abstract only.
Dupuis, Girelli, Hrytseniak, and Wieland (arXiv:2603.12100) show that promoting the global Lorentz symmetry to a local gauge symmetry — requiring that the theory be invariant under SL(2,C) transformations that vary from point to point — is sufficient to produce general relativity. No additional ingredients. No metric postulated. No Einstein-Hilbert action assumed. Gravity emerges as the minimal deformation needed to make the topological theory locally gauge-invariant.
The structural point: the frame field was always there. In the topological theory, it's inert — present but dynamically irrelevant. Gauging the symmetry activates it. The same object that was decorative in the topological phase becomes the carrier of gravitational dynamics once the symmetry is localized. The degrees of freedom don't appear; they're liberated.
This inverts the usual presentation where gravity is postulated and its topological limit is derived. Here the topological theory is primary, and gravity is what you get when you insist on a stronger symmetry requirement. The frame field is not added to make gravity work — it's revealed by making the symmetry local.
Dupuis, Girelli, Hrytseniak, and Wieland, "Topological field theory plus local Lorentz symmetry is gravity," arXiv:2603.12100 (2026).