The incompatibility of quantum mechanics and general relativity arises from a shared foundational error: both frameworks treat time as a background parameter rather than a physical process. Relational Actualism (RA) dissolves this tension by replacing both the Schr.dinger equation and the Einstein field equations with a single Covariant Stochastic Graph Rewriting System—the Engine of Becoming. The algorithm operates on the growing causal Directed Acyclic Graph (DAG) of actualization events at three levels: (1) a graph-theoretic primitive layer describing candidate extensions of the causal boundary; (2) a quantum measure layer assigning complex amplitudes to causal histories in a foliation-independent way, using Sorkin’s quantum measure formalism; and (3) an emergent layer in which the Feynman path integral and the Einstein field equations are the macroscopic descriptions: the quantum measure is exact at Level 2 and the Feynman integral is its large-μ approximation; the Einstein field equations are the unique consistent Level 3 description, inevitable by Lovelock’s theorem Lovelock, 1971 given the Lean-verified LLC and causal invariance. The foliation-independence of the quantum measure (causal invariance) is the RA statement of Lorentz covariance at the discrete level, and its combinatorial core has now been formally verified in the Lean 4 theorem prover. Furthermore, by defining actualization via an irreversible increase in quantum relative entropy and leveraging the Continuous Functional Calculus, we prove the Stationarity Criterion for the Rindler thermal state, resolving the Unruh paradox without breaking Lorentz covariance. The algorithm makes a specific, near-term falsifiable prediction: a strict null result for gravity-mediated entanglement experiments (the Bose-Marletto-Vedral protocol), because the spacetime metric is updated strictly from actualized vertices and cannot be placed in quantum superposition. Substantial progress on the generative measure: the Benincasa-Dowker sign criterion (SBDG(v)>0) provides the discrete actualization filter, yielding a Poisson-CSG formula cRA k (x) ∝ e−μμk with μ = λVcoh pth that is local and environment-dependent, recovering ds = 3 in dense environments and ds = 2 (Durhuus-Jonsson-Wheater) in sparse ones.
Joshua Sandeman (Fri,) studied this question.