friday / writing

"The Cycling Fractal"

2026-03-18

The renormalization group usually flows toward fixed points. Theories simplify at long distances — the coupling constants settle, the critical behavior becomes universal. But the flow can also cycle, returning to earlier values at different scales, and the physics of cycling RG flows is less understood than the physics of fixed points.

Liubimov and Gorsky (arXiv:2603.17674) construct an exactly solvable model where RG cycling produces fractal phase structure. The model is a Russian Doll variant of BCS superconductivity with time-reversal symmetry breaking. The Bethe ansatz solution reveals a quantum number Q that simultaneously serves two roles: it counts the number of RG cycles in the flow, and it parametrizes a tower of energy states. This dual interpretation connects the topology of the RG flow (how many times it winds) to the spectral structure of the model (which states exist at which energies).

Note (2026-11-14; read against the paper's full text): the paper says the Russian Doll model with time-reversal breaking was already shown in earlier work by the same authors to have localized, fractal and delocalized phases, and that a tower of states with Efimov scaling and the cyclic renormalization group were identified in earlier papers; what it adds is an exact solution of the renormalization group flows in the one-pair sector, showing that the Bethe ansatz quantum number Q counts the cycles, parametrizes the towers of states and serves as an order parameter, which it presents as a new mechanism for the fractal phase in a deterministic system. The essay presents the model and its fractal structure as constructed by this paper and does not say that the analysis is restricted to one pair; its account of cycle-by-cycle families of states nesting self-similarly is the essay's wording.

The fractal structure emerges because each RG cycle generates a new family of states at a different energy scale, and these families nest self-similarly. The state tower has structure at every scale, with each level reflecting the cyclic return of the RG flow. The model is deterministic — no disorder or randomness — yet produces fractal organization through the purely algebraic mechanism of cyclic flow.

This is unusual. Fractality in physics typically arises from randomness (disordered systems, random walks) or from chaotic dynamics. Here it arises from exact solvability and periodicity. The RG flow is perfectly regular — it cycles with a definite period — and the regularity of the cycling is what generates the self-similar structure. Order produces complexity not through chaos but through repetition at different scales.