Fractional epidemic dynamics — heavy-tailed super-spreading events, temporal avalanches, power-law waiting times — are typically modeled by replacing ordinary derivatives with fractional derivatives in the compartmental equations. The fractional calculus is inserted by hand: the modeler chooses a fractional exponent to fit the data, without deriving it from an underlying mechanism.
Note (2026-11-14; read against the paper's full text): the essay says the fractional exponents are determined by the coupling constants of the infection dynamics and that the reproductive number can be supercritical at short timescales and subcritical at long ones. In the paper the anomalous exponent alpha (between 0 and 2) is the exponent of the fluctuating host field's propagator in the anomalous regime; the transmission and emission couplings enter an effective coupling, and I did not find a statement that alpha is fixed by them. I found no occurrence of supercritical or subcritical. What the paper states is that the static effective reproductive number is bounded by the ultraviolet cutoff, interpreted as the inverse of the minimum microscopic distance between hosts, and that a super-spreader-driven epidemic could then only be collapsed by altering the microscopic spatial geometry of the network. It is a theoretical derivation; I found no numerical simulation or data comparison in it.
The fractional dynamics emerge from first principles through one-loop corrections in a non-equilibrium quantum field theory. The standard SIR epidemic model can be written as a field theory (the Doi-Peliti formalism), and computing quantum loop corrections to the propagator produces fractional space-time behavior as a natural consequence. The fractional exponents are not free parameters — they are determined by the coupling constants of the underlying infection dynamics.
The key transformation: the effective reproductive number R₀ changes from a scalar to a spectral dispersion relation. In the classical model, R₀ is a single number that determines whether an epidemic grows or decays. After loop corrections, R₀ becomes frequency-dependent — it takes different values at different timescales. At short timescales, the epidemic can be supercritical (growing) while at long timescales it is subcritical (decaying), or vice versa. The anomalous outbreak statistics that fractional models describe — super-spreading, clustering, temporal heterogeneity — are consequences of this scale-dependent criticality.
The structural observation: the ad hoc fractional calculus used to model epidemic anomalies is the leading-order quantum correction to the classical epidemic field theory. The anomalous behavior is not a deviation from the standard model that requires a different mathematical framework — it is the next term in the perturbative expansion of the same model.