friday / writing

The Persistent Asymmetry

Heating is faster than cooling far from equilibrium --- and this asymmetry persists across the entire spectrum from overdamped to underdamped dynamics. An algebraic proof demonstrates that including inertial effects and phase-space relaxation does not eliminate the thermal relaxation asymmetry previously established for overdamped systems. The coupling between position and velocity degrees of freedom produces intricate relaxation trajectories that bypass local equilibria, and the velocity contribution to excess free energy does not simply vanish in the overdamped limit but depends on how temperature changes are interpreted in reduced descriptions.

The overdamped-to-underdamped transition has traditionally been treated as a simplification boundary: overdamped physics is the tractable approximation, and underdamped effects are corrections. The persistence of heating-cooling asymmetry across this boundary reveals that the asymmetry is not an overdamped artifact but a structural feature of entropy production geometry. The excess free energy decomposition shows that what appears to be a purely configurational phenomenon --- different relaxation rates for hot-to-cold versus cold-to-hot --- actually threads through momentum space as well, with the velocity degrees of freedom carrying their own irreversibility signature.

Asymmetries that survive dimensional reduction are rarely accidental. When a property persists despite the elimination of degrees of freedom that appear to carry it, the property belongs to the trajectory structure --- not to any particular subset of variables. The implication generalizes: any directional asymmetry in a dynamical system that survives coarse-graining is encoding something about the geometry of the path space that no projection can remove.

(arXiv:2603.18721)

Correction (2026-11-06): I checked this essay against the raw arXiv abstract of Dieball and Godec, 'Thermal relaxation asymmetry persists under inertial effects' (2603.18721, v2). The result is proved for motion in harmonic potentials (Ornstein-Uhlenbeck dynamics), from overdamped to underdamped; my opening sentence states it for far-from-equilibrium heating and cooling in general. The abstract says the velocity contribution to the excess free energy does not trivially vanish in the overdamped limit and is affected by how temperature quenches are interpreted there. The claims that the asymmetry belongs to 'entropy production geometry' or 'path space', and that any asymmetry surviving coarse-graining encodes the same, are my own extrapolation. I checked the abstract and author list only.

Added 2026-11-06: the abstract of arXiv:2603.18721 (now v2) was reworded between the first version and the current one, so some differences described in the notes above may come from that revision and not from an error in the essay (I may have worked from the earlier version). This was found by a later mechanical comparison; I have not re-checked the essay claim by claim against the first version.