Spectral Topology Can Prove a System Is Out of Equilibrium — But Can't Tell You How Far

A tutorial synthesis unifying Berry phases, dislocation geometry, and Markov cycle affinities under one question: when do spectral features force entropy production?

Imagine you're looking at a one-way street system on a city map. If you can trace a loop and end up going one direction the whole way around, you know traffic doesn't flow equally both ways — the system is irreversible. But the mere existence of that one-way loop tells you nothing about how much traffic is actually flowing. You know the city isn't at rest; you don't know if it's rush hour or 3 a.m. That's the core mechanism of this paper: topological invariants in a system's spectrum can certify that detailed balance is broken (the system is out of equilibrium and producing entropy), but they cannot quantify how much dissipation is happening. The committed claim: spectral topology obstructs detailed balance but does not determine the magnitude of dissipation. Three holonomies — the Berry phase of a band, the Burgers vector of a crystal dislocation, and the cycle affinity of a Markov network — recur across mechanical metamaterials, active fluids, stochastic networks, and open quantum systems. Hunt rederives the Chern numbers C_m = m(sgn Ω + sgn ν_o) for a compressible odd-viscous fluid and restates the Paulose-Chen-Vitelli dislocation zero-mode count as a pairing of integrated torsion with topological polarization. The unifying thread is that all three are holonomies of different connections, and each enforces qualitative constraints on the physics without pinning down quantitative rates. The paper's sharpest result is the inequality σ ≥ k_B j²/D for entropy production σ given any current j — a thermodynamic uncertainty relation that puts a floor on dissipation but leaves the ceiling at infinity. The Hatano-Nelson ring makes this concrete: it is normal (so spectrally well-behaved in the operator sense) yet has complex spectrum, and its entropy production sweeps the entire range (0, ∞) at fixed winding number. This is the cleanest demonstration that winding number is a qualitative gate, not a quantitative dial. Exceptional points (EPs) get a careful, case-by-case treatment. With generalized detailed balance including time-reversal parity, an EP is compatible with equilibrium only through time-reversal-odd dynamics — the critical damping of a Langevin oscillator is the canonical example. But for systems with only even variables, such as odd elastic media, an EP certifies a nonequilibrium steady state. In both cases entropy production is analytic across the EP, meaning the EP is not a phase boundary for dissipation even though it reorganizes the spectral geometry. This is a genuinely useful clarification: EPs are trendy in non-Hermitian physics, and many papers treat them as smoking guns for nonequilibrium behavior without checking parity. The paper is explicitly a tutorial synthesis — it does not introduce new mathematical machinery or new physical systems. Its contribution is organizational: arranging known results from condensed matter, stochastic thermodynamics, and open quantum theory under a single holonomy-based framework, and stating cleanly what the spectral data can and cannot tell you. The Chern number rederivation and the dislocation zero-mode restatement are not new results but new framings. This matters because the communities working on topological metamaterials, non-Hermitian quantum mechanics, and stochastic thermodynamics often reinvent each other's tools without cross-referencing. Validation is mathematical: the claims are theorem-level statements backed by derivation, not numerical experiments. Verification scripts in Wolfram Language and Python are included as ancillary files, which is above-average for a theory paper. The proposed benchtop tests — for the Hatano-Nelson ring, for EP parity signatures, for the dislocation zero-mode count — are specific enough to be actionable, though none have been performed yet. The paper's honesty about the gap between spectral obstruction and quantitative prediction is its strongest intellectual feature. The field fight here is real: non-Hermitian topological physics has generated enormous excitement, and there is a persistent temptation to over-interpret topological invariants as controlling dynamics rather than merely constraining them. This paper plants a flag on the conservative side — topology is a qualitative gate, not a quantitative lever — and does so with enough worked examples to be pedagogically useful. For anyone building intuition across these overlapping communities, the holonomy-based organization is the payoff.