The category error at the heart of clinical medicine: Lyapunov divergence, fractal geometry, and the geometry of physiological complexity

Clinical medicine operates under an epistemological architecture inherited from eighteenth-century deterministic mechanics — a framework whose collapse as a universaldescription of natural systems was documented across the twentieth century by quantum mechanics, chaos theory, and fractal geometry. This Perspectives article formalizes the mathematical core of that collapse as it bears specifically on biological systems. We develop the Lyapunov spectrum as a rigorous characterization of sensitive dependence on initial conditions, derive the Kaplan–Yorke dimension as a bridge between dynamical instability and attractor geometry, and reconstruct the Grassberger–Procaccia algorithm as the instrument that makes fractal dimension empirically accessible from physiological time series. The convergence of these formalisms with Goldberger and West’s fractal scaling of heart rate variability, and with the Tononi–Edelman information-theoretic complexity measure, constitutes not a collection of adjacent findings but a systematic demonstration that the organism is a dissipative structure whose health is a dynamical property irreducible to point-in-time biomarkers. The central epistemological claim is precise: the application of linear, equilibrium-based statistical tools to nonlinear, far-from-equilibrium systems is not an approximation. It is a category error in the sense Wittgenstein assigns to grammatical mistakes — a misidentification of the object that generates structurally false predictions regardless of the quality of the data collected under its assumptions. A concluding section addresses the objection that a structural diagnosis carries an obligation to produce its own methodological resolution — an objection that, in demanding an immediate pipeline from a geometric argument, reproduces exactly the institutional reflex the article diagnoses. The division of scientific labor has neveroperated otherwise: Poincaré did not owe Lorenz a weather model; Riemann did not owe Einstein a theory of gravity. The precision of the diagnosis is the contribution.The resolution belongs to the research programs it makes possible.