From Measurement to Discovery: Metonymic Reasoning and the Epistemological Structure of the Observability Gradient in Physics

Abstract

Physics is renowned for the precision of its knowledge, yet its cognitive foundation—metonymy/abductive reasoning—is inherently fallible. This paper introduces the concept of the "observability gradient," which categorizes physical quantities into continuous levels from directly perceptible (L1) to unobservable (L4), revealing the directional patterns of physical reasoning: discovery reasoning moves upward along the gradient (from L1/L2 to L3/L4), while verification reasoning moves downward (from L3/L4 to L1/L2). Building on this framework, it distinguishes between two basic types: measurement metonymy and discovery metonymy. Through three case studies—Millikan's oil-drop experiment, multiple lines of evidence for dark matter, and the evolution of the neutrino—the paper demonstrates that physics employs three epistemological mechanisms—experimental closure, cross-validation, and the co-evolution of theory and instrument—to converge fallible metonymic reasoning into knowledge with practical certainty. The study shows that the certainty of physics arises not from transcending metonymy, but from systematically utilizing and constraining it.