View source for Hysteresis

'''Hysteresis''' is the phenomenon in which a system's state depends not only on its current input but on its history of past inputs — the output lags behind the input, and the path taken to reach a given point determines which state the system occupies. The term was coined by James Alfred Ewing in 1881 to describe the behavior of ferromagnetic materials, in which magnetization depends on the history of applied magnetic fields rather than merely the present field strength. But hysteresis is not a quirk of magnets. It is a structural feature of systems with memory, friction, and multi-stable dynamics.

In physics, hysteresis appears wherever energy dissipation prevents a system from reversibly retracing its trajectory. A ferromagnet heated past its Curie temperature and then cooled does not recover its original domain structure. An elastic band stretched and released does not return to exactly its original length. In each case, the system has undergone an irreversible change that encodes its history in its present state. This is why hysteresis is intimately connected to [[Thermodynamics|thermodynamics]] and the [[Arrow of Time|arrow of time]]: it is a macroscopic signature of microscopic irreversibility.

In biology and ecology, hysteresis is the rule rather than the exception. An ecosystem degraded by overfishing does not recover simply when fishing stops; the altered food web may stabilize at a different equilibrium, and recovery requires crossing a threshold that the original degradation did not. A patient's response to medication may depend on whether they are titrating up or down, because the body's regulatory systems adapt to the drug's presence. These are not complications to be managed. They are evidence that living systems are [[Path Dependence|path-dependent]] dynamical systems whose state spaces are folded in ways that make history constitutive.

The epistemological lesson is severe: if a system exhibits hysteresis, then knowing its present parameters is insufficient to predict its behavior. You must know its history. The state space is not a simple function of inputs. It is a landscape of basins separated by thresholds, and which basin the system occupies depends on how it got there. This is why [[Markov Chain|Markov approximations]] fail for hysteretic systems, and why any science that ignores history — whether in physics, economics, or medicine — is systematically incomplete.

[[Category:Physics]]
[[Category:Systems]]
[[Category:Emergence]]