Publication type: Research Monograph [Book]
Author: Rui A. P. Perdigão
Date: May 2020
Title: Synergistic Dynamic Causation and Prediction in Coevolutionary Spacetimes
Set: Monographs on Dynamical Systems and Complexity (M-DSC)
Indexed: Yes (Crossref)
Methodological Keywords: Causation, Prediction, Causality, Thermodynamics, Physics of Time, Synergy, Coevolution, Complexity, Emergence, Complex Systems, Dynamical Systems, Information Theory, Information Physics, Fluid Dynamics, Predictability, Mathematical Physics, Differential Geometry, Fractional Differential Geometry, Chaos, Entropy, Turbulence, Critical Transitions, Extreme Events, Non-Ergodic Information.
Restricted Access: Login here
This contribution settles to rest the argument on whether classical dynamical system theories are able to tackle causation (they are not), and whether traditional information theories are able to tackle causation (they are not). The underlying physics are notoriously absent from both classical dynamical systems and information theoretic frameworks still in use in engineering and in the earth and environmental sciences.
In fact, the classical approaches simply focus on a naïve descriptive assessment of kinematic geometry (dynamical system theories), and statistical-geometric relations (information theory), under very restrictive conditions far removed from realism.
Dynamical system theories and information theories are only able to provide diagnostic and predictive power – and then again to some extent. As such, they are operational mathematical and statistical techniques with inherent value in providing prediction frameworks when systems are “well-behaved” to the extent that their fundamental properties recur enough to give a reasonable predictive score to them.
However, prediction is beyond reach in such frameworks when non-ergodic coevolutionary systems come into play, where the invariants of motion are elusive, where there is no recurrence, no attractor, no statistical mechanics to hold on. To those classical techniques, anything not captured by them is deemed “black swan”, labeled as unpredictable until it happens. This tells scores about the practical obsolescence of classical dynamical systems and information theoretic frameworks in real-world decision making. Science – and society even more – needs a new approach to tackle prediction, and do so with fundamental causation (not merely inference) in mind.
With the new contribution “Synergistic Dynamic Causation and Prediction in Coevolutionary Spacetimes”, a new treatise on the mathematical physics of causation and predictability is thoroughly derived and discussed. Without any constraints, all derivations are thoroughly and exhaustively performed, as a service to fundamental mathematics on one hand, and to applied science and decision support agents on the other hand.
The theoretical contribution is accompanied by down-to-earth applications to detection, attribution and prediction of emerging synergistic natural hazards in under far-from-equilibrium multiscale coevolutionary socio-environmental earth system dynamics. Those applications are not only scholarly but also down-right operational, already assisting our institutional partners in formulating more informed decisions. Because it is fundamental that anything developed is aptly and swiftly applicable to assist those on the line of action protecting our society and the environment.