Use and Abuse of General Criterion for Discrimination of Causal Correlations from Individual Peculiarities and Provisional Correlations

摘要

A fundamentally novel approach to the issue about existence of a general criterion for autonomous discrimination of causal correlations from individual peculiarities and provisional correlations at stable complex systems is put forward. It is grounded on a recently proven by the author decomposition theorem whose subject has no cross section with the subject of the Central Limit Theorem. The fundamental advantage of that criterion lies in insensitivity to the details of the underlying dynamics and to the details of the hierarchical structure regardless of the nature of the corresponding system. It holds in an unspecified ever-changing environment. It also holds when information is incomplete and/or uncertain. Another advantage of the criterion is the ability to forecast a change in a system. The limitation of the criterion is substantiated as a ban over predictability whether that change would develop in an adaptation or in destruction. It is worth noting that while the criterion itself holds in the frame of the recently proposed theory of boundedness, the ban over prediction of the nature of a change is model-free.