Causality and Dependence as Container Projections

Causality is traditionally understood as an objective relation between events: some events bring about others. Similarly, dependence is often treated as a directed relation of the form “A depends on B.” However, this paper argues that both causality and dependence are not intrinsic features of reality but arise from the way a system is modeled. Specifically, they emerge from the selection of a decomposition (container) and a layer of analysis. Using detailed examples (including train motion, decision-making behavior, the three-body problem, and biological relations), it is shown that causal and dependency structures are not invariant: they depend on the level of decomposition, the type of relations considered, and the direction of interpretation. Axioms and theorems are formulated demonstrating that such relations can be understood as projections of multidimensional dependencies within a chosen model. It is concluded that the search for a “true cause” or “absolute dependence” outside a modeling framework is ill-posed.