Crisis-driven evolutionary learning: Conceptual foundations and systemic modeling

This study is premised upon the hypothesis that the global international system is capable of learning. Learning is hypothesized to occur in response to systemic crises. This learning occurs, initially, at the level of the individual decision-maker. It is subsequently incorporated into the structure of the international system in the form of generalized norms, or actualized decisional bodies, such as international regimes.; An international system so constituted possesses complex structure which incorporates many informational and decisional pathways. It is capable of evolving over time in response to crises. A system which is capable of learning may be characterized as comprising a complex adaptive system.; An evaluation of crisis-driven learning's theoretic premises is carried out, successively, at the individual, nation state, and, systemic levels. These findings are compared and contrasted with those of rational actor/expected utility theory.; Based upon the preceding investigations at several levels of analysis, a computer model of a simplified international system, consistent in its design parameters, as well as its underlying assumptions, with these findings, is constructed. This model is designed to represent a specific predicted instance (SPI) of evolutionary learning theory, wherein changes in the frequency of warfare between system elements which are capable of 'learning' in the context of a payoff matrix which varies non-linearly with respect to power are evaluated. Where individual cognitive parameters are relevant, the model incorporates them.; The basic idea is that the model's actors, representing nation-states, interact with one another along several dimensions, which are derived from the preceding analysis. These interactions lead to absolute, as well as relative, changes in power between them.; Given the ability of these actors to 'learn', a complex adaptive systemic learning model would predict that certain behavioral changes with respect to the frequency of warfare, in the context of increasing power, over time, ought to be observed. These predictions are distinguishable from non-systemic learning models. This section operationalizes the systemic learning model. It demonstrates that fairly simple rules, which are logically consistent with evolutionary learning theory, but not expected utility theory, may account for the system's seemingly complex observed behaviors.