This dissertation builds a philosophical model, SEER, of fitness-maximizing, highly interconnected systems in highly dynamic environments. I argue that such systems are maximally robust, where robustness is the property of maintaining functionality across multiple environments. I then argue that maximally robust systems utilize a dual phase evolutionary strategy, the developmental strategy. The developmental strategy consists of an exploration phase, whereby the system experiments with various configurations, and a selection phase, whereby the system selects that configuration, from the set of experimental configurations, leading to the highest fitness levels. The use of the developmental strategy entails that the system is plastic, modular, model building, and path dependent, which researchers can then test for in systems.

Although SEER is not a biological model, I construct the model based on questions within the philosophy of biology. Through this application, I show that, at the systems level, a model of fitness-maximizing, highly interconnected systems in highly dynamic environments yields testable implications. Thus, claims regarding the nonexistence of methodologies based on such assumptions are false. If a methodology is not possible, it is because of what we know about a particular system and not due to the properties of systems in general. I then apply SEER to discussions in organizational economics and show that maximally robust systems will not give preferential weight to its previous configurations. Next, I apply SEER to discussions in regional economics and show that maximally robust systems will utilize the developmental strategy regardless of their current fitness levels. Lastly, I apply SEER to ecosystem management and show that for robustness to apply to a system, that system must possess a converging coherence network, which we can conceptualize as a network of cooperating parts.