This project proposes to develop a scenario for the self-organization of goal-directed systems out of networks of (chemical) reactions. Related scenarios have been proposed to explain the origin of life starting from autocatalytic sets, but these sets tend to be too unstable and dependent on their environment to maintain. The new formalism of Chemical Organization Theory shows mathematically under which conditions reaction networks self-organize into self-maintaining, autopoietic "organizations". Through computer simulations, we will investigate what is needed to make an organization resilient, i.e. able to reach its goals in the face of environmental challenges. Using specially developed software, we will randomly generate and perturb computer-generated organizations, and select the ones that survive these perturbations, thus evolving increasingly resilient organizations.
The theory of cybernetics specifies the conditions needed for effective goal-directedness: reference levels for goals, negative feedback, requisite variety of actions and knowledge, hierarchy, feedforward and buffering. We will express these conditions mathematically in the language of COT, and then check in how far the evolved organizations satisfy these conditions. This will allow us to develop a taxonomy and space of resilient organizations. To test and refine these theoretical results, we will apply them to data about existing goal-directed systems, including metabolic networks and social systems. We will then determine the trajectories through the resilience space that are easiest to evolve, while comparing them with known evolutionary transitions. This should result in a realistic, step-by-step scenario for the evolution of goal-directedness, thus providing a theoretical solution to the age-old question of the origins of purpose. The results of our research will be delivered as publications, talks, videos, websites and demonstrations, aimed at both working scientists and the public.