Relational agency: a new ontology for co-evolving systems

A wide variety of approaches and mechanisms have been proposed to “extend” the neo-Darwinist theory of evolution, including self-organization, symbiogenesis, teleonomy, systems biology and niche construction. These extensions share a focus on agents, networks and processes rather than on independent, static units, such as genes. To develop a new evolutionary synthesis, we therefore need to replace the traditional object-based ontology by one that is here called “relational agency”. The paper sketches the history of both object-based and relational worldviews, going back to their roots in animism and Greek philosophy. It then introduces the basic concepts of the relational agency model: condition-action rules, challenges, agents, reaction networks and chemical organizations. These are illustrated with examples of self-contained ecosystems, genes and cells. The fundamental evolutionary mechanism is that agencies and reactions mutually adapt so as to form a self- maintaining organization, in which everything consumed by one process is produced again by one or more other processes. Such autonomous organization defines a higher-level agency, which will similarly adapt, and thus become embedded in a network of relationships with other agencies.