Worse than Complex

Doctoral thesis, 2017

This thesis engages with questions on the boundary between what has traditionally been understood as social and natural. The introductory essay contextualizes the specific contributions of the included papers, by noting and exploring a reinvigoration of "naturalism" (the notion of a continuity between the human realm and the rest of natural phenomena) under the banner of Complexity Science. This notion is put under explicit light, by revisiting the age-old question of naturalism and connecting ideas in complexity science with the work of e.g. Roy Bhaskar, Mario Bunge, William Wimsatt, and David Lane. A philosophical foundation for a complexity science of societal systems is thereby sketched, taking the form of an integrative and methodologically pluralist "complex realism". The first two papers provide a theoretical perspective on the distinction between social and natural: Paper I notes that societal systems combine two qualities that are commonly referred to as complexity and complicatedness into an emergent quality that we refer to as "wickedness", and that is fundamentally and irreducibly different from either quality in isolation. This explains the recalcitrance of societal systems to the powerful approaches that exist for dealing with both of these qualities in isolation, and implies that they indeed ought to be treated as a distinct class of systems. Paper II uses the plane spanned by complexity and complicatedness to categorize seven different system classes, providing a systematic perspective on the study of societal systems. The suggested approach to societal systems following from these conclusions is exemplified by three studies in different fields and empirical contexts. Paper III combines a number of theories that can be seen as responses to wickedness, in the form of evolutionary developmental theories and theories of societal change, to develop a synthetic theory for cultural evolution. Paper IV exemplifies how simulation can be integrated with social theory for the study of emergent effects in societal systems, contributing a network model to investigate how the structural properties of free social spaces impact the diffusion of collective mobilization. Paper V exemplifies how digital trace data analysis can be integrated with qualitative social science, by using topic modeling as a form of corpus map to aid critical discourse analysis, implying a view of formal methods as aids for qualitative exploration, rather than as part of a reductionist approach.