Modeling and Simulating Complex Conflict Management Using Reaction Networks

Evidence suggests that protracted conflicts persist because several forms of socio-political organization run simultaneously on the same population, resources, and territory. Reading Service's typology of bands, tribes, chiefdoms, and states not as evolutionary stages but as coexisting and superposed social organizations, we model conflict as a reaction network where each social form is a self-maintaining set of stocks-a chemical organization-and conflicts arise where competing productive logics between organizations generate stocks with negative connotation, such as grievances and displacement. Taking the Lake Chad Basin as inspiration, we build a ladder of progressively richer models arriving a mixed chiefdom-state configuration compatible with current views on the conflict. As the model complexifies, kinetic approaches become uninformative; we therefore develop complementary stoichiometric methods that are parameter-free and thus are far easier to measure and compute. These diagnostics reveal a structural bias toward conflict: transitions into conflict regimes are systematically richer than transitions out. We show how a dual chiefdom-state form acts as a conflict attractor within a closed conflict-peace loop that transits among documented different forms of organization. Conflict management then becomes the identification of the mechanisms that redirect rather than change the state of a self-sustaining organization-here, elite-surplus redistribution-and of the timescales at which such redirection is observable, turning intervention design into a structural rather than a parameter-tuning problem.