Assessing the Potential of Participatory Modeling for Decolonial Restoration of an Agro-Pastoral System in Rural Zimbabwe

Abstract

Colonialism has devastated the lifeways, more-than-human relations, and collective stewardship practices of Indigenous people. Decolonial restoration may be assisted by collaborative methods like participatory modeling, but further careful evaluation is needed to ensure that these methods that have the potential to connect ways of knowing actually do secure long-term benefit for Indigenous life and land. In this study, we engage in a mixed-methods analysis to assess our participatory modeling of a Zimbabwean agropastoral system. The Muonde Trust, a community-based research organization, partnered with international researchers from outside the community to create an agent-based model (ABM) representing the dynamics of land use in Mazvihwa Communal Area, Zimbabwe. Using interviews and participant observation during modeling workshops, Muonde and their allies assessed the immediate increases in confidence and self-efficacy for members of the research team (during workshops), intermediate-term changes in local land-use policy and management discussions (months to years later), and long-term changes in on-the-ground land use (up to four years later). We find that the model successfully assisted the Muonde Trust in working with local leaders to create policies allowing recultivation of fallow fields rather than further deforesting woodland grazing areas. This success is due to the involvement of the community at key times in the modeling process, resulting in a model that felt accessible and that Muonde (as a key information broker) could use to bring the community together to discuss collective management. Though aspects of our process still relied on colonial tools and power structures, the community in Mazvihwa finds the model useful and feels ownership over it.

Keywords

• mixed methods
• agentbased model
• ecocultural restoration
• situational analysis
• betweenness centrality
• participatory modeling