Working Paper
The experimented society: interventions, social science, and the failure of post-conflict reconstruction in Afghanistan

This paper critically examines the shortcomings of post-conflict reconstruction efforts in Afghanistan from 2001 to 2021, arguing that an overemphasis on measurable results and causal inference led to overly narrow, community-driven development interventions that failed to appreciate the complex political realities of the country.

While these interventions espoused community control, they were in fact the result of a top-down approach reminiscent of earlier state-building efforts, neglected the importance of customary authority structures, and treated Afghanistan as a blank slate for experimentation. Such community-based programmes also lead to a myopic focus on quantifiable metrics at the expense of political considerations, reflecting a broader methodological bias in development studies exemplified by the popularity of randomized controlled trials.

Drawing on extensive field observations, the paper contends that such programmes worked against the grain of Afghan society and paradoxically undermined the legitimacy and effectiveness of local governance institutions.

The paper concludes by advocating a more holistic, adaptive, and politically informed approach to grassroots development in conflict-affected regions—one that empowers authentic local ownership, aligns with endogenous social and political structures, and grapples with the messy realities of fostering legitimate governance in fragile contexts.

The hard-earned lessons from Afghanistan’s recent past should prompt a fundamental rethinking of what constitutes effective development assistance in post-conflict environments.