Send message to Authors

Paper short abstract

This paper examines how predictive technologies and preparedness exercises can inadvertently close down rather than enhance anticipatory capacity in disaster governance. Using Japan’s nuclear preparedness as a case, it distinguishes prediction from anticipation and highlights institutional mindsets.

Paper long abstract

Anticipatory governance is often associated with predictive technologies, simulations, and preparedness exercises designed to foresee and manage future crises. However, anticipation is not equivalent to prediction. While prediction estimates future events, anticipation translates uncertain futures into strategies, adaptive responses, and learning under uncertainty.

This paper examines how predictive technologies can paradoxically undermine anticipatory capacities, using the case of nuclear emergency preparedness in Japan before and after the Fukushima Daiichi accident. Building on previous research by the author on predictive technologies in nuclear disaster governance, the paper revisits the role of the System for Prediction of Environmental Emergency Dose Information (SPEEDI). Before 2011, Japan’s preparedness regime relied heavily on SPEEDI, a dispersion modelling tool for evacuation decisions. In practice, SPEEDI became embedded in scripted disaster drills aimed at demonstrating that responses could be executed under predefined plans. Prediction thus functioned less as a resource for exploring uncertainty than as a device stabilizing imaginaries of controllable disasters.

Following Fukushima, policy shifted in the opposite direction: predictive modelling was abandoned in favour of preset evacuation zones and criteria-based decision rules. While intended to ensure robustness under uncertainty, this approach limited incorporation of emerging knowledge during unfolding crises.

Drawing on literature on anticipatory governance and institutional mindsets, the paper argues that the anticipatory value of predictive technologies depends less on the technologies themselves than on the institutional contexts shaping their use. When embedded in closed preparedness regimes, prediction may stabilize imaginaries of controllable disasters rather than support anticipatory learning under uncertainty.