From pre-emption to slowness: Assessing the contrasting temporalities of data-driven predictive policing

Abstract

Debates on the temporal shift associated with digitalization often stress notions of speed and acceleration. With the advent of big data and predictive analytics, the time-compressing features of digitalization are compounded within a distinct operative logic: that of pre-emption. The temporality of pre-emption attempts to project the past into a simulated future that can be acted upon in the present; a temporality of pure imminence. Yet, inherently paradoxical, pre-emption is marked by myriads of contrasts and frictions as it is caught between the supposedly all-encompassing knowledge of the data-processing ‘Machine’, and the daily reality of decision-making practices by relevant social actors. In this article, we explore the contrasting temporalities of automated data processing and predictive analytics, using policing as an illustrative example. Drawing on insights from two cases of predictive policing systems that have been implemented among UK police forces, we highlight the prevalence of counter-temporalities as predictive analytics is situated in institutional contexts and consider the conditions of possibility for agency and deliberation. Analysing these temporal tensions in relation to ‘slowness’ as a mode of resistance, the contextual examination of predictive policing advanced in the article provides a contribution to the formation of a deeper awareness of the politics of time in automated data processing; one that may serve to counter the imperative of pre-emption that, taken to the limit, seeks to foreclose the time for politics, action and life.