Technology, Transparency, and Policing

Description

Free and open to all but Online Registration is required.

Barak Ariel (Cambridge)
Adam Edwards (University of Cardiff)

 

Abstracts

Barak Ariel’s co-authored paper is titled, ‘“Contagious Accountability: A Global Multisite Randomized Controlled Trial onthe Effect of Police Body-Worn Cameras on Citizens’ Complaints Against the Police.’ The use of body-worn cameras (BWCs) by the police is rising. One proposed effect of BWCs is reducing complaints against police, which assumes that BWCs reduce officer noncompliance with procedures, improve suspects’ demeanor, or both, leading to fewer complaints. We report results from a global, multisite randomized controlled trial on whether BWC use reduces citizens’ complaints. Seven discrete tests (N = 1,847 officers), with police shifts as the unit of analysis (N = 4,264), were randomly assigned into treatment and control conditions. Using a prospective meta-analytic approach, we found a 93% before–after reduction in complaint incidence (Z = −3.234; p < .001), but no significant differences between trial arms in the studies (d = .053, SE = .11; 95% confidence interval [CI] = [−.163, .269]), and little between-site variation (Q = 4.905; p = .428). We discuss these results in terms of an “observer effect” that influences both officers’ and citizens’ behavior and assess what we interpret as treatment diffusion between experimental and control conditions within the framework of “contagious accountability.”
 

Adam Edwards’ paper is titled 'Big Data, Predictive Machines and Security: The Minority Report’ (please contact convenors for a copy).  Emergent technologies, including machine-learning predicated on ‘Big Data’, are promoted by enthusiasts of ‘smart policing’ as a means of better anticipating problems of crime and security and targeting preventive interventions. Critics dismiss these claims on libertarian grounds and following revelations from WikiLeaks and the US security contractor, Edward Snowden, about the abuses of mass surveillance technologies and the need for more responsible human oversight and scrutiny. Other critics question the possibility of smart policing given the renowned limitations of Artificial Intelligence for acknowledging and responding to the improvised qualities of social relations. The presentation provokes a discussion about the scripted qualities of social relations involved in policing, which may accommodate improvisation but are nonetheless sufficiently routinized to be captured by predictive machines. In turn, the presentation provokes further discussion about the unintended consequences of predictive machines, including their role in a new ‘arms race’ between organisers and preventers of serious crimes who can use emergent technologies to outflank one another. In these terms, responsible research and innovation in emergent technologies requires a sociological imagination.
 

 

Please join the mailing list here to receive further information on joining the reading group as well as on the lecture schedule.
 

Part of The Politics and Paradoxes of Transparency, Research Group series
Administrative assistance: gradfac@crassh.cam.ac.uk