Translations and translation gaps: the gunshot acoustic surveillance experiment in Brazil
Leonardo Cardoso · Sound Studies · 2019
This article examines crime prevention as it relates to recent developments in audio monitoring for gun crime control. More specifically, I discuss the adoption of ShotSpotter, a device that detects and locates impulsive sounds. ShotSpotter can alert the police promptly whenever gunshots are fired, providing the location, number and exact time of the rounds fired, the number of shooters and even the shooters’ direction of motion. To convert sound monitoring into crime control, the technology needs to perform a series of delicate translations (a term I borrow from actor-network theory) between a range of actors. The actors include concealed sensors distributed across urban space, software for filtering and analysing sounds, and experts working 24/7 to analyse and classify data. In the early 2010s, ShotSpotter was installed in two urban areas in Brazil with high rates of gun violence: Canoas (in southern Brazil) and Rio de Janeiro. Public officials and private companies involved with the project stated that the technology would revolutionise crime control in the country. However, I argue that several local translation gaps ended up affecting ShotSpotter’s performance.