Reducing Fatal Police Shootings as System Crashes: Research, Theory, and Practice

Abstract

Can fatal shootings by American police be reduced? If so, what theoretical framework would be most useful in saving more lives? What research agenda would that framework suggest? The purpose of this review is to answer those three questions. It applies the system-accident framework ( Perrow 1984 ) as a pathway to help police agencies reduce fatal police shootings, adapting it as system-crash prevention to encompass a wider range of systemic causes of catastrophic events. In contrast to deterrence, the dominant policy perspective on reducing fatal shootings, a system-crash prevention approach applies lateral thinking ( Johnson 2010 ) from lessons learned about airplane crashes, surgical errors, nuclear power plant meltdowns, and other rare events in complex systems. This framework spotlights the rare combinations of risk factors and errors that can produce fatal shootings, the prevention of which may need to vary widely between large and small communities. Of the 986 fatal police shootings reported nationally in 2015 ( Wash. Post 2016 ), an estimated half occurred in cities with fewer than 50,000 residents; only a third occurred in cities over 250,000 residents ( Sherman 2015 ), where the majority of all research on police shootings has been done. The system designs for fewer police killings in very large versus very small police agencies require a general framework for policy-relevant criminology. The framework suggested by Sampson et al. (2013) requires research to identify (a) causal mechanisms that may generally help to reduce fatal police shootings but may also have (b) heterogeneous effects that may work differently on different subpopulations or when their application is (c) contextualized in different social settings. Future research must therefore include very small communities in order to understand and help prevent the majority of all fatal police shootings.