Organizational Accidents and Deadly Police-Involved Violence: Some Thoughts on Extending Theory, Expanding Research, and Improving Police Practice

• Published date: 01 January 2020
• Date: 01 January 2020
• DOI: 10.1177/0002716219892913
• Author: David Klinger
• Subject Matter: Patterns of Fatal Police Shootings

28 ANNALS, AAPSS, 687, January 2020

DOI: 10.1177/0002716219892913

Organizational

Accidents and

Deadly Police-

Involved

Violence: Some

Thoughts on

Extending

Theory,

Expanding

Research, and

Improving

Police Practice

By

DAVID KLINGER

892913ANN THE ANNALS OF THE AMERICAN ACADEMYORGANIZATIONAL ACCIDENTS AND DEADLY POLICE-INVOLVED VIOLENCE

research-article2019

In recent years, violence by and against the police has

been examined from the perspective of organizational

accident theory. This article extends that work by

reviewing some key ideas, identifying some limitations

of organizational accident theory for understanding

police-involved violence, and detailing some specific

research topics for future empirical exploration. It con-

cludes by offering some specific policy and practice

recommendations to reduce police-involved violence.

Keywords: deadly force; police violence; normal acci-

dents; high-reliability organizations; crimi-

nal justice policy; tactical reviews

The production of this volume indicates that

the social scientific community may be

ready to take seriously Charles Perrow’s (1984)

work in Normal Accidents (and the larger body

of academic literature in which it sits) in rela-

tion to deadly violence involving the police

(Klinger 2005; Sherman 2018). There are sound

empirical and theoretical reasons to believe

that viewing deadly police-involved violence

through the lens provided by Perrow’s normal

accident paradigm (and related works) holds

great promise to accomplish many positive

ends. These include (1) enhancing understand-

ing of why and how violent interactions between

police officers and citizens come to pass and

how and why they play out, (2) reducing the

number of police-citizen interactions that

include the application of deadly force by

police officers, and (3) reducing the number of

serious injuries and deaths sustained by police

officers.

David Klinger is a professor of criminology and crimi-

nal justice at the University of Missouri–St. Louis. Prior

to pursuing an academic career, he worked as a patrol

officer for the Los Angeles and Redmond (WA) Police

Departments.

Correspondence: klingerd@umsl.edu

ORGANIZATIONAL ACCIDENTS AND DEADLY POLICE-INVOLVED VIOLENCE 29

The purposes of this article are to (1) show how looking at deadly police-

involved violence through the lens provided by the normal accident paradigm

and related social scientific work might accomplish these three goals, (2) identify

some limitations of the normal accident and related intellectual traditions for

accomplishing these goals, and (3) identify several places where work in these

traditions should be extended to better understand and take action on deadly

police-involved violence. The work begins with a brief overview of Perrow’s

(1984) normal accident framework and how this author (Klinger 2005) first

applied it to the world of deadly police-involved violence.

Normal Accidents and Deadly Police-Involved Violence

Perrow (1984) developed the core of his normal accident thesis while serving as

a consultant to the Presidential Commission investigating the near-catastrophic

1979 accident at the Three Mile Island nuclear power plant near Harrisburg,

Pennsylvania. In seeking to understand how this accident came to pass, Perrow

identified a pair of intertwined culprits: the incredible complexity of the techno-

logical system that generated power via nuclear fission in the reactor that failed

and the high degree of coupling between the complex, critical parts of this sys-

tem. He came to believe that the accident in question occurred because the

supremely complex and tightly coupled power generation system in place permit-

ted an unimaginable series of technological and human failures to cascade out of

control, dooming the highly engineered reactor. From this starting point, he

developed a general theory of how and why bad things happen in high-tech sys-

tems. He then illustrated its utility by applying it to account for disasters that had

occurred in a broad array of fields beyond nuclear power generation, such as

marine transport and commercial aviation.

Elegant in its simplicity, Perrow’s theory holds that bad outcomes can be

expected to occur in high-tech systems that involve a high degree of complexity

and that are tightly coupled. The more complex the system, the greater the

degree of interaction between its elements and the more likely it is that some-

thing will go wrong. A problem in one aspect of a system will lead to additional

problems in other system parts. Moreover, as systems become more complex, it

becomes less likely that humans can quickly apprehend what is going wrong (or

has already gone wrong) when something goes awry, which means that things can

cascade out of control before the humans “in charge” can react to contain an

initial problem. When the elements of a system are closely linked (coupled), only

limited buffers exist between them. This means that failure can quickly spread as

a problem in one element of the system propagates to other elements and a full-

fledged disaster ensues. In asserting that bad outcomes are thus baked into highly

complex systems that are tightly coupled, Perrow argued that they are a regular

aspect of such systems. And so he dubbed the bad outcomes that he expects to

sometimes occur in high-tech systems that are highly complex and tightly cou-

pled, “normal accidents.”