Constructing Risk and Safety in Technological Practice

Boel Berner, Jane Summerton

• Publisher: Routledge
• Year: 2003
• Print ISBN: 9780415285711
• eBook ISBN: 9781134450411
• Topic: Technology & Engineering
• Subtopic: Sociology
• Index: Technology & Engineering

Modern technological systems entail risks and uncertainties of hitherto unknown dimensions. This book discusses the construction of risk and safety within a variety of empirical contexts where technologies and their risk are debated and handled by individuals, groups or organizations. With contributions from leading scholars from Europe and the USA, it presents original theoretical discussions, linked to detailed empirical case studies.

Part I

Interpreting accidents

Introductory comments

The chapters in Part I center on problematic understandings of the accidental. Judith Green discusses how technologies and knowledge systems have constructed “the accident” as a particular category of misfortune in late modern society. Jörg Potthast and Sidney W.A. Dekker focus in different ways on the discursive constructions of what caused specific accidents in sociotechnical systems (specifically air transport systems). The studies point to the various strategies by which actors attempt to make sense of injuries and catastrophic events, continually striving to control “the accidental.”

In an important recent book, Green (1997) traces the development of European thought in relation to the accident. She identifies three dominant discourses that could be crudely summarized as “fate,” “determinism” and “risk.” Before the second half of the seventeenth century, events in the course of one’s life were viewed as part of one’s personal destiny. Accidents as something that “just happened” had no place in life since all events had to fit into a pattern. By the end of the seventeenth century, this view of fate and destiny had been replaced by a modernist discourse of determinism. Accidents were now a residual category of misfortune that marked the limits of a rational universe. Given the vagaries of “nature,” accidents were inevitable and thus to be expected from time to time – yet not worthy of serious investigation. This consensus was fractured in the second half of the twentieth century. In today’s “risk society,” a discourse of risk management and prevention has emerged, marking the transfer of “the accidental” from the margin to the center of concern. If misfortunes such as accidents occur as a result of knowable and calculable risks, they can be transformed into potentially preventable events. All accidents thus become comprehensible in terms of “risk factors,” and “the accidental” becomes something to avoid.

These themes are further developed in Green’s chapter in this part. She focuses on the science of epidemiology as the dominant framework for generating knowledge about accidents since the mid-twentieth century. Epidemiology is viewed as a technology for understanding accidents as the outcome of mismanaged risks. Accidents have been reconfigured as not only potentially preventable but as events that ought to be prevented. The technologies of epidemiology and the artifacts that they make possible (such as safety devices for reducing risks) have, she argues, given rise to an individualization of accident management. Everyone is responsible for making rational decisions about avoiding and managing risks, using the everyday risk technologies that are available. Individuals are accountable for their personal misfortunes. There is a gap, however, between the myth of preventability and the occurrence of what should have been prevented. The risks are separated from the uncertain and fluid life-world in which they are experienced; they are fragmented and reified as specific “risks” to be calculated, catalogued and monitored. The ways in which risks and the accidental are classified and handled are thus seen as processes of both construction and control. This focus links Green’s analysis to work on the disciplinary nature of knowledge and expertise in the Foucaldian tradition (see, for example, Burchell et al. 1991; Lupton 1999). Yet Green does not only discuss the proliferation of safety technologies but also the resistances they engender, based on the individual’s chaotic experience of everyday life.

Until the second half of the twentieth century, there was little difference between the “expert” view of accident causation and that attributed to the public, which Green points out. Since then, professionals have tried to move away from a “popular” understanding of accidents toward a uniquely “professional” understanding based on what is viewed as a rational approach to risks and their management. The “irrationality” of the popular understandings was then – as was discussed in the introductory chapter to this volume – to be studied in psychological and psychometric work. Jörg Potthast gives an interesting perspective on these emerging tensions between expert and lay opinions in his study of different interpretations of the causes of the Swissair crash near Halifax, Canada, in 1998. He analyzes a large number of contributions to a discussion site on the Internet in the aftermath of the crash. Potthast’s analysis shows that the boundaries between expert and lay opinions are blurry at best, despite the attempts of some contributors to invoke an authoritative voice.

Potthast’s analysis shows the ways in which reconstructions of accidents by both expert and lay actors are often characterized by a high degree of uncertainty. Actors with differing competences and orientations struggle to reconstruct and understand a sequence of events in which the outcome may be clear but the process leading to this outcome is largely unknown. As various explanations of the accident are contested or reaffirmed, fragments of a negotiated understanding emerge among the contributors to the website. Potthast views the sense-making that is involved in these negotiations as reflecting three distinct “narratives of trust” that show how the actors work to reconstruct their trust in the artifacts, the operators and other experts and the safe operation of the system.

The chapter by Sidney W.A. Dekker focuses on another theme from Green’s analysis, that is, the tendency for expert accounts to be individualized and reified. Dekker discusses the accounts given by official investigating commissions in the wake of another, less publicized airline accident, namely a 1995 crash in Cali, Columbia, that resulted in the deaths of all on board the flight. Dekker’s story provides a detailed and incisive account of the tendency of investigators to confuse their own reality with the crew’s situation at the time of the accident. In these official accounts, the crew’s actions are removed from the context in which they are embedded and are ascribed with meaning in relation to after-the-fact-worlds. Such accounts exhibit all the errors of hindsight: through the matching of available and isolated fragments of behavior, an artificial story that is divorced from the situated reality of the actors at the time is successively constructed. Dekker convincingly argues for a different approach to accident investigations which would take into account the explicitly situated character of cognition and decision making. Dekker’s analysis thus aligns with the work of Karl Weick, Gene Rochlin and others whose studies of so-called High Reliability Organizations explore the contextual rationality and sense making of actors in situations of stress and uncertainty (Rochlin 1991; Weick 1993; see also the introductory chapter to this volume).

References

Burchell, G., Gordon, C. and Miller, P. (eds) (1991) The Foucault Effect: Studies in Governmentality, London: Harvester Wheatsheaf.

Green, J. (1997) Risk and Misfortune: The Social Construction of Accidents, London: UCL Press.

Lupton, D. (ed.) (1999) Risk and Sociocultural Theory: New Directions and Perspectives, Cambridge: Cambridge University Press.

Rochlin, G.I. (1991) “Iran Air Flight 655: Complex, Large-scale Military Systems and the Failure of Control,” in T.R. La Porte (ed.) Social Responses to Large Technical Systems: Control or Anticipation, Amsterdam: Kluwer.

Weick, K.E. (1993) “The Collapse of Sensemaking in Organizations: The Mann Gulch Disaster,” Administrative Science Quarterly, 38: 628–652.

1 The ultimate challenge for risk technologies

Controlling the accidental

Judith Green

Introduction

This chapter addresses technologies of risk in one arena, that of accidental injury. “Technologies of risk” refers here to artifacts and knowledge systems that are utilized for the management and reduction of risk. Accidents present perhaps the ultimate challenge for risk technologies – how to control apparently unpredictable events. However, risk technologies do not merely act upon a natural category of events called “accidents,” they also simultaneously shape the classification of certain events as “accidents” and shape our understanding of and reactions to those events. This chapter explores this simultaneous process of control and construction through an analysis of how two interrelated sets of technologies have constructed “the accident” as a particular category of misfortune in late modern society.

The first technologies of interest are those mundane and unremarkable artifacts that are designed to reduce the risk of an accident happening or reduce the risk of injury if an accident should occur. These include such innovations as child-resistant medicine containers, bicycle helmets and air-bags in cars. These particular examples are now commonplace and accepted as sensible ways of increasing safety – no rational adult would resist their use. However, the introduction of such risk technologies is not inevitable and straightforward, and the example referred to later in this chapter (namely “soft hip protectors” for elderly citizens) is of a more recent and less widely accepted artifact.

Second, are those technologies of knowledge that make possible risk reducing artifacts. From the second half of the twentieth century, legitimate knowledge about accidental injury has been primarily derived from statistical analysis of the distribution of accidents and their risk factors within populations. As accidental injury has become “medicalized,” epidemiology, the study of diseases and their distribution throughout populations, has become a dominant framework for generating knowledge about accidents. Given that the term “accident,” as has already been indicated, is taken here to describe a socially constituted category of events, this chapter will start not by trying to define the term, but by delineating the historical space within which accidents have happened. The question is not so much, “what is an accidental injury?” but rather “when did it become possible to speak of accidental injuries?” After outlining the historical emergence of accidents, this chapter describes the development of epidemiology as a technology for understanding accidents as the outcome of mismanaged risks. Two implications of an epidemiological understanding are then identified: the individualization of risk management and the separation and reification of risks. Finally, this chapter explores the possibilities for resisting risk technologies.

Historical context

As a category of misfortune, “accidents” have a relatively brief history (Green 1997a). In Europe, they are essentially products of the seventeenth century, when emerging rationalist and probabilistic ways of thinking (Hacking 1975; Porter 1986) created a space for those misfortunes which were, at least at the local level, inexplicable. Before then it could be argued that Western cosmologies had no need for a category of accidental events. Fate, destiny and a belief in predetermination meant all happenings could have meaning. With religion and superstition declining in legitimacy as answers to the inevitable questions of “why me, and why now?” when misfortune strikes, the accident became first a possible, then a necessary, category. By the beginning of the twentieth century, a belief that some events happen “by accident,” that they have no purpose and cannot be predicted at the local level, had become a defining belief of modernity.

That accidents came to signal a uniquely “modern” system of beliefs is evident in the writings of both European anthropologists, anxious to distance the superstitious beliefs of “primitives” from their own, and in those of child psychologists newly interested in the development of “adult” mentality. Anthropologists such as Evans-Pritchard (1937) pointed to the cosmologies of small-scale societies such as the Azande in central Africa. For the Azande, argued Evans-Pritchard, all misfortune had meaning and could be traced to witchcraft or the breaking of taboos. The specific cause of any misfortune (that is, why it happened to this person, at this time) could be divined. For Evans-Pritchard and others looking at the apparently irrational ideas of non-Western peoples, the belief that the local distribution of misfortune was random, unpredictable and inexplicable was what distinguished the modern rational mind from that of the “primitive” who persisted with such supernatural explanations.

Similarly, for child psychologists such as Piaget in the 1930s, children’s failure to understand chance and coincidence were evidence of their immature minds (Piaget 1932). Moral maturity brought with it the ability to conceptualize the accident as a blameless event. In the early twentieth century, then, a belief that accidents “just happen,” and that there is therefore no profit in examining why they happen to particular people at particular times, was a legitimate and rational belief. The accident was merely a coincidence in space and time with neither human nor divine motivation. Only those yet to develop a modernist rationality (primitives, children) would see an accident as anything other than a random misfortune.

More recently the place of accidents as a category of motiveless and unpredictable misfortunes has shifted. From a marginal category of events not worthy of serious study, they have moved center stage. Within the risk society, accidents are pivotal. First, management of the accidental poses the ultimate challenge for risk technologies: to predict the unpredictable and then control the uncontrollable. Second, that accidents do happen demonstrates the failings of risk management to date and thus justifies even greater surveillance of risks. Despite ever more sophisticated safety technologies, trains occasionally crash and children frequently fall from bicycles. Major tragedies and minor everyday mishaps are subjected to formal or informal enquiries to establish “what went wrong” and to apportion blame if appropriate. Such enquiries apparently erase the notion of the accidental, for in retrospect all misfortunes are rendered comprehensible in terms of the risk factors that produced them. They were therefore not accidents at all, but rather failures of risk management. Like the Azande in Evans-Pritchard’s (1937) account, we can no longer, it seems, countenance the idea of a meaningless event, a misfortune “caused” merely by bad luck or coincidence (Douglas and Wildawsky 1982). However, risk technologies of course have not erased the accidental at all. It survives and even flourishes but has been shaped by the very technologies that seek to manage it. To illustrate the ways in which risk technologies, and epidemiological technologies in particular, contribute to particular understandings of the accident, this chapter will now outline the growth of epidemiological technologies of knowledge.

The growth of epidemiology

As a professional activity, accident prevention has an even briefer history than that of the accident. From a vantage point of the early twe...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Contents
• Contributors
• Acknowledgments
• Constructing risk and safety in technological practice: an introduction
• PART I Interpreting accidents
• PART II Defining risks
• PART III Constructing safety
• Index