Organizational Rituals of Risk and Error

In this chapter, I compare organizational rituals of risk and error in two U.S. government agencies: The Federal Aviation Administration's National Air Traffic System (NATS) and the National Aeronautics and Space Administration (NASA). I take the position that all organizations, even those categorized as High Reliability Organizations, are subject to routine nonconformity: they regularly produce mistakes and errors. For both agencies, I describe their technologies of control, which are the rules and procedures, work practices, and surveillance technologies for regulating risk. Then I examine definitional processes: how anomalies are identified, tracked, and converted into formal organizational categories. Because of differences in the certainty of space shuttles and airplanes, the technologies of control employed as these two agencies identify, define, and control risk and error vary. This variation has important consequences. I show that the variation in these agencies' technologies of control produces different cultural understandings about risk and error. These cultural understandings have social psychological consequences, affecting how technical workers interpret signals of potential danger. Finally, I consider the implications for this comparison for organizational encounters with risk. Organizational encounters with risk and error are not restricted to the sensational cases that draw media coverage when mistakes, near-misses, and accidents become public. They are, instead, a routine and systematic part of daily organizational life that only occasionally become visible to outsiders. Merton was the first to observe that any system of action can generate unexpected consequences that are in contradiction to its goals and objectives (1936, 1940, 1968a). Recent research affirms his observation: unanticipated events that deviate from organizational expectations are so typical that they are "routine non-conformity" a regular by-product of the characteristics of the system itself (Vaughan 1999). The public only learns about the most egregious of these. Because routine non-conformity is a regular system consequence, complex organizations that use or produce risky technologies may have encounters with risk daily. In this paper, I compare daily encounters with risk for two organizations for which mistakes result in public failures and have high costs: the Federal Aviation Administration's National Air Transportation System (NATS) and the National Aeronautics and Space Administration's (NASA) National Space Transportation System (NSTS), otherwise known as the Space Shuttle Program. My comparison of these two agencies is grounded in two related strands of research. Barry Turner investigated the causes of 85 different "man-made disasters." Turner found an alarming pattern: after a disaster, investigators typically found a history of early warning signs that were misinterpreted or ignored. A problem that seemed well-structured in retrospect was ill-structured at the time decisions were being made (Turner 1978; Turner and Pidgeon, 1997). Turner did not have micro-level decision making data to explain how and why this could happen. Instead, he explained it skillfully by drawing on organization theories about information flows. However, no one had tested this theories with data about how people decisions and why they make the choices they did. The second strand is Star and Gerson's (1986) study of anomalies in scientific work. Star and Gerson point out that every anomaly has a trajectory during which it is subject to processes of definition, negotiation, and control that are similar to the response to anomalies in other types of work. They found that how anomalies are defined and negotiated depends upon the occupational context and the evaluation systems that have been developed to meet unexpected deviations in the work flow. Star and Gerson concluded that a mistake or anomaly is never defined in isolation, but always is relative to the local and institutional context of work (1986: 148-50). These two strands of research led me to the following conclusions. First, Turner's discovery of warnings of hazards over long incubation periods preceding accidents suggested the importance of research on daily encounters with risk. Second, Star and Gerson's work indicated that an appropriate focus was the trajectory of anomalies how they are identified, measured, and assessed for risk. I had a chance to investigate both these issues in my research on NASA’s Challenger accident (Vaughan 1996). In the years preceding the accident, I found a history of early warning signs signals of potential danger about the technology of the flawed Solid Rocket Boosters that caused the disaster. These early warning signs were misinterpreted as decisions were being made, contributing finally to the disastrous outcome. Their seriousness only became clear in retrospect, after the tragedy. In contrast to the accidents Turner investigated, I had micro-level data on NASA