Disasters create opportunities for active learning, why do they repeat?

Introduction

Disasters have adversely affected humans since the dawn of their existence (Coppola, 2011: 1). If one looks at the past as a model for what may be expected in the future, it can be found that severe events, which wreaked havoc on human communities and inflicted high levels of mortality, were surprisingly frequent and widespread (Torrence and Grattan, 2002: 1). Considering the recent past, the number of disasters increased threefold within the 1980’s in comparison to the 1960’s (United Nations Environment Programme, 2002: 271), and the economic loss increased by a factor of almost nine (Munich Re, 1999: 40). From the 1980's to the 1990's the number of affected people rose from 147 million to 211 million a year (United Nations Environment Programme, 2002: xxv). Recognizing the rise in number and effects of disasters, this essay discusses the potential for active learning that disasters create and the reasons why those repeat.

This essay begins by defining the key terms 'disaster' and ‘active learning’. To identify pathologies that affect the ways organizations learn from disasters, factors that determine the potential for active learning are then defined and discussed. These factors are integral parts of Toft and Reynolds’s “Steps to active foresight model” (2005: 67), which illustrates how they are interlinked. Toft and Reynolds’s model is then used as a framework to study two separate nuclear power plant disasters. These high profile events, which negatively affected the nuclear industry, are examined to identify, whether the potential for active learning was realized after the first event, and if yes, why an event displaying similar features yet again occurred. The findings deriving from the case studies are then summarized and the essay question is answered and concluded with, why disasters repeat despite they create opportunities for active learning. Before any analysis or conclusions can be made and in order to establish a research framework, key terminology and concepts need to be examined, beginning with the key term 'disaster'.

There is no commonly agreed definition of the term disaster. Historically, the term disaster derives from the Latin roots dis- and astro, meaning “away from the stars”, in other words, an event to be blamed on an unfortunate astrological configuration (Coppola, 2011: 29). Organizational definitions of disasters tend to depend on the organizational focus (Stallings, 2005: 131). The definition provided by the United Nations International Strategy for Disaster Risk Reduction (UNISDR, 2009), which aims to promote common understanding of terms, focuses on societal aspects of disasters. The definition promoted by the World Health Organization mostly focuses on the impact on health (WHO, 1995), whereas the definition of the International Criminal Police Organization focuses on the number of casualties (INTERPOL, 2009). One common aspect different organizational definitions coincide on is that a disaster event threatens social stability and requires “extraordinary or emergency countermeasures to reestablish stability” (Stallings, 2005: 130). The different focus of each organizational definition shows that the term disaster is only of ephemeral importance, acting as a stimulus to an organization specific reaction (Dombrowsky, 2005: 15). Considering the characteristics of both nuclear power plant disasters studied in the essay, the definition offered by the U.S. Federal Emergency Management Agency (FEMA) is adopted, which defines disaster as "an occurrence of a natural catastrophe, technological accident, or human-caused event that has resulted in severe property damage, deaths, and/or multiple injuries" (FEMA, 1996).

The essay continues with defining active learning and other key terms that affect the potential for active learning (Toft and Reynolds, 2005: 66-86). In general, learning is defined as the transfer of knowledge (O’Neil et al., 2009: 28). Toft and Reynolds (2005: 66) suggest that there are two main types of learning, ‘active’ and ‘passive’. Active learning denotes the process, during which remedial action is taken to rectify recognized deficiencies, whereas passive learning is limited to the acknowledgement of a deficiency without taking action. Active learning is considered of higher value as “there is little point in knowing, how to prevent a disaster, if no active steps are taken to prevent it” (Toft and Reynolds, 2005: 66). However, before active learning can take place, the potential for active learning must be recognized, which depends on a number of factors explained below.

Because all disasters are unique and typically low frequency events it is unlikely that any organization would be able to predict events based on its own history. But, when viewed in the context of a whole industry, a number of similar features can be observed, recognition and analysis of which is termed as isomorphic learning (Borodzicz, 2005: 26).

Relevant authors (Borodzicz, 2005: 46; Toft and Reynolds, 2005: 69; Waring and Glendon, 1998: 84) point out that organizations, using similar socio-technical systems, can benefit from hindsight that derives from isomorphic learning. However, analysis based solely on hindsight needs to be treated with caution, as disaster events rarely manifest themselves in the precise same way again (Reason, 1990: 174). Kletz's (2003) examinations of industrial accidents illustrate that there is no shortage of tragic events with similar features. The limitations of hindsight are also highlighted by Toft and Turner (2006: 191), who emphasize that learning from the past can only occur if similarities between the past and the present situation are recognized and connections are made with the foreseeable future.

Borodzicz (2005: 46) underlined that foresight, deriving from hindsight, can prevent disasters of similar nature from re-occurring and improve response to disasters. This learning is often prevented by the perceptions of organizations, which consider themselves more resilient than they are in reality (Scanlon, 2011: 16; Toft and Reynolds, 2005: 135). The development of foresight in the context of disasters is explained by Turner and Pidgeon (1997: xii), who define it as the recognition of preconditions during the incubation period. However, unless action is taken to mitigate disasters, the acquired foresight remains passive. Toft and Reynolds (2005: 69-89) differentiate between active foresight and passive foresight, highlighting that active foresight derives from the realization of the potential for active learning. Active foresight combines foresight of the possible causes of disaster with action to reduce risk of those taking effect (Toft and Reynolds, 2005: 65). Theoretically, if it is possible to create systems, it should also be possible to predict their failures. However, one of the central problems of attempting to use foresight is that one can never have perfect knowledge about all variables, which may appear and the complexity of modern systems (Borodzicz, 2005: 25). Therefore, in hindsight people tend to exaggerate what could have been anticipated in foresight due to the emotional impact and willingness of an organization to change, depending upon the real and relative distance to an event (Toft and Reynolds, 2005: 70-76).

The relative distance to an event is explained by Toft and Reynolds, who distinguish between different types of organizations to show their levels of interconnectivity (Toft and Reynolds, 2005: 69-70). The closer, the real and relative distance of an organization to a disaster, the bigger the impact on the organizational safety philosophy; whereas, the further, the real and relative distance to a disaster, the lower the impact on safety philosophy and the organizational reaction (Toft and Reynolds, 2005:109-110). Hence, the distance to an event impacts the organizational safety philosophy that defines the fundamental approach to safety (Taylor, 2010: 25), and the safety culture that denotes the set of norms and practices within an organization being concerned with minimizing the exposure of individuals to dangerous conditions (Toft and Reynolds, 2005: 26). From a safety culture viewpoint, the failure to perceive threats to a system's stability represents a poor safety culture (Borodzicz, 2005: 154).

Organizations dealing routinely with high risk technologies, e.g. nuclear power plant operators, are expected and regulated to be in possession of a sophisticated safety philosophy and, as a consequence, a safety culture from the outset. Due to the lives involved and immense costs, a trial and error approach is imprudent in these types of organizations. There are examples of failure in organizations dealing with high-risk technologies, e.g. the National Aeronautics and Space Administration (NASA). During the analysis of events of the space shuttle Challenger disaster in 1986, Torrance (2001: 1) described NASA as an organization too young to understand the factors leading to the Challenger disaster. This confirms Toft and Reynolds’s view (2005: 26, 77) that changes in safety culture do not appear overnight and it takes time to develop a system of commonly shared values and lessons. However, considering the various layers of culture, an issue is to assess the quality and effectiveness of safety culture (Borodzicz, 2005: 41).

An additional factor influencing the safety philosophy of organizations is the so-called ‘organizational perception of chance and disaster’. Although it should be an obvious goal of organizations to reduce the possibilities for a disaster, some apparently do not. Factors that influence this organizational behavior can be manifold and include economic and political factors. Additionally, psychological factors determine the willingness for change of safety philosophy. Where an organization does not believe that it was responsible for a disaster, the willingness to change its safety philosophy is limited (Toft and Reynolds, 2005: 77). However, organizational perceptions of disasters often appear less influenced by history than by frames applied to history, as organizations often attribute desirable outcomes to internal factors, whereas external factors are more often blamed for failures (Toft and Reynolds, 2005: 77). Thus, the emotional impact of an event, combined with the ‘organizational perception of chance and disaster’, determines the organization's willingness to change its safety philosophy and its reaction to a disaster. Ideally, the organizational reaction then gives way for passive foresight.

Passive foresight, in combination with safety by compulsion, gives scope and potential for active learning to take place (Toft and Reynolds, 2005: 69). Management decrees and legislation may present an opportunity to change individual behavior in a particular situation; however, compulsion alone cannot change the safety philosophy of an organization (Toft and Reynolds, 2005: 85). Hence, if it cannot be ensured that everybody in an organization attains to rules, legislation can create a false sense of safety. The Chernobyl nuclear power plant disaster is an example, where detailed regulations on operations existed, but were disregarded by personnel (Kreps, 2005: 38). That said, disasters do not necessarily happen due to rule breaking, but sometimes because existing rules did not cover the conditions that existed prior to the disaster (Toft and Reynolds, 2005: 101). The discussion shows that safety by compulsion has its limitations; however, it is clearly desirable that organizations, dealing with high-risk technologies, should be regulated by legislation and rules for a structured approach to safety and ensure high standards of operation (Borodzicz, 2005: 58). Therefore, Toft and Reynolds (2005: 83) recommend an approach that also considers people’s skills, experience and intelligence to solve organizational problems.

[...]