Safety Culture?
Safety culture as a concept has gained importance in high-risk systems over the years to address safety issues from an organizational perspective.
Jean-Christophe Le Coze identifies two waves of safety culture studies:
From the late 1980s to the mid-2000s, there was a distinction made between a managerial ambition to shape and control safety culture and a more neutral ambition to understand it as a social construct.
From the mid-2000s to the present, a more nuanced and diverse range of appreciations have emerged, ranging from highly critical to highly favorable perspectives on safety culture. For instance, safety culture's association with culture should be rejected, and it’s more appropriate to focus on organizational culture rather than safety culture, which would introduce too many biases. Moreover, safety culture’s dominant rationale as expressed in companies can be seen as an expression of responsabilization, which reproduces individualist and reductionist explanations that are unable to reliably explain social or system performance.
To understand these different critical, neutral, and enthusiastic views on safety culture, one needs to appreciate the broader context in which they are produced. The safety field is socially structured, with competing interests about the definition of safety culture and safety topics more generally, which can become safety products as part of a market. The critical or radical views of safety culture reflect what they consider to be a safety culture market dominated by a thriving consulting industry, which primarily sells products for business purposes without always promoting an organizational perspective.
Susan Silbey identifies three conceptual challenges in the way that safety culture is often discussed and understood.
The first challenge is that safety is defined and measured by its absence, and the pursuit of absolute safety can obstruct the achievement of more realistic safety goals.
The second challenge is that measures designed to enhance safety can also lead to destruction, and the routine success of organizations is often due to the fact that workers interpret and adapt protocols and procedures in order to suit local conditions. Some approaches to understanding this resilience offer complex models of feedback loops and fail to consider the ways in which human judgment and interpretation shape system performance.
The third and final challenge is the tension between promoting individual responsibility and accountability for safety while also recognizing the structural and organizational factors that shape safety outcomes. The concept of safety culture often ignores issues of power, inequality and conflict within organizations. The idea of safety culture often assumes that all members of the organization have the same interests and can be expected to follow the goals of increased safety enthusiastically. The interests of upper-level management and lower-level workers are often not the same, and lower-level workers may lack the institutional support and resources to advocate for their own safety concerns.
Silbey concludes by calling for a more critical and reflexive approach to safety culture that takes into account issues of power, inequality, and the broader social and political context in which it operates.
Le Coze concludes that safety culture is a contentious notion that has seen divergent views, and this can be attributed to the interactions between academics, consultants, publishers, industries, and regulators that shape knowledge production and influence practices. Safety culture can help us think about the networked high-risk systems of today's world, where companies import and translate ideas, methods, and practices introduced by outsiders, including consultants. The lack of empirical data about how these methods, ideas, and practices are concretely translated in companies remains a blind spot in safety science.
Read Susan Silbey's modern classic Taming Prometheus for more information.
Read Jean-Christophe Le Coze's paper How safety culture can make us think.