Normative safety and/or adaptative safety : how resilience engineering revisits activity ergonomics

Theories and methodologies
By Lucie Cuvelier, David D. Woods


For over 10 years, scientific discourse around “resilience engineering” has produced numerous debates on concepts such as: sacrifice decisions, trade-offs, capacity for maneuver, adaptability/fitness, variability, culture, etc. These various notions all have in common a strong connection with concepts developed since the 1950s as part of the research trend known as activity ergonomics (Leplat & Cuny, 1977; Ombredane & Faverge, 1955; Wisner, 1972). Although the distinction between work-as-done (WAD) and work-as-imagined (WAI) is a long-standing one in Francophone ergonomics, the debates are not finished, and more conclusions can be drawn from the identification of the prescribed versus actual distinction (Duraffourg, 2003). The vast array of wordings used in the safety domain by Francophone authors in the wake of resilience engineering appears to bear testimony that more inquiry is needed: constraint safety versus safety in action, regulated safety versus managed safety, normative safety versus adaptive safety, etc. All these notions and the various attempts to clarify distinctions or explore possible combinations in between extreme points leads us to explore the various positions taken by activity ergonomists regarding the gap (Maline & Guérin, 2009). Based on distinctions developed by Béguin (2007b, 2012) for the field of design, this paper considers three different goals pursued by ergonomists when confronted by the gap between WAI and WAD: (1) Eliminating or reducing the WAD/WAI gap, viewed as harmful to both operators (for whom the cost of bridging the gap is high) and corporate performance; (2) Managing the WAD/WAI gap, viewed as providing some limited leeway in activity while still pressuring for compliance out of fear that adaptation could be excessive; (3) Highlighting the WAD/WAI gap is a fundamental, universal dimension of human work systems that requires dialogical approaches to go beyond the two opposing poles.


  • Safety
  • activity
  • resilience engineering
  • activity ergonomics
Go to the article on