A ground-breaking new book, Beyond Human Error: Taxonomies and Safety Science deconstructs the conventional concept of "human error" and provides a whole new way of looking at accidents and how they might be prevented. Based on research carried out in the rail, nuclear, and defense industries, the authors show how, by concentrating solely on "human error," systems and sociological factors are frequently ignored in contemporary safety science. They also argue that the "information processing" view of human cognition, the foundation of the majority of safety science and ergonomics, is hopelessly simplistic and leads to ineffective or even misguided intervention strategies. Wallace and Ross explore how what they call the "technically rational" view of science can hamper the process of creating a taxonomy of error events, and the implications this has for the current orthodoxy. In laying out the limitations of the "technically rational" viewpoint, they clearly define their own alternative approach. They begin by demonstrating that the creation of reliable taxonomies is crucial and provide examples of how they created such taxonomies in the nuclear and rail industries. They go on to offer a critique of conventional "frequentist" statistics and provide coherent, easy to use alternatives. They conclude by re-analyzing infamous disasters such as theSpace Shuttle Challenger accident to demonstrate how the "standard" view of these events ignores social and distributed factors.

The book concludes with a stimulating and provocative description of the implications of this new approach for safety science, and the social sciences as a whole. While providing a clear and intelligible introduction to the theory of human error and contemporary thinking in safety science, Wallace and Ross mount a challenge to the old orthodoxy and provide a practical alternative paradigm.