Normal Accidents: Living with High-Risk Technologies by Charles Perrow
My rating: 5 of 5 stars

It's hard to put labels on this book. It's about complexity and systems theory and sociological analysis of workplace relations on the surface, but in drawing all these areas together it also presented a fundamentally humanist analysis countering the pragmatic and rationalist perspectives that are more common in this field.

Beginning with the idea that in complex systems predicting all eventualities is, by definition, impossible, Perrow argues that we must accept that some kinds of accidents will be inevitable or "normal". A key concept he introduces is that of coupling. When a system is tightly coupled, one event will lead to another without opportunity for intervention; when a system is loosely coupled, buffers of time and human expertise allow for intervention and the possibility of avoiding accidents.

On the surface, this might sound dry and not a very compelling read, but Perrow writes with elegance and a humanist insight that made this book unexpectedly compelling. I found myself constantly comparing and reflecting on how his analysis informed the work I do as a teacher. How does my organisation compare to those that Perrow analyses? And what can I learn through using the lenses of "tight v loose coupling" and "linear v complex" systems?

The big take-away for me in my work was the role of stress and the impact that stress has in limiting the creativity and efficacy of decisions. Tightly coupled complex systems fail because, when the unexpected happens, the system lacks the spaces to release pressure and one event leads to the next in unforeseen ways. When systems are stressed, the capacity for creativity and agency that might ordinarily allow the unexpected to be managed, is reduced. Perrow was dealing with systems where consequences are much more pronounced that a loss of efficacy in a school (the Three Mile Island Nuclear Disaster is a major focus), but I think the basic insights are transferrable to many contexts. What I wondered, as I train the next generation of managers and CEOs, is how building agency into the systems we use to manage and lead in our school, might model the kinds of professional autonomy that students will need to lead effectively in their futures.

Perrow deals only fleetingly with education in his book (and only with universities). It's a sign of the power of his writing that he could prompt so much thinking outside of the direct focus of his analysis. I'm confident that this is a book I will come back to often.

View all my reviews