(c) Erik Hollnagel, 2020

Are 'Safety-I' and 'safety 1' homophones or synonyms?

A growing number of people, in the health care community as well as in other fields of activity, have heard about something called ‘safety one and safety two’ (the spelling of the numbers is intentional). They have therefore become aware, that some researchers and practitioners propose a distinction between different views of safety, but may not be entirely sure of what this means.

One particular type of confusion or misunderstanding is that the terms point to a difference between ‘safety 1’ and ‘safety 2’, where ‘safety 1’ denotes the current state of affairs and ‘safety 2’ some kind of next step or progression. In other words, to a progression in different types of safety. Furthermore, that ‘safety 1’ is not altogether sound or optimal, hence that there is some kind of argument that we should move on to ‘safety 2’. But this interpretation is a misunderstanding.

First of all, the two terms have from the start been written as ‘Safety-I’ and ‘Safety-II’ and should be written like that. The use of Roman numerals is deliberate, to show that the terms refer to two different views, but not to a succession of views. (In hindsight, it might perhaps have been better to write ‘Safety of Type A’ and ‘Safety of Type B’.) It is therefore incorrect to conclude that we must be finished with ‘safety 1’ before we can move on to ‘safety 2’. The reason for using the two terms is that we want to describe a distinction and not a progression.

The introduction and succeeding multiple written and oral presentations of Safety-I and Safety-II (Hollnagel, Wears & Braithwaite, 2013; Hollnagel, 2014; Wears, Braithwaite & Hollnagel, 2015) also make it abundantly clear that Safety-II is a complement to Safety-I rather than a replacement (hence that there is no intention of a succession). Safety-I is a label that refers to the common approach to safety, where attention is limited to events that result in adverse outcomes (accidents, incidents, etc.) and efforts are directed at preventing such outcomes from occurring. Our argument is that this is insufficient in both the short and the long run and that it is both impractical and impossible to become safe only by preventing things from going wrong. In addition to prevention we must try to make sure that things go right, and therefore also study how this happens in order to understand that better. This is furthermore something that we should have been doing all along, and at the very lest something that must be done here and now rather than something that can wait until we have eradicated all sources of failure. Safety-II simply argues that we must look at everything that happens in a place of work, that we must try to understand why and how it happens, and that we must try to facilitate and support that which goes well in addition to trying to prevent failures.

While ‘Safety-I’ and ‘safety 1’ may be homonyms, they are by no means synonyms. ‘Safety-II’ is therefore not something that can wait until we have done all we can for ‘safety 1’. A failure to recognise this difference is crucial and may be the largest obstacle to real progress in safety.


Hollnagel, E. (2014). Safety-I and Safety-II: The Past and Future of Safety Management. Farnham, UK: Ashgate.

Hollnagel, E., Braithwaite, J. & Wears, R. L. (Eds.) (2013). Resilient Health Care. Farnham, UK: Ashgate.

Wears, R. L., Hollnagel, E. & Braithwaite, J. (Eds.) (2015). The Resilience of Everyday Clinical Work.  Farnham, UK: Ashgate.