by Carsten Busch
Mainstream and "new" approaches to safety are meant to complement each other, but for best effect, they need to co-exist.
Over the years there has been much debate around whether "new view" safety should be integrated with mainstream OHS. Intuitively, I am inclined to say "yes" to integration. One reason is because I think that getting everything under one banner might help to end or at least diminish the increasing tribalism in our profession. Recent "us and them" debates have wasted a lot of time, space and effort. Probably they added little to the development of the profession as such. If anything, they have widened the chasm with little increased understanding on either side.
One of the important things that tends to get lost in the various debates is that the "old" and "new" approaches to safety are meant to complement each other. For best effect, they need to co-exist, although perhaps some traditional practices should be discarded, while some new approaches need to be tested in practice to a greater degree before we can fully judge them on their merits.
In theory, there is promise. As most things in safety, however, the answer is probably not as straightforward as my spontaneous "Yes!". The only correct answer to most questions is often "It depends". Let us explore the subject from a couple of different sides.
New is the future old
The first thing to ask ourselves is about the meaning of "new view". The schools of safety that are commonly captured under this moniker go under various names. They all differ slightly from the others, emphasising other elements. We find, for example, Safety II, Safety Differently, Human and Organisational Performance (HOP) and Resilience Engineering. Their origins can be placed roughly around the turn of the century when papers by David Woods
and Sidney Dekker spoke of a "new view". Shortly thereafter, resilience engineering emerged as the first recognisable school. I have become increasingly reluctant to speak about "old" and "new" views in safety.
The "new view" is now around two decades old, which may remind us of the fact that "new" and "old" are relative terms Gust as "differently", for that matter). Everything we call "old" now, used to be "new" at some point in time, and since things often move in cycles, sometimes "old" becomes "new" once more. That makes them terms with best-before dates and perhaps another reason to integrate ASAP.
My main issue, however, is that the word "new" may suggest that it is something that has come to replace something "old". Something that is automatically better. It is not. While it may be useful for pedagogical purposes to contrast "new" and "old" approaches, it is not very useful for practice (except maybe to remind you that you want to do something different from last time).
Also, the power of definition "old" and "new" lies unsurprisingly with the newer kinds of safety. This is problematic when the "new" builds strawmen out of "old" practices. It can be useful to communicate about differences in their extreme forms, making clear what the "new" bring to the table. However. when people take these rhetorical statements about "the others" as gospel, it may work against its intention by creating black and white thinking that leads to throwing out the baby with the bathwater.
Integration: Beware
While I am generally positive about integration, shallow interpretations are one thing to beware of. Interpretations that pick up the buzzwords and some of the practices, then adding a veneer of "new view" to existing approaches. Interpretations that are superficial by creating a cargo cult of "new view". Interpretations that possibly turn against the original goals. This has happened all too often with other initiatives and concepts and I fear history will repeat itself.
This is a major risk that rides on the back of the success of essentially useful initiatives and concepts. When "LEAN" is reduced to standing around a board each morning, when "just culture" becomes another word for how we are going to discipline the person making an error, when "safety culture" becomes a synonym for following safety rules, and when "psychological safety" is yet another way to shove responsibility and accountability towards
the sharp end.
This is what happens when one tries to do things the easy way with least amount of effort. When one just copies some tools, but not adopts the mindset to use them. Deming already cautioned that copying without knowledge and understanding of principles is a problematic thing to do. So, if we want to integrate, we need to be prepared to make an effort.
The danger of dissonance
This effort comes with another challenge. Looking at the characteristics that are commonly attributed to the "old" and "new" views of safety, one gets the impression that these are on opposite sides of the spectrum. Take, for example, reductionism. Traditional approaches are to a greater degree focused on identifying broken parts (be they technical, human or social and then fixing them. Newer approaches stress the importance of context and seeing the system as a whole. This seeming opposition can be resolved by assessing what approach fits
the situation best. Simple problem solving (for example, when the light is not working)
benefits from reductionist approaches, while systemic improvement requires a holistic
approach. In that sense, "old" and "new" can very well coexist.
But there are also things where ideological differences are more serious. An example is the concept of "error". In "old view" approaches, these are seen as a useful category to tabulate, trace down and attempt to reduce. Errors are causes for bad things happening. The "new view" regards them as a consequence of other factors in the system.
Errors are merely a hindsight attribution, fuelled by knowledge of outcome. While I subscribe to the latter view, I acknowledge that this may clash with the everyday understanding and use of the term. I myself also use the term regularly as a practical shorthand for a certain kind of event. The Integration of "old" and "new" may, in cases like these, create mental dissonance
because some things are just hard to combine. As we say in Norway, to deal with this, you have to be able to keep two thoughts in your mind at the same time.
But then, maybe we have to reflect upon what we mean by integration. I think that integration should not mean that things are placed under one unifying theory. Instead, we should rather think about integration in terms of putting things into a wider framework that celebrates and utilises diversity and variety. We should think in terms of a mental toolbox from which we can pick the right tools for a situation, just as a carpenter has more tools than only the proverbial hammer. This may be hard, because it means that there are no right and wrong concepts per se, but there are right and wrong applications. Choosing the right tool for the situation may be demanding.
Are we ready?
This brings us to another reason that makes me hesitate. I sometimes wonder whether
organisations are ready for something "new" when it seems that they do not even master the "old". In 1941, safety pioneer Heinrich reflected uporn the simple accident prevention approach he advised to follow. Industry was not ready to embrace and practise more advanced approaches, he thought. Many did not even understand and practise the simpler basics: " ... the exclusion of detailed treatment is well justified because industry in general is not fully conversant with the more simple and direct approach to accident prevention and
does not apply it in practice." Therefore, they first had to understand these simple principles and act on them before being able to proceed.
It is tempting to think along these lines. The question is whether this applies today. Is it necessary to go first through one and then adopt the other? I don't think this need always be the case because there is no necessary progression from the "old" to the "new". Earlier on I called them complementary. They are different tools at our disposal.
Safety does not ask for one-size-fits-all approaches, and where "old" approaches do not give the results we desire, we could try "new" ones. So that means that integrating different practices provides us with opportunities, which we should embrace and use for the
creation of safety.
It will be hard work
One final and important reflection to finish with ... Do the organisations we work for actually want this? Some surely do and have shown so in the past. However, aren't many organisations - perhaps even the majority most of all looking for problem solvers that fix safety issues, keep the regulator and inspectorate from their backs and ensure the workforce behave in certain ways?
If that is what industry expects from the profession, then the "new view" does have challenges. First, because it is not established (yet). What we do not know, we often fear. So many may rather opt for the status quo of what they know than trying something they are unfamiliar with. Something that actually may rock the boat. What, with all this talk of autonomy, reducing bureaucracy and abandoning familiar metrics?
I think this is one reason why much safety education is the way it is - to provide industry with the type of problem solvers that they need. This is most likely one main reason why critical thinking is not part of the mainstream safety curriculum, not even on an academic level. So, there may have to be an uphill fight to integrate "new views".
It may also mean that the current teachers need to be upgraded In one way or another, or perhaps better, they need to be supplemented with different teachers. Second, some of the "new view" approaches may be harder work, and do not promise to give easy answers. Instead of a "root cause", investigations may suddenly present you with a complex context
without a "smoking gun" to fix. Instead of a linear narrative, you may be presented with multiple, seemingly conflicting narratives. If something goes sideways, instead of finding someone to blame and then discipline or train, you may need to address the greater
system. Personally, I am convinced that this harder work will pay off in the long run with
better results (including, but not limited to safety). but if you are aiming for short-term
results this may seem unattractive.
James Reason said in his 2008 book The Human Condition, "Safety is a guerrilla war that you will probably lose (since entropy gets us all in the end), but you can still do the best you can." I have always seen this as an encouragement to do our very best. I also see it as an argument to answer the question we started with. The "old" and "new• views are in this guerrilla war together, so I guess they better join forces.