What if we got it wrong?

file1851274688166People don’t come to work to be safe. They come to work to work. Or as a project manager once told me:

-I probably spend about 50% of my day thinking about behaviours and safety. You probably close to 100% of yours. But what about the guys doing the job? We remind them of safety messages and what we think is important during pre-start meetings and toolbox talks. But then what? Then they’re off, focusing on other things. I don’t think they spend even 5% of their day thinking about what we tell them is important. They can’t. They are too busy getting on with work!

Having safety as a separate topic brings about some problematic consequences. When safety becomes something separate (albeit a priority) it is something we do first or separately, and then we can get down to business. Through this, safety may be reduced to a tick in the box topic that does not carry through to the issues and areas where it potentially needs to be integrated and considered.

In a world in which safety is paramount, organisational life is naturally subject to intense efforts to manage and control what can go wrong. This takes place from within a longstanding reductionist tradition in which tricky issues and challenges are divided (reduced) into smaller, more understandable and manageable subparts and components. Consequently, the whole of organisations may be divided into departments or functions – Leadership, Production, Information Technology, Innovation, Engineering, Safety, etc. But the whole may also be cut across processes, so that for example design and construction are handled by different units. This refinement can continue as far as it is believed it provides efficient and meaningful chunks to manage.

So, separating ‘safety’ into a distinct thing to manage, to talk about or to do, makes sense and may facilitate specialisation and further refinement around specific safety issues. ‘This is the process for cleaning the admin building’. ‘These are the golden rules’. And so on. However, the further the reduction of chunks into sub-chunks goes – the more detailed safety issues we talk about – the more specialised the perspective and knowledge becomes. And the more specialised a particular perspective grows, the more disconnected it will be from other aspects.

At some stage, however, someone will have to integrate the many parts in a work setting. Someone will have to overcome the tension between the various bits of specialised knowledge and incorporate ‘safety’ into the seamless stream of everyday activities. This is the dilemma that many people working in safety-critical industries will find themselves in; they have to be efficient, get the job done as quickly and cheaply as possible, but they also have to be safe. Moreover, various contradictory goals need to be managed under time pressure and under uncertainty, and it is hard to foresee the outcome of certain actions; small effects can have large consequences. People are the only resource in the system that can make these kind of trade-offs, as they are the only ones who can adapt and prioritise under pressure, as well as act under uncertainty. These are the kind of issues we need to talk about. These are the issues we need to facilitate the management of. And these are the issues we need to help each other to make informed decisions about. Over-focusing on the safety aspect of work does not only do great injustice to the people that integrate these various competing goals on a daily basis, it is nonsensical.

Paradoxically, it seems the more we refine roles, accountabilities, rights and duties around safety the more we lose sight of how the many parts of an organisation go together. Analysis, decision and actions become miss-calibrated or misguided as we increasingly try to push the safety agenda. For example, much analysis goes into trying to solve what people think is a behavioural safety problem, when a more effective solution for safety may be found in looking for an easier or more engaging way to do the job.

In effect, when an organisation develops a safety procedure for cleaning the admin buildings (slip hazards, manual handling and chemicals were the reasons given to me), or put up signs about its golden rules, it may be a sign not of safety integration, but indeed of safety separation.

While safety is an important outcome of organisational processes, it does not necessarily mean that it is best achieved by putting a particular focus on safety. While many organisations have employed safety professionals, it does not mean it’s a good idea to give them the task to solve safety issues. And while having a 5 or 10 minute ‘Safety Share’ in the beginning of every meeting is pretty much standard in any ambitious Australian business, it may be detrimental for safety.

So what is the alternative? Safety emerges, is tested and shaped in the interaction between myriad aspects and inputs, such as design, culture, construction, business pressures, environmental changes, physiological and psychological fluctuations, competing goals, resource constraints, administration, training, information technology (to mention a few). That is to say that safety is about how many different aspects are interconnected, interdependent, and interact.

If we believe that safety comes out of the interaction between people, tools, resources, goals, environment and the organisation, then safety should be about gradually seeking solutions outside of safety. When faced with something that is considered a safety issue, we should not see it as a safety problem, but an efficiency, production or operational issue; about getting the job done. How else can we bring what happens “off the frequency” into the conversation and seek to integrate as much as possible of all the factors that influences safety?

What if we got it wrong in putting safety first? What if the reductionist approach has outlived its usefulness in safety? A more holistic way of thinking naturally invites more people to be part of the discussion and to shed light on how things go together. One of the greatest challenges for future safety professionals may be to come up with ways and mechanisms which allow organisations to comprehend what is going on, not just on the safety front.


Note: Thanks to Roel van Winsen, PhD candidate at the Safety Innovation Lab at Griffith University for the help in connecting the pieces..


  1. Fred Schenkelberg Reply

    Safety first is little more than a slogan. Make safety a clear value to the individual.

    I learned about a program for employee driving education – and for employee spouses. The people that took the training became better drivers and made the connection that the company program demonstrated concern and care for them as employees. The benefit to the employee was very clear.

    If at work we emphasis meeting the quarterly numbers and by the way be safe… we really are missing the mark. I suggest stopping emphasis which that which increases tension, risk, and accident rates – if safety is first, we don’t have to talk about it. It is in the way we operate the business and how we behave everyday.



  2. Bill Mullins Reply


    Reductionist deconstruction of work is a logical (note I do not say natural) consequence of extending the Newtonian paradigm beyond the scope of engineering with inanimate materials. Most of the 20th Century saw psychologists, sociologists and their colleagues in organizational development and economics ( and more recently engineering and regulatory affairs) intensely engaged in what’s been termed “physics envy.”

    And that means Newtonian physics – sans quantum entanglement and relativity theory’s demolition of the “privileged stance” – that later of course is the simply presumed “logical ground” of every said-to-be “objective” experiment in the social sciences.

    You correctly observe how the dusty museum of ever more complicated Ptolemaic models of fine-structured work control systems still attracts thousands of adherents to conferences each year. Compartments beget professions, professions beget dogma and ever-less porous theoretical boundaries and so it goes.

    Here are a few notions of how a post-Ptolemaic, post-Newtonian world view might be described in an axiomatic frame:

    – Performance is intentional work: only humans can compose performance
    – Performance is an emergent phenomenon; it emerges from the entanglement of productive and protective actions
    – Protection serves to sustain the performing institution and its capital assets in being
    – Production serves to accomplish the performing institution’s mission and to earn the energy of sustainment
    – Production and protection are a dyadic pair: grossly distinct but intimately entangle in the way a sailing vessel can take a port or starboard tack, but not just one
    – Because protection encompasses preservation of human, constructed and financial capital, protective actions only serve a single preservation objective under rare (e.g. life-saving emergency) circumstances
    – Put succinctly, the objective of competent control systems is to Do Work Safely; this strategy stands in contrast to the Reductionist standard: Do Safety, Work when Safe Enough
    – While efficiency and timeliness are more typically salient from the production perspective on Performance; effectiveness is often most significant from the planning and acting for adequate protection stance
    – In the navigation of challenging circumstances there are no absolutes of relationship between protective and productive efforts – because of human natural perception limits we “tack” shifting our attending back and forth from one aspect to another throughout all work
    – For most institutions, “Minimize avoidable rework” is a sensible over-arching Performance objective – clearly performing so as to go home safely day-after-day fits under this objective.
    – Do Work Safely enables comprehension of the “both/and” character of how production and protection activities can be quilted together
    – Assurance, learning from experience, standards-based norms and procedures, efficiency-thoroughness trade offs, and many other component of Quality Performance fit comfortably in this scheme
    – Safety First does not – but that’s okay because it serves no true purpose anyway!

    Happy Holidays

  3. Drew Rae Reply

    A good example of what you’re talking about might be HAZOP (Hazard and Operability Studies). It’s a “safety” technique in chemical process industries. One of the more interesting empirical findings about it is that around 80% of the issues it discovers are not directly safety related. When discussing the technique with students I emphasise that this is a good thing – it means that people without safety in their job title see the technique add value by improving designs in ways that save money and make them easier to use.

    If it happens to do a decent job of hazard analysis also … well that’s something for the safety manager to smile about quietly as they see people integrate the technique into their design process.

    I think it is possible to take your line of reasoning too far though. While I agree with it as far as you’ve taken it, I’ve also seen similar arguments used to say “we don’t need safety engineering, so long as we do good engineering”. This ignores the fact that safety _is_ an externality, in the sense that those who benefit from risk are not necessarily those who experience the consequences (or are removed enough in time or thought not to envisage those consequences). This means that there has to be some element of safety as a forcing factor – a thumb on the scales to make sure that an optimised design/process/behaviour pays due attention to the safety consequences. Believing that this will happen naturally without people who have responsibility to make it happen is a mistake. I think the point you were making, which I agree with, is to do with how these people go about influencing others.

  4. Tanya Hewitt Reply

    Some time ago, I was at a patient safety conference whereby an aviation representative gave his talk (a fashion in patient safety is to learn from aviation specifically). I asked him to expand upon the “safety first” myth – he clarified by stating that instead of a priority, which tends to be quantitatively measured (accident rates, LTI, etc.) safety is a core value that underlies all decisions taken by everybody at the company. This needs reinforcement in messaging by leaders, and everybody must “walk the talk”.

    Jim Collins, a writer/ management consultant, stated though that workers rarely can articulate values (or what underlies “how we do things around here”). When “sacred” practices are challenged, however, they are very adept at “cloaking themselves in core values” – which is a significant impediment to change, especially if the core values are poorly understood.

    Safety, when delegated externally to “the safety guy” becomes somebody else’s problem, and can rapidly be seen as a impediment to production. There is seemingly little problem in marrying production and financial concerns. Pretty well everyone understands how money underlies all decisions and “how work is done around here”. Dr Peter Pronovost, a patient safety guru from John Hopkins, has questioned why the rigor of financial auditing has not been translated over to quality and safety in a healthcare context. If production can be seen as the interplay of many different influences, safety being one of them, hopefully safety can become just “how we do things around here”.

  5. Rob Long Reply

    Daniel, one of the greatest cultural impediments for safety is hubris. Overconfidence in organisations about safety is a sweet condition for blindsidedness. The very question in your title entertains doubt and yet the contradiction about certainty in risk (all that is uncertain) seems to be missed in a sector addicted to certainty.

  6. Caryl Anne Reply

    Safety should be at the top of the list when it comes to priorities within an organization. Workers come to work to “work”, but if they do not feel safe, they might not be as productive as they could be. This leads to lower efficiency or effectiveness within the business. Thanks for sharing your thoughts!

Leave a Reply

Your email address will not be published. Required fields are marked *