Safety is defined in terms that may not support its achievement.
The term implies, ultimately, a condition of ‘perfection’ that defies natural laws and logic. Previously, safety referred to the absence of accidents, and essentially as an unwanted loss for the organization, and a rational construct. At some point the construct became an issue of value and of righteousness and evolved into a moral imperative. Linking rationality and morality in a management model leads inevitably to a goal statement of ‘zero’, progressing to the self-deceiving and self-discrediting one of ‘zero harm’ – not attainable, not aspirational, not inspirational and certainly not manageable.
The ‘condition of perfection’ defies the real world of business or of technology. The flow of processes in business or society is subnormal, and always so. It never gets near a condition of risk-free, and yet the raison d’être of the safety profession is to eliminate risk, failure, variability and error. It then states false goals of compliance, prevention, quantification and defences Nature, and the organization, drifts towards its intended mission, and randomness is not only ignored by the profession, it is denied. Our profession operates on a number of delusions, which defy logic and the reach of our standard, current management science. The most damaging delusion is the one of simplification, while safety is all but. “Risk is not rocket science, it is far more complex than that” (John Adams). Accident models, whether dominoes or Swiss cheese, perpetuate this delusion.
Safety management continues to be a bolt-on function in the business world, as a result of the above and other delusions. It continues to be a legalistic matter, a program, an investigation if catastrophic events occur, and a focus on tangibles. Any organization, every day, every moment, operates at a subnormal level, with risk and randomness intertwined in complex ways. The most potent defence we have, the human being/operator is targeted with processes and programs that reduces his/her potency, such as behavioural interventions. The operator’s cognitive competence is destroyed with a compliance paradigm, instead of creating a culture and work environment where the coping is transformed to enablement. This is achievable through transformational leadership, at all levels, and not through simple models of commitment, management by walking around or changing behaviours.
It comes down to our “worldview” in safety. Do we see the organization as a logically, rationally producing ‘machine’ with gears and cogs and ratio’s that could be optimized by fiddling with its parts and its mechanisms, or as a chaotic, unpredictable process? The first one is the way we produce our products, and it seems right and practical. But we can’t see the ‘management’ of risks and safety that way, because accidents are not ‘produced’ that way, even though the entire safety profession operates on that principle.
Think of a radio station, whose sole purpose it is to entertain us. Some do it with prank calls, and it is very funny. We all laugh and the ‘clowns’ really succeed in their jobs. They do it a lot and every day. Then someone at the other end of the prank call commits suicide and it is not funny anymore and we condemn the clowns. What has been totally ‘right’ before, in fact, was its purpose and what made it successful, is suddenly wrong. We now trace back to see what wrongdoings there were. We pick them out, lift them out in the bright light and everybody can see what was (wrong). Inevitably, we blame the two “wrongdoers”. But had we looked at the same thing a week before, it would’ve looked right and good and funny and successful. The Prince himself even laughed…
Organizations do exactly the same, after the disaster ‘struck’. Then the experts/safety pros come out, point to the dead bodies, then point to the preceding activities and “causal events”, show the clear “links”, leading directly to the wrongdoing and the wrongdoers – and you dare not disagree.
Note: This post was co-authored together with Corrie Pitzer
Hi Lincoln,
I agree in principle with what you (and Corrie) have written.
But I’m not convinced that the term ‘safety’ implies a condition of ‘perfection’. In my understanding safety doesn’t mean an absence of risk, nor necessarily an absence of incidents.
I also think we need to be careful that our ‘professional’ debates and dialogues do not lead to a loss of confidence in us as ‘technical experts’ that fail to agree on what terminology we will use or wha our terms mean.
When we place ‘absolutes’ on our terms, as in the opening of your article, we actually tell people what the word DOES mean rather than clarifying how it should be used. Eg: I did a quick web search and Wikipedia defines safety as:
“Safety is the state of being “safe” (from French sauf), the condition of being protected against physical, social, spiritual, financial, political, emotional, occupational, psychological, educational or other types or consequences of failure, damage, error, accidents, harm or any other event which could be considered non-desirable. Safety can also be defined to be the control of recognized hazards to achieve an acceptable level of risk. This can take the form of being protected from the event or from exposure to something that causes health or economical losses. It can include protection of people or of possessions.”
Note that this definition is quite broad and includes several alternatives. Yes it includes the part that you have used to introduce your article but it is not limited to that definition.
So we need to find a way to clarify how a term is used (what we want it to mean in a given context) rather than undermining it’s value by establishing strict limits on its definition that are not supported in other places.
Rather I prefer to think of Safety as one of the alternatives in the Wikipedia definition “Safety can also be defined to be the control of recognized hazards to achieve an acceptable level of risk”. Hence Safety is not a ‘perfect’ system but rather it is a ‘continuous improvement’ process.
With Safety Management we establish a process for how people should analyse their work – eg: task risk analysis – to at least examine and attempt to identify and quantify the risks BEFORE work activity occurs. Where relevant and possible some risks should be eliminated. Others can be mitigated and still others can be agreed as ‘acceptable’ unless, or until, further evidence demonstrates that either the risk was incorrectly assessed and/or it is subsequently determined to be unacceptable.
Taking this approach, and relying on the legal requirement for ‘consultation’ with affected parties, we are not looking for a perfect system, but we are seeking to exercise some control over the process and this in turn should lead to a safer (less potential for injury or damage) outcome.
Just as with continuous improvement in a production process, the incident (near hit or accident involving injury) is a non-conformance outcome that can be examined to further inform and continually improve the ‘safety management’ process.
I think I’m showing by this argumant that I agree with the gist of your article but disagree with the way you defined the term safety. And that the alternative definition also suports your view of how ‘safety’ should be interpreted.