The ‘failed state’ of safety

The ancient city of Sabratha, LibyaI recently gave a talk about Safety Differently to a group of, mostly, safety professionals. As usual, I offended some people with my spontaneous jokes (this is easier to do in some places than in others, believe me), and made those who have vested interests in the old paradigm squirm or look shell-shocked (“but, but, my posters saying that ‘safety is our number ONE priority’ actually work…!”). Also, as usual, I divided the room, leaving nobody the option of remaining neutral in the face of what I said. At the leading edge of science, there is always a continual and sometimes hard clash of diverging interests.

So far nothing new under the sun.

But a more sinister, corrosive tone seemed to run underneath it all. To understand it, remember that I was born and bred right next to the most liberal, tolerant city on earth—Amsterdam. This is a place stamped out of the mud by individual citizens whose real collective enemy was water, where rulers (worldy or ecclesiastical) were mistrusted, ignored, mocked or thrown out, a place where Rene Descartes finally felt safe to express his ideas in a time dominated by the dogma’s of Church and Crown, the birthplace of Baruch Spinoza (the child of Portugese Sephardic Jewish refugees who weren’t welcome anywhere else in Europe) who laid an important part of the foundation of the 18th century Enlightenment. It is, and always has been, a cosmopolitan, tolerant place, indelibly resilient and attractive because of its celebration of diversity; the place on which New York (New Amsterdam) was forever modeled. The real taboo in Amsterdam is to have a taboo in the first place. It is the kind of place where ideas cannot be self-evidently true for a long time, as other ideas soon emerge to show other ways of thinking or being. It is a place where rules and rulers are treated with equal skepticism, a place where riding your bike through a red light is statistically safer than riding it through a green light, because you don’t get creamed by cars or trams turning into you, and because you are probably more vigilant and adaptive.

And so here I was. Giving that talk. One of the charges I commonly lay at the feet of the safety field (our own feet) is the moral capitulation, historical vacuousness and intellectual complacency inherent in our own commitment to the current paradigm (to think that I would use the word complacency). We are not only economically wedded to the way we have done things for ages; we are also afraid to let go of what we know, we are suspicious, if not intolerant, of outsiders, we are literally conservative—clinging on to how things always have been. Read the beginning of the book Safety Differently, and you will see what I mean. We don’t ask tough questions about whether our interventions really are the right thing to do, we don’t show much historical curiosity to find out where our ideas might have come from, and we happily submit to rote repetitions of tired, retreaded and proven useless behavioral ideas—posters, safety songs, compliance controls in the form of a rush of job observations at the end of the month because someone, somewhere is waiting for the excel sheet that shows you conducted them.

Some safety people tell me that “we ourselves are the biggest obstacles to doing safety differently.” Safety professionals are as much cause as consequence in this; configured as they are in a set of bureaucratic relationships where they are held accountable for showing low numbers of negative events, or high numbers of personal protective equipment (PPE) compliance rates, completed safe work observations or risk assessments. It was not always this way. The neo-liberal (or neo-conservative, for those in North America) governance turn in many Western countries has created a situation where safety regulation and inspection is increasingly delegated down, into organizations themselves. Organizations, in turn, push it down into local work units, or onto the heads of individual workers. It is, in a sense, MacDonaldization on a massive scale: make the customer do most of the work. Because things are a lot cheaper that way. And the single common currency that makes that world go ‘round is safety numbers. In a world where safety is increasingly a bureaucratic accountability that safety professionals need to show up, and to a variety of stakeholders who are located far away from the sharp end, it makes sense that safety gets organized around reportable numbers. Numbers are clean and easy to report, and easy to incentivize around. They are gratefully inhaled by greedy, if stunted and underinformed consumers: insurers, boards of directors, regulators, media, clients. Though frequently and commonly experienced, the ills produced by this are better known than they are publicized. They include:

  • Unethical ‘management’ and suppression of incidents and injuries;
  • The hollowing out of expertise and the forced exile of common sense;
  • The cynical, mindless, rote tick-flick of risk assessments and work observations that largely look for what we already know;
  • The cynical and invasive surveillance and monitoring that reaches into the capillaries of people’s working lives;
  • The deep illusion that nudging the small stuff to go away can prevent the big stuff from ever appearing (like the plant going ‘boom’ in the night).
  • The abuse of the bureaucratic safety umbrella by safety reps in some countries; it shields and sustains union leverage over minutiae of work and turns safety into a club to beat Das Kapital with, often in pursuit of petty gains and perquisites that have no demonstrated link to safety. But those gains and perks are the panem et circenses (the ‘bread and games’) offered by union leaders to keep the masses pliant and the dues coming in;
  • The kind of authoritarian high modernism that makes even those in positions of power submit meekly, with no more trust in common sense, and no hope that things can be better under a different system of safety governance 

A Canadian study uses the inelegant word ‘responsibilization’ to capture the neo-liberal or neoconservative trend. It notes, along the way, that such governance is accompanied by an increasing number of fines, citations and prosecutions hitting individual workers for workplace safety violations—rather than the employer, as used to be the case. Performance-based safety regulation was once a great idea: the organization could show the regulator that it had its risks under control, and that it could deal with harmful influences that come its way—by whatever means and in whatever ways. It was a regime that promised liberalization, the end to overspecified and overprescribed rules, and a legitimation of diversity, context-sensitivity, innovation, open-mindedness. To say nothing of budgetary savings. But perhaps it is turning safety into a ‘failed state:’ with eroded governance conditions, capacities and responsibilities, and abused, mistrusted or ignored institutions. And ultimately of course, as the very definition of a failed state: it is rendered increasingly ineffective. The current safety paradigm is no longer really getting those reportable numbers any lower. And what is worse, it has not made serious dents in fatalities or process accidents across many industries for a decades now.

What it seems to boil down to is a silent, and in a sense fundamentalist, pact of small-mindedness that many safety professionals have to submit to: stay comfortably within the narrow bandwidth of what is politically correct, occupy yourself with only the sliver of technical callings of the field at hand, and for your own sake and that of your contracts or continued employability, do not touch the taboos. Do not ask the questions you don’t want to know the answers to, or that you don’t want others to know the answers to. Well, if we want to change the paradigm of safety, it won’t work that way. Safety is not going to be changed on the back of technical tinkering alone. If we want to make fundamental changes, then we need to go right to the heart of how we think about, and hold beliefs about, deeply sensitive and tricky issues like ‘sin’, or accountability and blame, about retribution and social justice, about trust and relationships, about organizational governance and managerial integrity, about inequality and the distribution of capital, about hierarchy and power, and indeed also about the excesses of union abuse of their putative ‘voice from below.’ If we don’t dare to reach into such aspects of political economy, social history, or moral convictions, then we cannot ever begin to change the paradigm.

Back to the talk. I told my audience early on that I didn’t want to be right. That is not what this is about. It is not even fun to be right. I’d rather be proven wrong (though you have to come armed with strong arguments or persuasive data). Being proven wrong is indeed the basis for scientific progress. So I’d be happily proven wrong. But for that debate to work well, we all have to become willing and able to see where a well-argued or well-founded idea begins and where a particular conviction ends that we slavishly adhere to for its mercantile advantages. In this, everbody has to give and take.

A good friend of mine, himself an innovative safety thinker, told me that when confronted with my own outrage about safety people’s moral capitulation and intellectual laziness, I should not put their failings on full display, for everyone to see, and for the courageous ones to mock. I know I can do that very effectively and entertainingly (and then I offend some people). Rather, my friend said, I should muster sympathy, understanding, and compassion—even if this is accompanied by the inevitable sadness of having to depart from the righteous joy of seeing the splinters in other people’s eyes. I can only wish I had the wherewithal and patience to follow his advice consistently. I admire those who can. I, instead, often revert to my shock therapy, jolting and shuddering people into a new era—and losing a bunch on the way.

But, haltingly so, that is where I take safety people, or try to. I think people get uncomfortable when I say that if we want to change safety, we have to look into ourselves. We should not look at all those other people (who, we might believe, need to ‘pay more attention and be more careful’). I believe that if we in safety don’t change, nothing is going to change in safety.


  1. Mike Edwards Reply

    Hi Sidney, I can’t help but wonder how many safety professionals have been discarded under the current paradigm because they had the courage to stand up against the poor corporate and unprofessional behaviours from management and so called Safety Professionals that you have flagged.
    I would suggest that one of the critical elements of being a professional, in any playground, is the ability to establish and maintain your ‘professional distance’ at all times. In my experience this means you have to be prepared to never be part of the team. You can on work on a system from the outside. If you’re on the inside you are ‘part of the problem’. As a first year teacher, I was told you can always be friendly, you can never be friends with those you may have to influence or work with – I have also found this to be sage advice in the EHS environment. Mmike

  2. Kelvin Genn Reply

    Sidney, maybe we, the safety profession, are more Leninist than Marxist, aspiring to the establishment of the dictatorship of the proletariat!

    In the words of Pink Floyd, we are comfortably numb.

    The child is grown,
    The dream is gone.
    I have become comfortably numb.

    On that note, time for some introspection.

  3. Rodney Currey Reply

    We are all delusional. Organisations spend millions on fudging figures. When you spend that much you want results. All these organisations have low to no LTI rates. Imagine if you were the new safety manager, and you didn’t maintain the safety record, you would be seen as hopeless, so you keep the illusion alive and running. You join the game. These company’s spend millions only to throw safety under a bus. It’s all an illusion. They fear a fatality that will hinder job opportunities, then fear safety will get in the way once they have the job. Quite a conundrum.

  4. Rodney Currey Reply

    Had a thought the other day on “safety posters”. I have never seen “profit posters”. What I mean by that is these companies as we know create and strive for profit. How is this promoted? How do the managers maintain this, but cannot maintain the same consciousness towards safety? Is this what we should be finding out how they promote and achieve profit compared to how safety should be achieved?

    1. Bill Mullins Reply

      You see a profit poster every couple of weeks.

      A hint; there’s a reason that people are encouraged, often under threat of reprisal, not to share each other’s profit posters (aka pay and bonus checks).

  5. Bill Mullins Reply

    This situation has for some time put me in mind of another time of paradigmatic “failed state” – that of the Ultraviolet Catastrophe in classical physics.

    At the middle of the 19th Century there as an expanding frontier of industrialization built largely upon theory and experiment in classical thermodynamics and later the newly discovered equations of Maxwell’s electro-magnetic synthesis.

    In 1900, a century of astonishing, reason-inspired development from that foundation still lay ahead; but already the seed of paradigmatic (i.e. Newtonian-Cartesian determinism) disruption had been planted. Leading physicists had noted the Ultraviolet Catastrophe – a clearly defective prediction of these classical theorems. Despite their marvelous utility on industrial scales, these theorems lost potency, spectacularly, under down scaling.

    At that point in history, nothing less than a transformative conceptual reframing of classical physics would suffice if the measurements of increasingly concise atomic level observation were to be accommodated.

    The challenge raised by the Safety Differently observation of stagnant performance measures and calcifying theorems of performance improvement fits with the Kuhnian characterization of “paradigm exhaustion.” But the observation of that “failed state” does not, by itself, point toward any direction for paradigm transformation.

    William Bridges, in his analysis of institutional transformation, points out that early in the change movement there is a period of unfreezing and letting go of the power of some cornerstone premise features of the exhausted operating paradigm. This conscious action, particularly by leadership, clears the stage for alternative memes to emerge out of fresh thinking about the institutions failed circumstances.

    In the case of the Ultraviolet Catastrophe, Einstein snatched Planck’s theoretical notion of “oscillation quanta” and set it to experimental duty – thus boosting quantum theory into conceptual orbit. Much of modern electronic technology, and understanding of bio-chemistry, would remain undiscovered but for this leap of imagination to let go of certain classical constraints.

    For a century Safety Management has ridden upon the back of classical determinism – blithely ignoring the inevitable day of reckoning with its own version of the Ultraviolet Catastrophe. That is, the trope that: “Every accident is preventable;” and thus that perpetual experience of Safety is possible.

    Quantum theoretical description resulted in a raft of previously unimagined memes – among them is the Uncertainty Principle. It codifies the constraint imposed by nature upon knowability – and thus predictability – both forward and backward in time.

    In a sense the Safety Differently critique begs the question: How is quantum indeterminacy (i.e. the Uncertainty Principle) to be made manifest in the management of hazards at scales where non-linear system dynamics and irreversibility of accident sequences are in evidence?

    I want to suggest that in the new paradigm we hazard control professionals will find ourselves “letting go” of the meme “Safety Management.” This will be needed because it embodies a blatant contradiction (ala the Ultraviolet Catastrophe) – it conflates a transient feeling state – safety – with a meme implicating enduring intentionality – management. In ordinary and traditional usage, the momentary experience of “safety” is entirely limbic reflexivity; originating in the release from the fight-or-flight response, there is no forward-looking mindfulness associated with it. We variously experience that limbic relief for a time before our pre-frontal cortex is release to begin thinking about “What just happened?”

    The misappropriation of “safety” as a functional modifier for “management” is understandable in the context of times before modern neuroscience; this is particularly so given the early and regrettably continued dominance of behaviorist psychology after Skinner’s experimentation with limbic conditioning. But with current cognitive science’s understanding about the structure of mindfulness, the combination is thoroughly rendered obsolete for analytical and systems design purposes.

    Just as “quanta” emerged to resolve those classical theorems’ breakdown, it seems likely that “hazard control differently” will only appear when a fully post-limbic formulation of the profession is conceptualized. Despite the ease with which these constructs slip off the tongue, Safety Differently, and Safety II embody the obsolete cognitive contradiction and thus should be seen as dissonant – part and parcel of the very obstacle we seek to circumvent.

    While Safety Differently can be useful in describing the need for change, history suggests that the post-limbic paradigm for hazard controls management will not include a grand-fathered reference to “Safety Management.”

    In my experience, substituting Protection Management for Safety Management avoids the contradiction; to be clear, by itself it does not detail the substance of the paradigmatic re-thinking needed. There remains much work to express the growing body of “differently” risk insights that is emerging from the on-going transformational reflection. But, it does accomplish a critical piece of the necessary “letting go” from the Great Stuckness.

    Protection Management, while functionally accurate, may prove to be only a scaffold meme; however, it does acknowledge the prerequisite work of introducing modern neuroscience into the conceptual basis for advanced hazard control in complex, high-consequence circumstance institutions and enterprise settings.

    Speak of “safety” – accurately in the ordinary limbic sense – whenever useful; but, avoid the conjunction “safety management” as both obsolete and detrimental.
    Are we ready to take that step? Do we really have a choice?

  6. Tanya Hewitt Reply

    This is a challenging post to swallow.

    In skimming the responsibilization reference noted above, it made me think of how I needed to justify why safety is similar to Population Health (the current doctoral programme I am in). Blaming the victim (in health) is directly analogous to blaming the front line worker, and upstream interventions (health) are analogous to organizational and higher safety foci.

    This also brings to mind a concept in social psychology of the Fundamental Attribution Error (FAE) – unlike a bias, misappropriating attribution to dispositional factors of individuals to the exclusion of environmental factors is labelled an error. The Milgrim and Stanford prison experiments (which were the genesis of present day research ethics boards) highlighted this very powerfully, and directly. To not acknowledge the demonstrable influence of the larger context in which people behave is not only subject to the FAE – it is unethical, and morally wrong. And yet, we seem not to be able to shake this fundamental paradigm in safety.

    One tiny little anecdote that also comes to mind is when a middle income country’s company took over a Canadian company (in the 1980’s?), and promptly removed the Safety office. The Canadian consultants saw this as the beginning of the end – but the new owners installed a Sustainability office. The Canadian consultants were blown away – not only is this a broader focus than the former Safety office – it is light years ahead in thinking.

    Thanks so much for the discussion!

  7. C Brown Reply

    I recently attended a talk of yours on Safety Differently. Indeed, Sidney, your ideas are an affront to the status quo, and there was a lot of buzz in the week following your talk. For the record, I was entertained. It is admirable to be on the front line of such a provocative crusade, provoking an introspective look at the origins that have shaped our methods in safety, and challenging the fundamental types whose deeply rooted ideas are entwined with personal values and beliefs. This is not an easy feat by any means, and requires a person who doesn’t care much for being popular or compassionate. I think that resistance here can be construed as a positive outcome in that your message was heard, despite the level of acceptance. What can be learned by the safety professionals who are open to dissecting their beliefs and want to take this paradigm shift forward, is an understanding of what obstacles lay in the path towards shifting the insulted mindset, or working around them to start. So, fellow safety crusaders, absorb what you can from the intellectuals, and the ‘intellectually lazy’ and continue moving onwards and upwards.

  8. Tanya Hewitt Reply

    I too recently attended Sydney Dekker’s talk (and got to chat with him the next day!), and he was very emphatic on, for example, HR having nothing to do with safety investigations. Naturally, the HR professionals in the room got their backs up, and one even confronted Sydney with not understanding our country’s context, and that the HR contribution to safety investigations is value add. I was impressed that Sydney didn’t lash out at the guy, but he didn’t – he was just rather dumbfounded that obviously the individual listened to some of what Sydney was saying, but not all. I believe that individual did not return after the next break, but I wonder if the whole interaction might have impressed upon others that there is merit to challenging the status quo, challenging “this is how I do my job”. It is not easy – but being able to critique our safety beliefs is necessary – and overdue.

  9. Jeff Washburn Reply

    Kenneth Burke defined a metaphor as “a strategic predication upon and inchoate pronoun (an I, a you, a we, a they[as in Safety Professionals]) which makes a movement and leads to performance.”
    This definition is apt for this conversation. Shock therapy, sympathy and ridicule separate, but do not make movements. Based on the above definition, where there is no movement, there will be no performance. Also, it follows that since safety is in a “failed state,” then the metaphors that have been in use have failed to cause better performance.
    The prevailing metaphor in the corporate world is: “Business (life) is war.” Under this metonymy/metaphor there are a number of ways that we refer to the conduct of business and life. You can see these in Lakoff and Johnson, Metaphors We Live By. Organizations have captains and lieutenants. Corporations take territory. Apple wins or loses, Samsung loses or wins. Strategy is developed or pursued. We talk in terms of supply chain. These re just a few. Robert McNamara produced lots of numbers under this kind of
    Safety culture is a common metaphor in the safety realm. It has failed to produce improvement, so it has not been a metaphor which causes a movement in the people that are at the sharp edge. Perhaps part of the failure is caused by the prevailing “Business is war” metaphor because it is difficult to see winning. Another failure of the safety culture trope is that its definition relates to characteristics or traits and does not lead easily to clear action which develops those traits.
    I have spent some time blogging about much of this. I will not post it unless requested
    One way forward is to use an overriding metaphor that makes a movement of the people of an organization. That metaphor should easily lead to focusing on safe performance. One I have thought of is: “Business is a work of art.”

  10. Jeff Washburn Reply

    I see that I didn’t finish this thought:

    Robert McNamara produced lots of numbers under this kind of

    I would revise it to say this about measures: Robert McNamara, US Secretary of Defense during the Vietnam War, produced lots of numbers which “proved” the US was winning the Vietnam War. It is a rather stunning failure of performance measures and a great illustration of Goodhart’s Law. Wikipedia states the law: “When a measure becomes a target, it ceases to be a good measure.” My counsel to safety professionals is that safety performance numbers are, at least, suspect. At worst, the almost uniformly good numbers indirectly point to the “failed state” of safety.

  11. Rodney Currey Reply

    Sidney, I think everyone is scared to let the safety genie out of the bottle. Its been in there since safety ever was. The government won’t even try. Safety professionals dare not as it could mean there obliteration from the face of the earth. Organisations and commerce still think the economic world will faultier into a depression or some other financial abyss. So the genie remains in there with many wonderful powers that will never be realized as we are all too scared of what might happen. Whatever that is.

  12. Mike Edwards Reply

    I think you are right on the money Jeff.
    I would add that in my experience, the problem is not the fact that we use statistics as performance measures, rather it is the often dishonest or selective way in which the numbers can be fudged and misreported to cover up inherent and endemic management practices at all levels of the organisation. Usually to protect personal rewards such as bonus payments etc. or just the questionable behaviours from people who seem to have disproportionate impacts on how their companies, sites, teams and people behave with apparent impunity.

    1. Jeff Washburn Reply

      Absolutely! It doesn’t need to have a dishonest intent. The result is still the same.

      I learned this from an uncle (the context was woodworking): “Measurement is the enemy of accuracy.”

  13. David Provan Reply

    Thanks for the post Sid and everyone for your comments. I didn’t understand all of Bill’s comment but his mention of Safety Management being a misappropriation triggered a thought. I think titles are important descriptors of roles. Safety Manger’s are fraud’s, they don’t manage safety, workers do. I’m struggling to think of another profession that so inappropriately titles itself. Perhaps it’s the very name of the job that influences beliefs about the role and the way it is carried out – think Police Officer. What if tomorrow every Safety Manager was re-titled Social Worker. Would they then make themselves available to listen to others challenges without judgement or solution? Would they wait to be asked before they imposed their paradigm on others? Would they see the positive actions and growth in people, not the problems? Would they hold up a mirror to the system so that it could understand and change itself not manipulate data? Would they step back and away when they are no longer needed? I wonder what a title change might influence. Although, now that I think about it, I remember when “safety facilitator” became the rage so maybe we would all still be Leopards, unable to change our spots.

  14. Bill Mullins Reply


    You are making points along the direction I was speculating; some further thought triggered by your comment about titles and roles follows.

    In the governance framework I employ the end-in-mind is the Doing of Work – Safely. That is the domain of Whole Performance; it has Production and Protection aspects which seem quite clearly distinct for purposes of attending to work controls. Of course the sum of controls must flow in a cooperative fashion if the Mission is to be accomplished.

    This is the “Focus on the Work to be Done” perspective; and thus far I’ve said nothing about organization charts, roles, or titles. Those all relate to division of labor – that is to organizing Inputs; where the path to completed work is linear-sequential (Newtonian/Cartesian in Dekker’s sense) then all the parts fit together like Lego blocks and the desired outcome is evident in the assembled parts.

    Much of tried and true practice for achieving the “safely” aspect of the desired outcome has emerged from determinist reduction of work paths backward to timely hazard reductions or eliminations (e.g with reliable PPE). In the well-ordered world of the corporation’s hierarchical organization chart, this is a tidy system of roles and descriptive titles; over time it lends itself to improved reliability and even to more efficient ways of delivering protection.

    The profession of Safety Management developed as the Community of Practice with expertise in the reliable delivery of protective “work controls.” While the frontier of improvement was with reducing acute injuries this system “made sense.”

    But as the mastery of those challenges was largely gained then the matters of “healthful work place” became more salient. Health is more in the form an ecological challenge and not a predominantly mechanistic one. It also engages mindset intentionally, beyond the behaviorist psychology of individual transactions in the stimulus-response tradition. Collective life is authentically complex; work and controls co-evolve (or devolve).

    There is a case to be made for re-conceptualizing the practice formerly known as Safety Management toward one that encompasses the disciplines known as Community Organizing. In this instance we would be talking about Community of Work Practice – and it is increasingly not just Social Work.

    As the scale of industrial and government action impacts continue to grow exponentially, Community Protection Organizing (clunky I know) must ponder the issues of Safe and Healthful Work Place from an integrating socio-techno-political perspective.

    It would be great if Line Executives, or Policy Makers would take the lead on this challenge, but I don’t see that happening for a number of reasons we could discuss.

Leave a Reply to Mike Edwards Cancel reply

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