Learning or Blaming: The Choice the Diving Industry Needs to Make. Part 3 of 3.
Blog 3 of 3: The Discussion and the Way Forward
This is the final blog in a three-part series drawn from Storytelling to Learn: What Happens Underwater, Stays Underwater — an MSc thesis completed at Lund University. Blog 1 covered the problem and the literature. Blog 2 covered the data. This one discusses what it all means — and what can realistically change.
There is a phrase that has been used in safety science for some years now: "You can learn or you can blame. You cannot do both." That challenge sits at the centre of everything my research uncovered. Across the diving community, recreational divers, technical divers, cave divers, instructors, and lawyers, the data told a consistent story about a system that has chosen blame, largely by default rather than by design. The consequence is that the stories which could most improve safety are the ones least likely to be told. The ones where trade-offs, tensions, workarounds, adaptions, and drift led to an adverse event happening.
This final blog draws on the discussion section of the thesis to explore what the findings mean in context, where the most important leverage points are, and what a genuinely different approach might look like.
The System Is Not Broken: It Is Working As Designed
One of the most important reframings that human factors and systems thinking offers is this: when you see a pattern of behaviour that looks like a problem, ask yourself whether the system is actually producing it consistently and reliably. If it is, the behaviour is not a deviation, it is what the system is designed to produce.
The sports diving industry is, in this sense, working precisely as designed.
Training agencies maintain deliberate organisational distance from the instructors who deliver their programmes, which limits their exposure to liability.
Instructors are trained, by culture and sometimes by explicit instruction, to call the agency's lawyer before speaking openly about an adverse event.
Incident report forms carry disclaimers noting they may be used as evidence in legal proceedings.
The result is a system that consistently produces silence — not because individuals are negligent or dishonest, but because the structural incentives make silence rational.
Rasmussen's model of dynamic risk management describes how feedback loops within a socio-technical system allow it to adapt and improve over time. In a well-functioning system, information about what is actually happening at the front line reaches those with the authority and capability to make changes. In the diving system as it currently operates, those feedback loops are largely broken or absent. There is no reliable pipeline from the experiences of divers and instructors at the sharp end to the organisations that design the training, set the standards, and shape the culture. The imagined system has feedback. The real one does not.
Just Culture. Why the Label Might Be Getting in the Way
The research found that 19.4% of survey respondents identified the presence of a just culture as the single most important change needed to improve storytelling in diving. But the same research found that 55% of respondents either did not know what a just culture was or defined it in terms that were counter to its actual meaning.
That is a problem! The work done over the last thirteen years to introduce just culture thinking to the diving industry has genuinely shifted the conversation. But if people are asking for a just culture without understanding what it requires, the organisational commitment, the leadership behaviour, the protection mechanisms, the clarity about what is and is not acceptable, then the label may be doing more harm than good. It gives the impression of progress without the substance.
The thesis raises the possibility of a different framing: not just culture, but a culture of justness. This is not semantic wordplay. The distinction is meaningful. "Just culture" carries institutional weight — it implies a formal, legislated framework of the kind that exists in aviation under EU Regulation 376/2014.
A culture of justness is something that can be built from within, at the level of a dive centre, a team, a focus group, or a community of practice, without requiring a regulator to mandate it. It is a commitment to fairness, to understanding before judging, to separating the person from the action, and to treating adverse events as learning opportunities rather than evidence of culpability.
A culture of justness starts with leaders — not organisational leaders necessarily, but the experienced instructors, mentors, cave divers, and technical divers who others watch, emulate, and follow. These are the people who can change norms from the inside. And the research showed that where they choose to share stories honestly, others follow.
Fear Is the Mechanism, But Silence Is the Outcome
The research identified fear as one of the most pervasive factors in the system. Fear of litigation, fear of peer criticism, fear of being seen as incompetent, fear of losing certification, fear of jeopardising an agency's commercial position. These fears operate at the individual level, the organisational level, and the cultural level simultaneously.
What is striking is that many of these fears are not grounded in documented outcomes. When the question was posed in various social media forums — "Has anyone experienced actual legal consequences from voluntarily sharing a context-rich story about an adverse event?" — there was no substantive response. The fear exists in the stories that circulate about what might happen, not in the evidence of what does happen. One of the lawyers interviewed confirmed that, once litigation is resolved, nothing legally prevents the sharing of the contextual evidence and learning that emerged. That information is almost never shared.
This is Heffernan's wilful blindness in practice — not deliberate deception but an active choice not to look too closely at what is known or suspected. Organisations that ask for compliance data but not contextual learning data are choosing, consciously or not, to remain uninformed. The absence of data becomes the justification for the absence of change, which maintains the conditions for the absence of data.
Leadership can disrupt this cycle. The research cited the example of Alan Mulally at Ford — a CEO who publicly congratulated his executive team for reporting problems, explicitly reversing the culture of punishment that had preceded him. Performance improved. In the diving context, this kind of leadership does not require a CEO. It requires the experienced instructor who says in front of their students: "I once made this mistake, and here is what it looked like from inside it." It requires the technical diver with respected standing in their community who posts an honest account of a close call without the bravado. It requires the dive centre owner who makes near-miss review a normal part of how the team operates, not a disciplinary proceeding.
The Cave Diving Exception — and What It Shows
One of the more instructive findings in the data was the different relationship the cave diving community has with incident analysis. Sheck Exley's Blueprint for Survival, written in the 1980s following a wave of cave diving fatalities, remains a standard reference in cave diver training today. It is built on the systematic analysis of incidents and accidents, communicated through stories, with the explicit goal of extracting transferable learning. Following its publication, the cave diving fatality rate dropped significantly.
The blueprint is not a formal incident reporting system. It is a collection of context-rich stories organised around recognisable patterns, written by someone who held genuine authority and respect within the community. The cave diving focus group showed that this approach — embedding the analysis of adverse events into training from the outset — produces a different relationship with failure. It is still imperfect; fear of peer criticism was present in that group too. But the normalisation of learning from incidents within the training curriculum changes what is expected, what is acceptable, and what is possible.
This is not an argument that every form of diving should become cave diving. It is an argument that the model — of respected practitioners sharing honest accounts as a normal part of how a community develops — has demonstrable value, and that recreational and technical diving have largely not adopted it.
What Context Actually Does
The research showed clearly that context changes everything in how a story is received. When focus group participants were given a simple narrative — incident happened, diver did something, outcome was adverse — the responses clustered around judgment. When the same event was presented with the wider context included, the responses shifted: participants were less likely to blame, more likely to identify with the situation, and more likely to say they would share the story themselves.
This is not a surprise if you understand how attribution works. When context is absent, observers default to dispositional explanations: the person made a mistake because of who they are. When context is present, situational factors come into view: the person made a mistake because of the conditions they were in. Learning requires situational understanding. Blaming requires only a proximate cause.
What the data showed was that most divers do not know what context matters, because no-one has taught them. Training programmes do not cover it. Agency materials attribute causes to individual error. The language of debrief — what happened, what contributed, what would we do differently — is not part of the dive culture for most divers outside of club environments. The skill of telling a learning-focused story is itself a skill that needs to be developed.
The Way Forward: Multiple Levers, Realistic Expectations
This research does not offer a single solution, and it would be intellectually dishonest to present one. The factors that shape whether context-rich stories are told are interdependent, layered, and deeply embedded in the culture of an industry that is fragmented, globally diverse, and largely unregulated. There is no regulator to mandate change. There is no single body that represents all diving. There is no enforcement mechanism.
What there is, is influence. And influence in complex systems works through demonstration, repetition, and the gradual shifting of norms over time. The research identified several practical starting points.
The language matters. Incident, risk, just culture — these terms carry baggage and mean different things to different people. The industry needs a shared vocabulary, and developing it requires education, not just exhortation. The work of organisations like The Human Diver has begun this process, but the reach is still limited relative to the scale of the community.
The structure of incident reporting matters. If the first text on an incident report form tells the reporter that the contents may be used as evidence in legal proceedings, the form is not a learning tool — it is a liability document. The design of reporting systems sends a message about what they are for. Changing the design changes the message.
Leadership behaviour matters more than policy. The most effective learning environments in the research data were not the ones with the most sophisticated systems. They were the ones where someone with credibility and standing had modelled honest reflection in public, and others had followed. This is replicable. It does not require a budget or a mandate.
Finally, the training curriculum matters. If incidents are never discussed during initial diver training — if the word "fatality" does not appear in agency materials for fear of deterring new divers — then the community will continue to enter the water without the conceptual framework to make sense of what goes wrong, or to tell useful stories about it afterwards. Cave diving showed that building this awareness in from the start produces divers who are more honest about failure, not less confident in their ability.
The Central Point
The research showed that stories are suppressed in diving not primarily because people are dishonest or callous, but because the system — its culture, its legal environment, its organisational structures, and its training frameworks — makes silence the rational choice. Changing that requires working on the system, not the individuals within it.
The goal is not a perfect incident reporting infrastructure. It is a community where a diver can say "this happened to me, and here is why, and here is what it felt like from inside it," and be met with recognition rather than judgment. That community would learn faster, operate more safely, and produce better divers.
It is possible. Some of it is already happening, in small groups, in trusted conversations, in the caves of Florida and the technical diving community's quieter corners. The question is whether the wider industry has the will to scale it.
Be better than yesterday.
Part 1: https://www.thehumandiver.com/post/msc-part-1-the-problem-space
Part 2: https://www.thehumandiver.com/post/msc-part-2-the-data-and-results
The full thesis, Storytelling to Learn: What Happens Underwater, Stays Underwater, was submitted in partial fulfilment of the requirements for the MSc in Human Factors and System Safety at Lund University, 2024. Gareth Lock is the founder of The Human Diver
References
Dekker, S. (2009). Just culture: Who gets to draw the line? Cognition, Technology & Work, 11(3), 177–185. https://doi.org/10.1007/s10111-008-0110-7
EC. (2014). Regulation (EU) No 376/2014 of the European Parliament and of the Council of 3 April 2014. European Commission.
Exley, S. (1986). Basic cave diving: A blueprint for survival. National Speleological Society – Cave Diving Section. https://nsscds.org/wp-content/uploads/2018/05/Blueprint-for-Survival.pdf
Heffernan, M. (2011). Wilful blindness: Why we ignore the obvious. Simon and Schuster.
Hoffman, B. G. (2012). American icon: Alan Mulally and the fight to save Ford Motor Company. Crown.
Rasmussen, J. (1997). Risk management in a dynamic society: A modelling problem. Safety Science, 27(2–3), 183–213.
