Where five adjacent accounts locate the harm

Five bodies of work approach this territory from starting points the six occupational health models do not occupy. Each opens a line the DFS analysis can extend.

Abusive supervision

Anyone reading Robert’s working life from inside management research will reach for a construct that already exists.

Tepper (2000) defined abusive supervision as subordinates’ perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and non-verbal behaviours, excluding physical contact. It is among the most heavily researched constructs in the field, linked to psychological distress, reduced wellbeing, turnover intention, and diminished performance. Robert eventually acquired a manager whose conduct fits it closely and who did him enormous damage: the fall from being one of the go-to people in his field to redundant and isolated was that manager’s doing.

HDM did not produce the man. Nothing in the architecture creates hostile conduct.

What HDM did was let him function and let him stay. Function: his choices bound others, and nothing required him to account for them. Stay: nothing in the architecture produces a signal against a position holder whose choices were never answerable, and the people whose perceptiveness would have registered what he was doing are the ones the selection mechanism is filtering out.

Substantiating a choice covers what was decided and why. It says nothing about how the choice is delivered: contempt survives substantiation intact. Where this becomes the ordinary practice across an organisation, DFS predicts that who gets selected for directive-authority positions changes over time (Verweij, 2026f), and abusive supervision recedes with the selection rather than being separately forbidden. None of that reaches Robert’s case.

Under a different manager Robert would in all likelihood have suffered less. What would not have changed is that his questions still had nowhere to go, and that he would have had other run-ins, because the prison was in place before that manager arrived, and he had already adapted to it by making himself smaller. He ran into two things: the man, and the architecture that was there before him and gave the man’s choices their force.

Robert’s case has an identifiable actor in it. Every case this paper can point to relating to work-related Complex PTSD does. His case illustrates the role of the HDM architecture. It is no evidence for the other claim this paper makes, that Complex PTSD does not need a hostile actor. Where one is present, it becomes more likely, more quickly.

This gives the two accounts a clean point of separation, and it is testable. Abusive supervision is defined by sustained hostile display from an identified supervisor, and its instruments measure exactly that. So the test runs from the harm rather than from the organisation: take cases at burnout level and beyond, and ask the people in them whether there was anyone doing this to them. DFS predicts a substantial group who answer no, and who show the full mechanism anyway: foreclosed challenge, forced internal attribution, deepening captivity, the perceptive employee carrying most of it. If harm turns out to track hostile display and nothing else, the argument in this paper is wrong.

The two also differ in what they make available afterwards. Where the harm has an identified aggressor, there is something to name, something to report, and a remedy: remove or restrain the person. Where it does not, there is none of that: the harm produced no incident anyone could enter on a register.

Moral injury, not burnout

Moral injury is harm arising not from exhaustion but from being made to take part in outcomes one’s professional expertise identifies as wrong. It was developed to explain healthcare worker distress that burnout frameworks were not capturing. Dean, Talbot, and Dean (2019) made the case; Shay’s (2014) definition — a betrayal of what is morally right, by someone who holds legitimate authority, in a high-stakes situation — corresponds closely to the HDM context. Molendijk (2026) argues that moral injury cannot be reduced to individual actors and involves the systems in which professionals and their leaders operate. Praslova (2022), reporting preliminary data from her own research, puts at least 25% of those reporting burnout as better explained by moral injury.

HDM makes the source precise. The expert who can see exactly what achieving the desired outcome would require, and who is structurally prevented from contributing that expertise, is not being asked to do something wrong. They are being prevented from doing something right. The more clearly they can see what is needed, and the more committed they are to the outcome, the more acutely they experience the impossibility of contributing. Perceptiveness here is the measure of what is being lost.

Robert’s account adds two things to this. The first is that the injury is mostly not episodic. Alongside the few clear conflicts, he describes a daily drip of small ones — the little misrepresentation, the process weighted towards the company — each requiring energy to resolve or explain away, until the boundaries blur and the pre-packaged justification arrives before the discomfort does. The second connects moral injury to what asking costs. He describes what happens to the person who does not blur: they are painted as awkward, disruptive, not a team player, and once their morals stop them doing as they are bid, they become a problem to be solved. That is the same selection mechanism, reached from the values side rather than the reasoning side.

When the harm is not a failure but a product

Institutional betrayal theory (Smith and Freyd, 2014) established that harm done by institutions to the people who depend on them does distinctive damage, producing betrayal blindness: suppression of awareness of harm to maintain the dependent relationship.

The HDM mechanism differs. Institutional betrayal turns on a violation of trust. The institution was supposed to protect and instead harmed. In HDM organisations, the harm is not a deviation from how the organisation functions. It is a product of how it functions. Under institutional betrayal, an employer failing to prevent harassment constitutes betrayal. Under DFS, the employer operating HDM as designed is the harm mechanism. That distinction closes off the solutions institutional betrayal implies — better intentions, improved procedures, whistleblower protections — and opens the only one that remains: structural change to the decision-making architecture itself.

Frustration, not deprivation

Self-Determination Theory (SDT) identifies three basic psychological needs — autonomy, competence, and relatedness — and distinguishes between need deprivation and need frustration. Vansteenkiste and Ryan (2013) established that being forced to comply with a decision one cannot support is a different experience from simply not having a voice. HDM frustrates rather than merely deprives: it overrides expertise while requiring compliance. SDT predicts burnout, exhaustion, and counterproductive behaviour from need frustration, and that is what the architecture produces.

DFS reads the frustration as produced by the structure, not by managers. Directive authority without a substantiation requirement cannot be managed into compatibility with those needs, and the captivity it produces means the frustration has nowhere to go.

How HDM drives purpose-driven burnout

Purpose-driven burnout, the pattern in which commitment to a cause becomes the engine of exhaustion, is well documented across mission-driven sectors. Schaufeli (2009) identifies frustrated idealism as a defining quality of burnout from its origins. The Center for Effective Philanthropy’s State of Nonprofits 2024 report found that 95% of nonprofit leaders were concerned about staff burnout, with 75% reporting it was already affecting their organisation’s ability to achieve its mission. Among humanitarian aid workers, Lopes Cardozo and colleagues (2012) found that between a quarter and a half of workers reported symptom levels associated with high-risk burnout dimensions. The pattern is consistent: those most committed burn out at disproportionate rates.

HDM logic suggests a structural pathway contributing to this. Directive authority sets a desired outcome — a target, a mandate — without substantiating whether the conditions make it achievable. That places the person in an impossible position without their knowing it. They believe in the cause. They work harder. The gap cannot be closed, because it was structural from the moment the target was set. The signal that would normally protect a person, the sense that this is not worth it, is unavailable. Commitment is the condition that removes it.

When failure arrives, the attribution does not land on one thing. The person is failing themselves, since their professional identity is bound to the cause. They are failing their colleagues, who share it and witness the shortfall. They are failing the cause itself. Each layer compounds the attribution, and the shared commitment of those around them intensifies rather than softens it. Lopes Cardozo and colleagues (2012) found that those who evaluated their NGO most positively showed the highest burnout on the personal accomplishment scale, because the stronger the identification with the organisation, the more completely the failure was attributed inward.

The tools compound this further. Mission-driven organisations typically operate with constrained resources: limited budgets, understaffed teams, inadequate infrastructure. These are visible and tangible. When outcomes fall short, the tools become the explanation. The person works harder, seeks better resources, advocates for what they are missing. The unsubstantiated mandate remains invisible behind a practical problem that appears, in principle, solvable.

The cause is being failed because the target was never achievable. As far as the person can see, it is being failed because they have not yet done enough.