The Capture of the Professions

Psychiatry, the academy, engineering

Psychiatry claims scientific standing while also making value-laden judgments about distress, impairment, risk, and treatment. Its diagnostic categories have always changed with evidence, clinical practice, law, and social values. Change is therefore not itself evidence of capture. The question is whether revisions remain transparent, evidence-responsive, open to criticism, and consistently applied.

That claim is now under strain, and the way it fails is instructive. It does not fail through a single dramatic betrayal. It fails through a set of small institutional adjustments, each defensible on its own, which together turn a truth-telling profession into an instrument of the tribe that captured it. Psychiatry is not the worst case, but it fails legibly enough to reveal a transferable pattern. Once you can name the symptoms in one profession, you can find them in any profession whose authority rests on being trusted to tell an inconvenient truth.

The Five Symptoms

Five indicators can justify closer investigation. None is diagnostic alone, and each has an innocent analogue.

Selective redefinition of terms. Diagnostic language gets rewritten in ways that blur or abandon objective definitions. Categories once anchored to observable reality are replaced with socially malleable ones, and — the tell — the same clinical logic is no longer applied across contexts. Some false beliefs become exempt from being called false, depending on how politically sensitive they are.

Policy preceding evidence. Institutional guidelines change in response to activist pressure or legal trends rather than new findings. Line up the shifts on a timeline and there is no underlying scientific breakthrough to explain them — only the arrival of a new political imperative. The guideline moves; the evidence does not.

Asymmetric application of principles. In an ordinary case, a belief that directly contradicts physical reality is treated as a symptom to be evaluated. In a politically protected domain, those same evaluative tools are suspended, softened, or reframed. The asymmetry is the abandonment of value-neutral science, visible in the seam.

A chilling effect on inquiry. Clinicians who question the new orthodoxy risk reputational damage, sanction, or their jobs. Even mild dissent is recast as a moral failing rather than a scientific disagreement, which silences exactly the debate a self-correcting field depends on.

Institutional gatekeeping. Professional bodies fold activist representatives into the drafting of standards, granting political actors direct influence over clinical definitions — and those same bodies begin issuing policy positions on matters that have nothing to do with core competence.

None of the five requires bad faith from any individual. Together they support a capture diagnosis only if the evidence also shows asymmetric treatment, protected conclusions, sanctions disconnected from professional quality, and failed correction or appeal. Legitimate revision can look similar at a distance: definitions change because evidence changes; policy may precede definitive evidence under uncertainty; affected communities may appropriately inform standards. The burden is comparative and institutional, not rhetorical.

There is a longer lineage here than a decade of institutional drift. Psychiatry is the institutional descendant of the Freud meme complex — a profession built, in no small part, on a self-sealing explanatory scheme that taught a century of practitioners to treat resistance as confirmation and doubt as symptom. I trace that inheritance in Echoes of Freud. A discipline that learned early to make its own unfalsifiability a feature was always going to be an easy host for the next totalizing narrative that came along. The symptoms of capture are, in part, the old immune weakness expressing itself in a new pathogen.

Independence Was the Contract

Universities have never stood wholly outside politics; patrons, churches, states, wars, class, and ideology have shaped them for centuries. Their distinctive claim is narrower: methods and results should not change merely because a faction dislikes them, and scholars should retain room to criticize both governing power and their own institutions.

Individual scholars do not abandon that protection by speaking politically. Nor does an institution become captured merely because it studies injustice or adopts a public position. The warning signs appear when assent becomes a condition of professional standing, dissenting work faces different evidentiary rules, institutional speech is presented as disciplinary consensus without a method for measuring it, or political administrators can redirect conclusions through appointments and funding.

This yields a reciprocal law: capture invites counter-capture. Once professional machinery is accepted as a tool for enforcing one orthodoxy, opponents have both an incentive and a precedent for seizing it. That does not justify retaliatory funding cuts or political interference. It explains why durable safeguards must be viewpoint-neutral: transparent criteria, distributed authority, academic freedom, due process, disclosure of conflicts, replication, and appeal. The dynamics of why “our side should capture it first” is a trap appear in The Politics of Safety.

The Category Error in Engineering

The third case is the clarifying one, because engineering is the discipline where the stakes are most physical and the confusion is most precise. Here the capture arrives under the banner of decolonization, and it rests on a single mistake that is worth stating exactly.

Engineering joins universal physical constraints to situated human purposes. A bridge must bear its loads regardless of the designer’s culture, but which bridge is built, whose land it crosses, which risks count, and whose knowledge reveals local conditions are social decisions. The category error can run in either direction: treating physical verification as merely one cultural story, or treating the choice of objectives and acceptable tradeoffs as if physics decided them. Quantification, modeling, testing, and revision earn authority over prediction; they do not settle every question of meaning, priority, or justice.

The confusion has a name, and it is worth carrying out of this chapter: it is the confusion of verification with meaning. A “decolonizing and indigenizing” engineering curriculum places engineering and Indigenous knowledge traditions on the same epistemic plane, as parallel ways of knowing that deserve equal space in the classroom. The posture aims at respect and commits a category error. Engineering evaluates a claim by verifying whether it corresponds to outcomes in the physical world. A cultural tradition evaluates a claim by its meaning within a historical and relational context. These are not interchangeable standards. Meaning explains significance; verification explains behavior. Both matter, and they answer different questions. A curriculum that collapses the two deprives engineering of its essential evaluative framework — the capacity to tell a true prediction from a false one.

The concrete case that surfaced this was a real course, flagged by the engineer Paul Raymond-Robichaud¹ and documented in a published course paper for a Canadian engineering faculty:²

A required course in an accredited engineering program: “Decolonizing and Indigenizing Engineering.” The assessment is reflective journaling, participation in discussions and ceremonies, and a summative reflective project. No modeling. No design exercise. No analysis of a physical system. No testing of a hypothesis. This is an engineering course.

The assessment model is the whole argument. Reflection is genuinely valuable in ethics and in reckoning with the human impact of engineering; it is not a substitute for the cognitive architecture required to produce a design that does not fail. Replacing modeling, analysis, and testing with introspection and narrative processing is not integration. It is skill displacement — swapping the competencies that define the practice for the ones that flatter a worldview. An engineer who reflects well but designs poorly is not an engineer. The bridge does not care how thoughtfully its designer examined his assumptions about power and knowledge; it cares whether the moment calculation was done right.

This is where the Axio framework makes the cost precise. Agency, in this framework, is the ability to construct, evaluate, and choose among the models that shape one’s dealings with reality. Engineering is among the most agency-amplifying domains humans have ever built, because it hands individuals and communities the power to predict and reshape physical systems. But agency depends on epistemic clarity — the ability to test an assumption, the freedom to reject an unfounded story, the discipline to privilege a model that works over a model that merely feels meaningful. When a curriculum prescribes an interpretive frame rather than teaching students how to evaluate and choose among frames, it does not expand agency; it constrains it. Substituting worldview relativism for engineered models is not empowerment. It is epistemic enclosure — the fencing-off of the very capacities that would let a student think past the frame he was handed. Capture, seen from inside the framework, is always an enclosure of agency dressed as a liberation of it.

And the physical stakes give the whole diagnostic its edge. When truth becomes negotiable, safety becomes negotiable. A captured psychiatry produces inconsistent care; a captured academy produces distrust; but a captured engineering produces failures with body counts, because the discipline’s entire value is its indifference to story. It cares only whether a model corresponds to reality. Take that away and you have not decolonized engineering. You have disarmed it.

Respect Without Relativism

None of this is an argument against the history, the ethics, or the community. The critique dismisses nothing about Indigenous history, treaty obligations, environmental stewardship, or culturally informed practice — these are real components of responsible engineering, and an engineer who does not understand the context his design will operate in is a worse engineer. The point is that respect for cultural knowledge does not require flattening epistemic distinctions, treating empirical success as culturally contingent, or importing rituals into professional training.

There is a version of integration that is coherent, and it is worth stating so that the critique is not merely negative. Teach the history of engineering projects and their impact on Indigenous communities: it illuminates real harms and prepares students to practice more responsibly. Analyze cases where disregarding local knowledge caused environmental or social damage: it sharpens ethical judgment without diluting technical standards. Incorporate ecological insights that carry genuine predictive value — traditional knowledge of watersheds, fire regimes, and species behavior often does — into the modeling itself, where it strengthens the engineering rather than replacing it. Build thoughtful consultation and consent into projects, so that engineering respects community autonomy while keeping its methodological spine intact. This is the shape of respect without relativism: learn from another culture without substituting its epistemic foundation for the one that makes the work reliable. Respect the culture, learn the history, engage the community — and keep engineering anchored to truth.

The Vice and the Obsolescence

There is a final turn, and it belongs to economics as much as to memetics. While the professions have been captured from within, they are being made obsolete from without. The academy’s credential and the licensed profession’s seal were always, at bottom, proof-of-work tokens — expensive, hard-to-fake signals that their holder had done the labor and could be trusted to tell the truth in a domain the public could not check directly. That signal is collapsing on two fronts at once. Ideological capture destroys its content, so that the credential no longer certifies neutral competence but membership in a faction. And cheap, capable machine intelligence destroys its scarcity, so that the underlying labor the credential attested to can be had for the price of a query. I develop that second dynamic in Ornament and Advantage; here it is enough to note that the two failures compound. A profession that has spent its neutrality to win a culture war finds, when it turns to defend itself, that the market has stopped believing its signal for entirely separate reasons. Capture-from-within meets obsolescence-from-without, and the profession is squeezed between them.

The three cases are one mechanism. A profession earns its authority by being trusted to tell an inconvenient truth; capture is the process by which a tribe converts that trust into a weapon and, in doing so, spends it. The five symptoms are the instrument for detecting it early — before the redefinitions harden, before the chilling effect empties the room of dissenters, before the gatekeepers finish drafting the new standards. They are a special case of the general anatomy I lay out in Anatomy of Capture, and they feed the larger claim of The Forbidden Pattern: that the institutions we rely on to see clearly can develop, collectively, a blind spot precisely where they most need sight. Psychiatry’s question generalizes to every captured profession. If the people we trust to tell the truth will only tell us the convenient half of it, then the truth has nowhere left to live — and we will have arranged its exile ourselves, one defensible institutional adjustment at a time.