Daniel sent us this one — he's picking up on something we touched on in a previous episode about narcissistic personality disorder, where we mentioned the diagnosis has traditionally been benchmarked against societal expectations. And his question basically is: what happens when those expectations aren't uniform, or worse, when a society or subculture normalizes and rewards the very traits we'd call pathological? Think about a corporate culture that celebrates ruthlessness, or a high-stakes trading floor where empathy gets you eaten alive. If you're diagnosing deviation from norms, but the norms themselves are messed up, what are you actually measuring? And the alternative — prescribing a universal standard of healthy behavior — that has its own problems. So is the whole enterprise of tying diagnosis to social expectations just inherently flawed?
That's the knife-edge this field has been walking for decades, and it's not an academic question — the same behavior that gets you diagnosed with narcissistic personality disorder in a community mental health clinic gets you a corner office and a bonus on Wall Street. So what are we actually diagnosing? The person, or the mismatch between the person and their context?
The DSM-five-TR, the current manual from twenty twenty-two, still hangs its hat on "clinically significant distress or impairment in social, occupational, or other important areas of functioning." That's the gatekeeper criterion for almost everything in there. And every word of that phrase is socially defined — what counts as impairment in a social context depends entirely on what that social context expects.
Right — and the DSM knows this. They've known it. The text explicitly tells clinicians to factor in cultural context when making judgments about impairment. But telling someone to "factor it in" is not the same as giving them a method for doing it. It's a hand-wave dressed up as a guideline.
The thing Daniel's really poking at is whether this is a fixable problem or a structural one — whether the whole approach of tying pathology to social norms is like building a house on sand and then writing a footnote about the sand.
Here's why it matters right now, practically. We're in a moment where the ICD-eleven — the World Health Organization's classification system — has just completed the biggest overhaul of personality disorder diagnosis in forty years. They threw out all the categorical labels — no more narcissistic, borderline, antisocial as separate boxes — and replaced them with a single dimensional diagnosis rated by severity. That went live in twenty twenty-two. Meanwhile the DSM still has one foot in each camp: the main text keeps the old categories, and Section Three has an alternative dimensional model that almost nobody uses in practice. So we're living through the transition Daniel's question anticipates.
The transition from "you have this disorder because you deviate from what we expect" to...
That's what we need to unpack. And the best way into it is to look at the skeleton in the DSM's closet — the moment that made it impossible to pretend diagnosis is just objective science.
Nineteen seventy-three. The APA Board of Trustees votes to remove homosexuality from the DSM. And the thing people forget is that it wasn't a scientific breakthrough that drove that decision — it was activism, protest, and a vote. A straight-up political process.
For a medical diagnosis. That tells you everything about how much social norms were doing the actual work behind the clinical language.
The language before seventy-three was specific — it was listed as a "sociopathic personality disturbance." The pathology was defined entirely by deviation from what society considered normal sexual behavior. There was no claim about distress or impairment that wasn't itself a product of the social context. The "impairment" was being gay in a society that punished it.
Which is the cleanest case study you could ask for. The diagnosis didn't identify something wrong with the person — it identified something wrong with the fit between the person and the surrounding culture. And when the culture shifted, the diagnosis evaporated.
That's the ghost that haunts every personality disorder diagnosis since. Because the question becomes: which of our current diagnoses are the nineteen-seventy-three homosexuality of their era? We don't know, because we're inside the culture that produced them.
Daniel's question lands right here. If a corporate culture rewards grandiosity, exploitation, and lack of empathy — and someone thrives in that environment, gets promoted, makes millions, feels great about their life — are they disordered? The DSM says no, because there's no distress or impairment. But that answer feels like it's dodging something.
It's dodging the possibility that the environment itself is pathological. The DSM is built to diagnose individuals, not systems. But personality disorders are inherently relational — they're about patterns of interaction between a person and their social world. You can't cleanly separate the two.
The central question this episode is really about is: how do you diagnose personality pathology without either just rubber-stamping whatever norms happen to be dominant, or imposing some universal standard of healthy personhood that might just be your own culture wearing a lab coat?
The field has actually been trying to answer that. The alternative model in DSM-five Section Three — the Level of Personality Functioning Scale — is the most serious attempt to solve this from inside the diagnostic framework. Instead of asking "does this person deviate from what we expect," it asks "can this person form a stable sense of self and sustain meaningful relationships." Those are meant to be capacities that matter regardless of culture.
The idea being that empathy and identity aren't just Western values — they're functional requirements for being a human among other humans.
That's the claim. Whether it holds up when you actually take it across cultures — that's where it gets interesting.
Let's trace the actual mechanism here. When the DSM says "clinically significant distress or impairment," it sounds like a safety valve — if the person's fine and their life's working, no diagnosis. But "impairment" is doing an enormous amount of hidden work. Who defines it? The clinician, who was trained in a particular culture, using norms they may not even recognize as culturally specific.
The narcissistic CEO is the perfect stress test. They're not distressed — they feel fantastic. Their board isn't distressed — the share price is up. Their employees might be miserable, but the DSM doesn't diagnose based on third-party suffering alone. So by the manual's own logic, this person who meets every trait criterion for NPD... doesn't have a disorder.
Robert Hare, the psychologist who basically wrote the book on psychopathy, explicitly described this as the "successful psychopath" problem. Someone with the full suite of traits — grandiosity, callousness, manipulativeness, lack of empathy — who functions beautifully in certain professional niches. They're not just undiagnosed. They're undiagnosable under the current framework, because the framework outsources the judgment to the environment, and the environment is giving them a standing ovation.
Which means the diagnosis isn't measuring the personality pattern. It's measuring the friction between the pattern and the context. Same person, same traits, drop them into a different workplace or a different culture, and suddenly they're pathological. That's not a medical fact. That's a relational one.
This is where the DSM's structural problem gets really uncomfortable. The manual was built on a medical model — disorders are things people have, like a virus or a broken bone. But personality disorders don't work that way. They're patterns of relating. You can't diagnose the pattern without diagnosing the relationship, and the relationship includes the society.
Why does the social-norm approach persist? If everyone can see this problem, why is it still the backbone of the system?
Partly inertia — the categorical model is what clinicians were trained on, what insurance codes require, what research has been built around for decades. But there's also a deeper reason. The alternative is genuinely harder. If you can't lean on "this person deviates from what we expect," you have to define what healthy personality functioning actually is, in a way that works across cultures. That's a monumental ask.
Enter Section Three.
So the DSM-five included an alternative model — the AMPD, Alternative Model for Personality Disorders — tucked away in Section Three, which is the "emerging measures and models" section. It's officially recognized but not the primary framework. And its centerpiece is the Level of Personality Functioning Scale, the LPFS.
Which tries to solve exactly Daniel's problem. Instead of asking "does this person fit in," it asks "how well does this person's internal machinery work.
Self-functioning — that's identity and self-direction. Can you maintain a stable, coherent sense of who you are? Can you set and pursue meaningful goals? And interpersonal functioning — empathy and intimacy. Can you understand other people's perspectives? Can you form and sustain close relationships?
The shift is subtle but profound. Those aren't cultural values — they're capacities. The claim is that every human being, regardless of culture, needs some version of these things to function. The content of a stable identity might look different in Tokyo versus Texas, but the capacity to form one is universal.
They're rated on a five-point continuum from zero — healthy — to four — extreme impairment. So you're not checking a box that says "narcissist" or "borderline." You're describing where someone falls on these fundamental human capacities. The diagnosis becomes dimensional rather than categorical.
Which means you can be impaired in specific ways without the whole thing hinging on whether your behavior matches or clashes with local norms. A narcissistic CEO might score high on self-direction — they're goal-oriented, they pursue things — but low on empathy and intimacy. That pattern shows up regardless of whether the board loves them.
Here's the key mechanism. The LPFS tries to anchor pathology in something the field calls "mental representations of self and others." Can you hold a stable, nuanced, integrated picture of who you are and who other people are? Or is your self-image wildly inflated and brittle, and your view of other people flattened into tools or threats? That's not a cultural judgment. That's a cognitive and emotional capacity.
It's the difference between saying "you don't act the way we expect" and saying "your basic equipment for being a person among persons isn't working properly.
That's a much harder case to dismiss as cultural imperialism. Not impossible — we'll get to the cross-cultural problems — but it's a genuine attempt to find ground that isn't just "whatever the dominant group thinks is normal.
The ICD-eleven actually went further and made the dimensional approach mandatory, but the DSM kept it optional, in the back of the book.
Section Three is the DSM's "we know this is better but we're not ready to commit" section. It's like keeping your old car in the driveway while the new one sits in the garage with a full tank.
A very psychiatric image.
I contain multitudes. But the serious point is that the LPFS represents a real philosophical shift. It's saying: pathology isn't about fitting in. It's about whether your personality structure allows you to do the basic work of being human — know yourself, direct yourself, understand others, connect with others. If those capacities are compromised, it doesn't matter how well your environment rewards the surface-level behaviors.
Here's where it gets uncomfortable. You take this beautiful framework — universal human capacities, stable self-representation, empathy — and you test it across actual cultures, and the numbers get weird fast. NPD prevalence studies show rates from basically zero percent in some collectivist societies to significantly higher in individualistic Western contexts. Are we measuring a real difference in personality pathology, or are we measuring how well the diagnostic lens travels?
The honest answer is we don't fully know, and measurement invariance is the technical term for the nightmare underneath this. When a clinician in Tokyo administers a diagnostic interview for NPD, they're asking questions that were developed and validated largely on American and European populations. The behaviors that signal grandiosity in one culture might signal something completely different in another. Bragging about personal achievements reads very differently in a culture that values self-effacement versus one that values self-promotion.
The zero percent figure might not mean there's no narcissism in that culture. It might mean the diagnostic instrument can't see it, or that it manifests differently enough that the checklist doesn't catch it.
And this cuts both ways. Some researchers argue that what looks like narcissism in an individualistic culture is partly just... the culture's script for success being performed enthusiastically. You're not disordered, you're just really good at being American.
Which brings us to the pathological culture problem Daniel raised directly. What about subcultures that actively select for and reward traits we'd call pathological in the general population?
High-frequency trading firms are the textbook example. They filter for extreme risk tolerance, low empathy, rapid decision-making with low regard for consequence, and a kind of emotional flatness under pressure. Those are traits that, in a different context, would map neatly onto antisocial or narcissistic patterns. But inside the firm, they're not just adaptive — they're the job description.
The diagnosis becomes context-dependent to the point of meaninglessness. Same person walks into a therapist's office — if the therapist doesn't understand the subculture, they see pathology. If they do understand it, they see a highly functional specialist. Neither view is wrong, and that's the problem.
The military makes this even sharper. A soldier who is hypervigilant, emotionally constricted, aggressive when threatened, and deeply suspicious of unfamiliar environments — that's adaptive in combat. It keeps you alive. Bring that same pattern back to civilian life, and you're checking boxes for PTSD, paranoid personality traits, or both. The DSM tries to handle this with the pervasiveness criterion — the pattern has to show up across contexts — but what counts as "across contexts" when someone spends years in a single context that reshapes their entire personality architecture?
Then there's the hikikomori case in Japan — severe social withdrawal, sometimes for years. It's pathologized within Japanese society because the social expectation around engagement and contribution is so strong. But you drop that same behavior into a culture with different expectations around social participation, and does it look the same? Is the pathology in the withdrawal, or in the mismatch between the withdrawal and what the surrounding culture demands?
This is where the knock-on effect gets dangerous. If you're diagnosing deviation from norms, and the norms happen to be whatever the dominant group prefers, then diagnosis becomes a tool for enforcing conformity. The dissenter, the eccentric, the person who refuses to play the social game — they're not just difficult, they're sick. And we have a long, ugly history of exactly that.
The old Soviet practice of diagnosing political dissidents with "sluggish schizophrenia" — a diagnosis that existed almost exclusively to pathologize nonconformity. That's the extreme end of the spectrum, but it's the same logic. Define healthy as "fits in," and anyone who doesn't fit in is by definition unhealthy.
Is there a way out of this? Because the extremes are both bad — either you're imposing your cultural norms on everyone, or you're saying everything is normal in its context and diagnosis is impossible.
There's a framework I think gets closer than anything else, and it comes from outside psychiatry entirely. Amartya Sen's capabilities approach — the idea that what matters for human well-being isn't resources or happiness or conformity, but whether a person can do and be what they have reason to value.
Which we touched on in a different context — the Human Development Index episode — but it applies here almost perfectly. Instead of asking "does this person fit into their society," you ask "does this personality pattern prevent them from living a life they have reason to value?
That shifts the anchor from social conformity to individual flourishing. It doesn't matter whether the trading floor rewards your callousness. What matters is whether that callousness prevents you from having the relationships, the self-understanding, the range of emotional experience that you yourself would want if you could step outside the pattern.
The clinician's job, on this view, isn't to judge whether someone matches or deviates from local norms. It's to help the person see whether their personality structure is foreclosing possibilities they would otherwise value. The person who doesn't want close relationships and is content with that — that's not a disorder, even if it looks unusual to the clinician. The person who desperately wants connection but keeps destroying every relationship because they can't regulate their grandiosity — that's where the pathology lives, regardless of whether their job rewards the surface behavior.
Practically, for clinicians, this means something specific. You don't just ask "does this person deviate from norms." You ask three things: Is this pattern causing harm to the person or others? Is it inflexible — does it show up regardless of circumstances? And is it pervasive — does it cut across multiple domains of life? If the answer to all three is yes, you've got something that matters, regardless of whether the local culture celebrates the surface presentation.
The ICD-eleven's dimensional model embeds exactly this logic. Severity is determined by the degree of harm and impairment, not by which flavor of personality pattern you have. The specific traits — grandiosity, detachment, disinhibition — are descriptors, not the basis of the diagnosis itself. It's a genuine attempt to escape the culture trap without falling into the everything-is-relative trap.
The middle path is hard. It requires clinicians to know enough about a patient's cultural context to distinguish between what's adaptive within that context and what's impairing across contexts. That's a lot to ask of a fifteen-minute diagnostic interview.
It's a lot to ask of any human being. But the alternative is worse — either you're a cultural imperialist in a white coat, or you've abandoned diagnosis entirely and left suffering people without a framework for understanding what's happening to them.
What do we actually do with all of this? Because it's easy to get stuck in the critique and forget that real people are sitting across from clinicians, in real distress, needing something useful.
The social-norm approach isn't wrong, exactly. It captures something real — the fact that personality pathology almost always shows up as friction between a person and their world. That friction is often the first thing that brings someone into treatment. You can't just dismiss it.
The problem isn't that we notice deviation from norms. The problem is stopping there. If all you've got is "this person doesn't act the way we expect," you haven't actually done the diagnostic work. You've just described a mismatch. The real question is whether that mismatch is causing harm, whether it's rigid and inflexible, and whether it shows up across enough contexts that it's clearly about the person's internal structure and not just a bad fit with one environment.
That's the practical takeaway for anyone listening who's either a clinician or a patient navigating this system. Don't let the diagnosis begin and end with "does this person fit in." Push to the next layer. Is this pattern causing suffering — to the person, to people around them? Is it the same pattern regardless of where they are and who they're with? If it only shows up at work and nowhere else, that's not a personality disorder — that's a toxic workplace.
The pervasiveness criterion is doing real work here, and it's the thing most pop psychology coverage misses entirely. A personality disorder isn't situational. It travels with the person. The CEO who's ruthless in the boardroom but a warm, connected parent at home — that's not NPD, that's a person with a very specific professional persona. The diagnosis is for when the ruthlessness is the only setting they've got.
The dimensional approach — the LPFS and the ICD-eleven severity model — is a genuine advance on exactly this front. It forces you to look at the architecture of the person rather than the fit with the context. But here's the uncomfortable practical reality: almost nobody uses it.
It's wildly under-implemented. Most clinicians still use the old categorical boxes because insurance forms want a code, referral letters want a label, and the entire administrative machinery of mental health was built around the idea that you either have a disorder or you don't. Dimensional assessment takes longer, requires more training, and produces a more nuanced picture that doesn't fit neatly into a checkbox.
Which means we're in this strange transitional moment where the science has moved, the WHO has moved, and clinical practice is still catching up. The ICD-eleven went live in twenty twenty-two and made the dimensional model mandatory for member states. But implementation takes years, sometimes decades. There are clinicians practicing today who were trained before the DSM-five even existed.
That's the meta-lesson Daniel's question really points toward. Diagnosis is never just science. It's a negotiation between what we can measure, what our institutions can process, what our culture will accept, and who has the power to define the categories in the first place. The nineteen seventy-three vote on homosexuality wasn't an aberration — it was the whole enterprise laid bare.
Being aware of that doesn't make diagnosis impossible. It makes it harder. And that's the price of doing it responsibly. You don't get to pretend you're just reading out objective biological facts. You have to hold the cultural contingency of your categories in mind while still making judgments that matter for real people's lives. That's uncomfortable, and it should be.
The alternative — pretending the cultural problem doesn't exist, or swinging to the other extreme and saying nothing can be diagnosed across cultural lines — both of those are easier. And both of them fail the people who actually need help.
That tension — between what we can measure and what we're actually trying to understand — is about to get a whole new set of tools thrown at it. Computational psychiatry is moving fast. The pitch is that we'll eventually ground diagnosis in neurobiology — biomarkers, brain scans, machine learning models trained on thousands of patients. The hope being we can bypass the culture problem entirely by pointing at something physical.
But you and I both know what happens when you train an algorithm on data that already encodes cultural bias. You don't escape the bias. You automate it, scale it, and give it a sheen of scientific objectivity that makes it even harder to challenge.
A diagnostic AI trained on American psychiatric data is going to learn that American cultural patterns are the baseline for healthy. It won't know it's doing that. It'll just output confidence scores that look wonderfully precise while reproducing every assumption baked into the training set.
The danger is that the precision becomes its own authority. A clinician can be challenged — "you're imposing your cultural assumptions on me." An algorithm with a ninety-four percent confidence score is harder to argue with, even if that ninety-four percent is measuring conformity to norms that were never universal to begin with.
The question isn't whether we can build the technology. We can, and we will. The question is whether we'll be honest about what it's actually measuring.
In the meantime, the ICD-eleven's dimensional model is the biggest structural change in personality disorder diagnosis in four decades, and we're living through the implementation right now. That's not a future hypothetical — that's the present. The categories are gone, severity is the backbone, and the whole thing is designed to be more culturally portable than the old system ever was.
Whether it actually delivers on that promise is going to depend less on the manual and more on the people using it. Which is probably exactly where the responsibility should be.
Now: Hilbert's daily fun fact.
Hilbert: In the Seychelles, a traditional coronation ceremony for island chiefs once required the new ruler to be publicly weighed against a stack of coconuts equal to the population of their district — one coconut per subject. If the chief weighed less than the coconuts, the coronation was void.
...so legitimacy was literally a matter of mass.
I have so many questions about the coconut-to-subject exchange rate. But I'm going to sit with them silently.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. If you want to dig deeper into any of this, the dimensional model is publicly available in the ICD-eleven clinical descriptions — it's worth reading the primary source. And if you enjoyed the episode, leave us a review wherever you listen. We're back next week.
