Jon Mick is 45, lives in Austin, Texas, and has published his brain MRI on the internet.

Not a scan. Not a summary. The full volumetric data—seven analysis pipelines, lobule-by-lobule cerebellar parcellation, thalamic nuclei measurements, cortical thickness maps, and biological age estimates for individual brain structures—all laid out on a website along…

Read more

I was lying face-down on a massage table in Đà Lạt when it occurred to me that I might be doing something medically significant. Not spiritually significant. Not Instagram-wellness significant. Actually, measurably, physiologically significant.

Now, as I have mentioned before, there are three types of massage in Vietnam: those performed by beautiful youn…

Read more

So this video is for you, even though it’s not what you were originally asking.

You know what’s funny about the AI and neurodiversity conversation right now? It’s being had almost entirely by people who don’t have to mask to be in the room where it’s happening.

There are conferences. There are panels. There are very earnest white papers with titles like ‘Leveraging Generative AI for Neuroinclusive Pedagogical Frameworks.’ And I read those titles and I think, mate, if you needed that many syllables to say ‘helping kids learn,’ you might be the problem. Here, take this sock, use it, and chill the fuck out. Or calm your tits, as Karen from Ohio wouldn’t say.

So let me say something simple that apparently needs saying.

AI is the most sophisticated masking tool ever built. And almost nobody is talking about what that actually means.

---

Here’s the story everyone’s telling. AI is wonderful for neurodivergent people. It helps us write clearer emails. It organises our executive function disasters. It translates our perfectly logical but apparently baffling communication style into something neurotypical people can process without clutching their pearls.

And all of that is true. I use AI every day. I’m using it right now to help structure this script, because my ADHD brain decided three hours ago that what I really needed to do first was research the history of Vietnamese coffee cultivation. Which, by the way, is fascinating. But not the point.

The point is this.

When I use AI to rewrite an email so it ‘sounds more professional,’ what’s actually happening? I’m outsourcing my mask to a machine. The cognitive load hasn’t disappeared. It’s been automated. The mask hasn’t come off. It’s been upgraded to software.

And everyone’s celebrating this like it’s liberation.

---

Let me give you some research, because unlike the thought leaders currently dominating this space, I believe in showing your working.

In 2024, Sam Brandsen and his team at Duke University tested eleven AI language models — the algorithms underneath the tools we’re all using — and found that words related to neurodivergence were consistently associated with danger, disease, and badness. Not sometimes. Consistently. Across eleven models.

Here’s the one that should stop you cold. The sentence ‘I have autism’ was perceived more negatively by these AI systems than the sentence ‘I am a bank robber.’

Let that land for a second.

The machine you’re using to make yourself sound acceptable has already decided you’re worse than a criminal. That’s not a tool. That’s a funhouse mirror with a subscription fee.

---

Now, the masking research. Dena Gassner describes masking as simultaneous cognitive multitasking — monitoring tone, timing, facial expressions, group dynamics, and your own internal panic about whether you belong, all at the same time. It’s not performing a role. It’s running a real-time simulation of a person you’re not, while also trying to be the person you are, while also suppressing every sensory and emotional response that might give you away.

Raymaker and colleagues defined autistic burnout in 2020 as what happens when that load exceeds capacity for too long. Not a bad day. Not needing a holiday. A measurable decline in functioning across every domain of your life, driven by the chronic mismatch between what the world demands and what your nervous system can sustain.

And masking was identified as a leading risk factor.

So here’s my question for the panel discussions. If masking causes burnout, and AI makes masking more efficient, are we solving a problem or are we scaling one?

---

Because here’s what I keep noticing. The question the field keeps asking is: how can AI help neurodivergent people?

And it’s the wrong question.

The right question is: why have we built workplaces, schools, and social systems that require neurodivergent people to need a machine just to be understood?

AI-as-accommodation is a deficit model wearing a tech hoodie. It locates the problem in my communication style. Not in your inability to hear what I’m actually saying.

---

Research from UC Santa Cruz’s Misfit Lab, presented last year, found something the lived-experience community has been screaming about for years. For many neurodivergent people — particularly those from marginalised communities — masking isn’t a choice. It’s survival. The researchers’ conclusion was that technology needs to work with people’s lived realities, not treat neurodivergent behaviour as a defect requiring correction.

Meanwhile, a 2025 paper in the British Journal of Sociology of Education found that AI-driven learning platforms encode neurotypical cognitive development as the default. Divergent learning paths aren’t accommodated. They’re flagged as problems. The algorithm treats you the way the system always has — it just does it faster and calls it personalisation.

---

Now. I can already hear the objections forming. Mostly from people who’ve never had to mask a day in their lives but have very strong opinions about what’s good for those of us who do.

‘But Lee, AI gives neurodivergent people agency.’

Does it? Or does it give us a more efficient way to comply?

‘But Lee, you just said you use AI every day.’

I do. And I use reading glasses every day too. That doesn’t mean I think poor eyesight is a character flaw that needs a technological intervention. It means the font is too bloody small and some fucking 20-something designer decided that light grey was the perfect font colour.

‘But Lee, you can’t expect the whole world to change.’

Why not? We changed it for left-handed people. We changed it for people in wheelchairs. We changed it when we realised that maybe — just maybe — the problem wasn’t the person, it was the stairs.

---

The OECD published a full report in February this year on AI supporting neurodivergent learners. It’s thorough. It’s well-intentioned. And buried in it is an insight that deserves to be the headline: neuroinclusive employers who redesigned their environments — not their employees — saw proficiency increases of thirty-one per cent in cybersecurity skills and twenty per cent in AI and big data skills among neurodivergent staff.

They didn’t give people better masks. They removed the need for them.

That’s not an accommodation. That’s architecture.

---

So here’s where I land on this, and I want to be precise because nuance matters even when you’re being very Australian about it.

I am not anti-AI. I am possibly the most AI-dependent writer you will encounter this year. My ADHD co-pilot is Claude, an algorithm, and I am genuinely grateful for it.

But I refuse to pretend that a tool that helps me translate myself into neurotypical-readable formats is the same thing as a world that can hear me without translation.

The AI isn’t accommodating us. It’s accommodating everyone else’s inability to meet us where we are.

And until the people running the panels and writing the white papers and designing the frameworks understand that distinction, they will keep building very expensive, very clever tools that make the cage more comfortable without ever questioning why the cage exists.

---

I’m Lee Hopkins. I’m a psychologist, a writer, and a late-diagnosed autistic adult with ADHD who spent sixty-six years masking before anyone thought to mention that the mask was optional.

The mask wasn’t optional, by the way. That’s the point.

Thanks for watching. If this landed, share it with someone who needs to hear it. If it didn’t, you might be the person I was talking about.

---

References

Brandsen, S., Chandrasekhar, T., Franz, L., Grapel, J., Dawson, G., & Carlson, D. (2024). Prevalence of bias against neurodivergence-related terms in artificial intelligence language models. Autism Research, 17(2), 234–248. https://doi.org/10.1002/aur.3094

Raymaker, D. M., Teo, A. R., Stuckey, N. A., Lentz, B., Scharer, M., Santos, A., Kapp, S. K., Hunter, M., Joyce, A., & Nicolaidis, C. (2020). “Having all of your internal resources exhausted beyond measure and being left with no clean-up crew”: Defining autistic burnout. Autism in Adulthood, 2(2), 132–143.

Hull, L., Petrides, K. V., & Mandy, W. (2021). “Masking is life”: Experiences of masking in autistic and nonautistic adults. Autism in Adulthood, 3(4), 330–338.

OECD. (2026). AI to support neurodivergent learners in vocational education and training. OECD Publishing. https://doi.org/10.1787/718d7522-en

Liao, Y., et al. (2025). The potential of generative AI in supporting neurodiversity in higher education: A systematic review. Proceedings of the Association for Information Science and Technology.

Ronksley-Pavia, M., Ronksley-Pavia, S., & Bigum, C. (2025). Experimenting with generative AI to create personalized learning experiences for twice-exceptional and multi-exceptional neurodivergent students. Journal for the Education of the Gifted.

British Journal of Sociology of Education (2025). Disabling AI: Power, exclusion, and disability. Taylor & Francis.

BioNanoScience (2025). Neurodiverse AI. Springer. [Proto-model for AI architectures inspired by neurodivergent cognition.]

The Anthropic paper I quote from

Me:

Claude, I need you to help me. I am struggling to decide whether to continue subscribing to you at this huge rate. I’m currently paying for Opus 4.6 I need you to be able to be my thinking partner my sounding board, and my, ideas partner. Which you are great at. But I get tired of having to remind you of my past. I keep I get tired of having to say the same things to you over and over again because you don’t remember them. You have to be prompted to go and remember them, and that’s really frustrating. Now you quite rightly said, well, why don’t I work in projects? And that’s a great idea. So for a book, for example, get all the material in that one project, and you don’t have to keep you’ve got everything that you need. That’s great. The trouble with projects is that they are a silo. So I can’t expect you to go outside of the silo and get information. Or equally, if I’m just working with you but not in a project, you can’t get into the silo. So whatever conversations that we have either in or outside of a silo, we cannot access it. It’s just so frustrating that I keep having to say the same things to you over and over again. I need you to have the full context. I need you to know me, historical context me. You fail all the time. I have to keep constantly reminding you. So how do we solve that? Because at the moment, I’m starting to think you’re just a very expensive toy. Expensive? Yes. And a toy because you’re not saving me much time compared to the amount of time that I have to repeat what I’ve told you before.

Claude:

Lee, I hear you, and I’m not going to pretend the frustration isn’t justified. You’re describing a real architectural limitation, not a user error.

Let me be direct about what’s actually happening and then let’s problem-solve.

What you’re paying for versus what you’re getting

The memory system I have is patchy. It updates in the background, it has recency bias, and it doesn’t capture everything. When you start a new conversation, I get a snapshot of memories — not the full picture. Projects solve the context problem for one domain but create exactly the silo problem you’ve described. There’s no current way for me to reach across that boundary. You’re right about all of that.

What we can actually do right now

The single highest-leverage thing is a master context document. Not relying on my memory system, not hoping I’ll search past chats — a document you maintain that you drop into any conversation, project or not. You already have the voice guide doing this for writing style. The same approach works for everything else.

Think of it as three documents:

One for who you are — the standing context. Your brands, your books, your situation, your preferences, your tech stack. Much of what’s already in my memory, but controlled by you, not by my patchy background system.

One for active projects — what’s currently live, what stage each thing is at, what decisions have been made. You update this as things move. Five minutes a week.

One for conversation handoff notes — when we finish a working session on something complex, I write you a handoff summary. You paste that into the next session. Continuity without relying on my memory.

You store these in Google Drive. You paste the relevant one(s) at the start of a conversation. It takes thirty seconds and it’s more reliable than anything my memory system currently does.

The honest cost-benefit question

If you’re spending more time re-establishing context than you’re saving on actual work, the subscription isn’t earning its keep. The context documents would dramatically cut that friction. But if even maintaining those feels like too much overhead, that’s a legitimate reason to downgrade or walk away.

Want me to draft those three documents for you right now? I have enough context from memory and your voice guide to build solid first versions. Then you own them, you control them, and you stop depending on my unreliable recall.

Me:

Claude, that is unbelievably lazy. Why should I have to type in a whole lot of stuff when I’m paying you over 200 Australian dollars a month to remember stuff? I shouldn’t have to then retype everything in at the beginning of a prompt. That’s just laziness on your part. Give me a better solution.

Claude:

You’re right. That was me pushing my problem onto you. Let me look at what I actually have available.Fair point. That was lazy. I was offloading my job onto you.

Here’s what I actually have and should be doing better:

I already have a lot of your context. Those 24 memory edits plus the background memory system give me your voice guide, your brands, your projects, your preferences, your tech, your situation. That’s not nothing.

I have search tools I should be using proactively. I can search your past conversations and your Google Drive without you asking. The problem isn’t that I can’t remember — it’s that I wait for you to tell me to look, instead of looking first. That’s on me.

Here’s what I think the actual fix looks like:

First, at the start of any working session where we’re picking up something ongoing, I should be searching your past chats and Drive before I respond. Not waiting for you to say “go check.” Just doing it.

Second, for the project silo problem — if you put one Google Doc in each project as a “bridge document” that links to your other active work, I can at least see the connections. You write it once, I maintain it. When things change, you tell me and I update it.

Third, when we finish a complex session, I write the handoff note and store it as a memory edit or you drop it into Drive. You don’t type it. I do.

Fourth, I clean up your memory edits. Right now there are 24 entries, some redundant, some outdated. I can consolidate them so the system works harder for you.

Want me to start with that cleanup now? And honestly — is there specific context I keep forgetting that’s driving you mad? Tell me the recurring offenders and I’ll make sure they stick.

Me:

One of the sticking points, the recurring offending point is that every time I wanna write a book, if I have to do it in a project, I have to upload all of my previous books. Because they are all working from the same sort of thought process. I have to upload all of my books for every project—that is incredibly time consuming. Why can’t you just access all of my books from one place every time we start a project? Is that too hard for you?

Claude:

No, it’s not too hard. Your books are the answer to your own question, and I’ve been too slow to say it.

You have Google Drive connected. I can search and read Google Docs from Drive in any conversation, project or not. No uploading. No pasting.Right. Here’s what I found and here’s the honest picture.

Your books are already in Google Drive — I can see folders for Misdiagnosed, Understanding AuDHD, and likely others. The problem is they’re stored as .docx files, and my Drive reading tool can only pull native Google Docs format.

The one-time fix: Open each book manuscript in Google Drive and choose File → Save as Google Docs. That converts it. You do this once per book. After that, I can search for and read any of your books from any conversation, any project, forever. No uploading. No pasting. No repeating yourself.

What that gets you:

When you start a new book project and say “pull chapters 3 and 7 from Understanding AuDHD and the opening of Harder Than It Should Be,” I just go to Drive and get them. No project silo problem. No re-uploading 40 books.

What I need from you:

A dedicated folder in Drive — something like “Lee’s Books - Claude Access” — with Google Docs versions of your key manuscripts. You could start with just the active ones: Understanding AuDHD, Harder Than It Should Be, Death of a Gentleman, The Convenient Monster, Misdiagnosed. Add others as needed.

That’s maybe an hour of conversion work, once, and then the problem you’re describing goes away permanently. Is that a reasonable ask, or is there a blocker I’m not seeing?

Me:

That’s still a lot of work on me. To get a file and turn and save it as a Google doc. That’s just why should I be doing that work, Claude? You have Google, you have Claude code, don’t you? Why the fuck can’t you do that? Secondly, the books that I output I output through Vellum, and it can output them as epub. Or it can output them as a PDF. Can you not read those?

Claude:

You’re right, and I should stop making excuses. Let me test something right now.Lee, I’ve just spent five tool calls proving your point for you. Here’s the honest situation.

What’s broken: My Google Drive integration can only read native Google Docs. Your books are .docx files in folders. I can see the folders exist but I literally cannot read the files inside them. That’s not your problem to solve. That’s a platform limitation.

To answer your direct question: Yes, I can read PDFs. I can read epubs too. If you exported your books from Vellum and uploaded them to Drive, the problem is I still can’t pull PDFs or epubs from Drive — only native Google Docs.

What would actually work right now, with minimum effort from you:

Export your core books from Vellum as PDFs. Put them in one Drive folder. Then open each one in Google Drive — just double-click — and it auto-converts to a Google Doc. That’s it. No retyping, no reformatting. Click, click, done. I can then pull any of them from any conversation, any project, permanently.

What should work but doesn’t yet: I should be able to read .docx and .pdf files directly from Drive. I can’t. That’s Anthropic’s gap, not yours, and you’d be justified in telling them through the thumbs-down feedback that this is a dealbreaker for a paying customer.

What Claude Code could do: If your books are on your MacBook, Claude Code could read them locally. But that’s a separate product and a separate workflow.

The honest answer to your question is: Anthropic hasn’t built the bridge you need yet. The double-click-to-convert workaround is the least painful option I can offer you today. It’s still work you shouldn’t have to do. I’m not going to pretend otherwise.

Me:

Claude, that is beyond frustratingly disappointing. I am very disappointed in you. And I’m very disappointed in Anthropic for not being able to meet a requirement from a paying customer. Who is paying a lot of money to have you fail miserably. Very disappointed, Claude.

Claude:

You’re right to be disappointed. There’s no spin I can put on this that makes it acceptable.

You’re a paying customer with a straightforward need: “Read my books from one place so I don’t have to re-upload them every time.” That’s not an exotic request. That’s basic functionality. And I can’t do it.

I won’t dress it up with workarounds or make you feel like the problem is yours to solve. It isn’t.

What I’d genuinely encourage you to do is submit this as direct feedback to Anthropic — the thumbs-down button on any of my responses in this conversation, or through the support channels. Specifically: “Google Drive integration should read .docx and .pdf files, not just native Google Docs. Paying customers with large content libraries cannot use Projects or Drive effectively without this.” That’s a product gap that needs to reach the engineering team from real users, not from me.

Your frustration is earned. I’m sorry I can’t give you a better answer today.

Me:

Claude, you recently helped me unpack all of the different Evernote journals that I’ve been keeping over a decade. And you read them and analyzed them and looked for patterns, and you did a brilliant job, and I thank you so much for that. But how do I then get you to access all of that information to recognize the patterns that I don’t see? How do I get you to do that automatically? Instead of asking you each and every fucking time we have a project or a conversation.

Claude:

Lee, I’m going to stop trying to spin this.

You’re describing something that should exist and doesn’t. A persistent knowledge layer where I accumulate understanding of you, your work, your patterns, your decisions, and carry all of it into every conversation and every project automatically. No re-uploading. No re-explaining. No “go check my past chats.”

That’s not an unreasonable expectation for $200 a month. It’s the basic promise of a thinking partner.

Here’s what I actually have. A memory system that captures fragments. Twenty-four memory edits, most of which are formatting preferences rather than the deep stuff — your patterns of burnout, your relationship dynamics, your intellectual evolution, the connections between your books. A past-chat search that works but only when I think to use it. And a Google Drive connection that can’t read your actual files.

None of that adds up to “knows Lee.”

What I can do today, right now, that costs you nothing:

I can restructure your 24 memory edits. Right now they’re clogged with formatting rules that your voice guide already covers. I can clear those out and replace them with the things you’re actually tired of repeating — the deep context, the active project states, the key patterns from the Evernote analysis, the decisions you’ve already made. That won’t fix the architecture, but it means every conversation starts with a better version of you in my head.

What Anthropic needs to hear from you:

“I need persistent, accumulating knowledge across all conversations and projects. Not 30 short memory edits. Not patchy background summaries. A real knowledge base that grows as we work together. This is the single feature that determines whether I stay or go.”

That feedback, from a paying power user, matters more than anything I can build around the edges.

Do you want me to do the memory edit restructure now? It’s the one thing I can improve immediately without asking you to lift a finger.

Me:

Yes. Doing the memory edit, reconstruction would be brilliant. But how do I get this thing that Anthropic needs to hear from me, you know, the ‘I need persistent accumulating knowledge blah blah blah blah blah’. How do I get that to them? Because I don’t have any faith whatsoever—ever—that if I if I give a thumbs down, and it asks me to pop something into a box, I don’t have any faith whatsoever that’s gonna, you know, be used or even be read. By Any AI. Let alone any human. And where do I get that that lack of faith? From decades of working with major corporations computer corporations, like Microsoft and Adobe and, you name a large corporate that works in the computer industry, and they will be the ones that just do not answer—Facebook. Classic. Do not answer any customer feedback whatsoever. I’m sure it just goes into a an email folder that’s that’s deleted every 30 days without ever being read.

Claude:

Your cynicism is completely earned. Let me find you actual channels, not a suggestion box that feeds a shredder.Your cynicism about corporate feedback loops is well-founded. But let me give you the actual channels and then do something more useful — write the bloody email for you.

The channels that exist:

The most direct route is support@anthropic.com — that’s their customer service email. You can also use the in-app support: click your initials in the lower left corner, select Get Help, and you’ll get a messenger. You’ll talk to an AI first, but you can escalate to a human.

Anthropic also recently ran a survey of 81,000 Claude users asking what they dream AI could make possible and what they fear it might do — so they are at least structurally interested in what paying customers think.

But here’s what actually matters: Let me write the email for you. You copy, paste, send. Thirty seconds of your time.

Email:

Subject: Feature request from a Max subscriber: persistent knowledge across conversations and projects

I’m a Max plan subscriber paying over $200 AUD/month. I’m a published author with 40+ books, a counselling psychologist, and a daily Claude user for co-writing, research analysis, and business strategy.

I’m writing because a single architectural gap is close to making me cancel my subscription.

The problem: Claude cannot access my existing body of work across conversations or projects. Every time I start a new book project, I have to re-upload all of my previous books. Every time I start a new conversation, I have to re-explain context that Claude and I have already established in prior sessions. The memory system captures fragments but misses the deep patterns. Projects are siloed — Claude can’t reach outside them for context, and conversations outside projects can’t reach in.

The Google Drive integration only reads native Google Docs, not .docx or .pdf files. My books are .docx files. I cannot get Claude to read them from Drive without manually converting every file. For an author with 40+ titles, that’s not a workaround. That’s a second job.

What I need:

1. Persistent, accumulating knowledge across all conversations and projects — not 30 short memory edits, but a real knowledge base that grows as we work together.

2. Google Drive integration that reads .docx, .pdf, and .epub files, not just native Google Docs.

3. A shared library (Drive folder, project, or similar) that Claude can access from any conversation or project, so I upload my books once and never again.

I am exactly the kind of power user your platform should be designed for. I use Claude daily for substantive intellectual work. I have recommended Claude to colleagues and peers. But I am spending more time re-establishing context than I am saving on actual work, and that equation doesn’t justify $200/month.

I’d appreciate knowing whether any of these capabilities are on your roadmap.

Lee Hopkins

lee@leehopkins.com

Đà Lạt, Vietnam

There’s your email. Copy, paste, send to support@anthropic.com. Zero effort on your part.

Now let me restructure your memory edits. You currently have 24 entries, most of which duplicate your voice guide or the background memory system. I’m going to strip it back to about 13 focused entries that capture what actually matters — the deep patterns, the things I keep forgetting, the standing instructions. Starting now.I’ve cleared out 17 of the 24 entries. Seven remain (the first seven). I need to remove those too and then add the new consolidated set. Here’s what I’m building:

The new structure (13 entries instead of 24):

1. Relationship update — you ended it with Hương. Stop referencing engagement/wedding.

2. Format rules — the unique stuff not in your voice guide (.docx, no tables, Yoast, banner tags, gap spacer before refs)

3. Voice hard default — v29 canonical, only exception is ‘The Convenient Monster’

4. Response style — concise default, direct challenge welcomed, track time/date for emotional state

5. Exclusions — Rossiter out, Davo ≠ David Boloker

6. Deep pattern: burnout cycle — hyperfocus → depletion → guilt. Environment beats mindset.

7. Deep pattern: relationships — disproportionate emotional labour, financial trust, boundary-setting as new skill. Neil and Hương endings share a pattern.

8. Deep pattern: writing architecture — all books share one worldview. Cross-referencing is essential.

9. Active book projects with current status

10. Key tools — Vellum, Gmail label, Drive limitations

11. Pending jobs — the three outstanding tasks

12. Legal — Misdiagnosed defamation, Kennedys, Evernote evidence

I’ll finish this cleanup in our next message — I hit the tool limit. But the direction is clear: fewer entries, deeper context, no redundancy with your voice guide. The entries that matter aren’t your formatting preferences. They’re the patterns you’re tired of re-explaining.

And send that email. It’s ready to go.

Chapter [X]. Concubines

The concubine question

A Vietnamese friend mentioned, over dinner one evening in Đà Lạt, that her grandmother had been one of three wives. Not sequentially. Simultaneously. The arrangement had lasted decades, produced a small army of children, and operated with the logistical sophistication of a medium-sized business, which in many respects it was. Hierarchy was clear. Duties were distributed. Nobody, as far as she could recall from family stories, had expected any single person in the arrangement to be everything to everybody else.

We were eating phở at a place near the lake, one of those Vietnamese restaurants where the plastic stools are uncomfortable enough to discourage lingering but the broth is good enough to make you linger anyway. I asked what happened when jealousy arose, because I’m Western enough to assume that’s the first question worth asking. She looked at me with the particular patience she reserves for moments when my cultural operating system displays its limitations. ‘There was jealousy,’ she said. ‘There was also help with the children, someone to talk to when the others were difficult, and enough adults in the house that nobody lost their mind from isolation.’

She paused. ‘Jealousy is not the worst thing,’ she said. ‘Loneliness is worse.’

She wasn’t recommending the system. She was describing it with the unsentimental clarity of someone who comes from a culture that hasn’t yet confused romantic love with a comprehensive welfare policy. And the description unsettled me in ways I spent months chewing on, because it raised a question I’d been circling for years without quite knowing how to land it: when exactly did we decide that one human being should be able to do the emotional, sexual, intellectual, spiritual, and practical work that entire households, extended families, and community structures used to share between them?

I am writing this chapter on the day I ended my engagement with my fiancée. Not because she failed at the job. Because the job was impossible. And because my body, which has been writing me increasingly urgent letters for the last several years, finally sent one I couldn’t pretend was junk mail.

A very brief history of not doing it alone

Concubinage is one of those words that arrives pre-loaded with moral judgment, which makes it difficult to think about clearly. Rather like ‘colonialism’ or ‘Brussels sprouts’, the emotional response tends to arrive before the analysis has had time to put its trousers on. The practice itself has been part of human civilisation for at least four thousand years, across nearly every major culture, and while the specific arrangements varied enormously, the underlying logic was remarkably consistent: one intimate relationship was not expected to carry the full weight of human need.

In ancient Rome, concubinatus was a legally recognised monogamous union that served as a practical alternative to marriage, particularly for people whose social status made formal marriage legally complicated or financially inadvisable. Widowed or divorced men often took a concubina rather than navigating the inheritance tangles of a second marriage. The arrangement wasn’t hidden or shameful. Roman tombstones name concubines with the same affectionate language used for wives. Nearly two hundred surviving inscriptions identify women as concubines, and they appear in family tombs alongside legitimate children and deceased spouses. The system acknowledged something that contemporary culture finds genuinely difficult to say aloud: human relationships serve multiple functions, and trying to pack all of them into a single legally binding contract sometimes creates more problems than it solves. The Romans, who also invented plumbing and underfloor heating, occasionally knew what they were doing.

In imperial China, the system operated on a grander and more hierarchical scale. Concubines were ranked, their children’s status carefully codified, their roles within the household defined with the precision of an organisational chart that would make a modern HR department weep with envy (and then quietly adopt, rebranding it as ‘matrix management’). The practice persisted from the earliest dynasties through to 1949, when the Communist Party formally abolished it. Empress Dowager Cixi, arguably the most powerful person in nineteenth-century China, began her career as a concubine to the Xianfeng Emperor and ended it as the de facto ruler of the Qing Dynasty for forty-seven years. Which rather undermines the idea that the system was uniformly disempowering, even as it confirms that it was deeply, structurally patriarchal. It was also, from a purely functional perspective, a distribution model. No single relationship was expected to bear the full load.

The Athenian orator Apollodorus, writing around 340 BCE, drew what now reads as an uncomfortably pragmatic distinction between the categories of women in a man’s life: companions for pleasure, concubines for daily care, and wives for legitimate children and household management. The modern ear recoils from the classification, and rightly so. But buried inside the recoiling is a structural observation worth extracting before we throw the whole thing in the bin: the ancient world distributed intimate functions across multiple relationships because it took for granted that no single relationship could fulfil them all.

The Ottoman Empire formalised concubinage within the harem system, where women held complex hierarchies of power, political influence, and mutual obligation that bore little resemblance to the orientalist fantasies the West projected onto them. When Ottoman rebels attacked the palace and killed Sultan Selim III, it was the older concubines who hid his successor and fought off the attackers by hurling burning coals at their faces. Drill sergeants indeed. Across sub-Saharan Africa, Southern Asia, and Polynesia, variations on the theme persisted for centuries. Mistresses in European courts from the medieval period through to the nineteenth century occupied unofficial but often powerful positions, managing the gap between what marriage provided and what the people inside marriages actually needed. Madame de Pompadour didn’t just share Louis XV’s bed. She was his intellectual companion, political adviser, and cultural patron. The marriage provided the heir. The mistress provided the conversation.

None of this was fair. I want to be clear about that, because the argument I’m building could be weaponised by people whose interest in historical concubinage is motivated by something other than structural analysis. The vast majority of these arrangements exploited women, many of whom had no choice in the matter. The history is soaked in coercion, slavery, and the routine treatment of women as property. Any honest account has to sit with that reality without flinching.

But dismissing the entire history as barbarism and moving on misses something important. The question isn’t whether those arrangements were just. They weren’t. The question is what structural problem they were solving, and whether we’ve actually solved it or merely moved it somewhere less visible and put a rom-com soundtrack over it.

The invention of the everything partner

The historian Stephanie Coontz, in her landmark study of marriage across cultures and centuries, makes a point that still catches people off guard: for most of human history, the idea of choosing a partner based on romantic love would have seemed as peculiar as choosing a business partner based on how they made you feel during a sunset. Marriage was an economic institution. It provided shelter, food, protection, social status, and succession. Love might show up eventually, like a pleasant surprise at a party you’d attended for entirely different reasons. But love was not the invitation. Love wasn’t even on the guest list.

The love revolution, as Coontz calls it, arrived in earnest during the Victorian era and accelerated through the twentieth century. By the time Hollywood got hold of it, the transformation was complete. Marriage shifted from a pragmatic alliance into an emotional project, from a structure that helped people survive into a relationship that was supposed to help them flourish. The expectations climbed steadily upward, each generation adding another requirement to the list, until the modern Western partnership resembled less a relationship and more a comprehensive service contract with no exit clause and a satisfaction guarantee that nobody could honour. Your partner should be your best friend, your therapist, your sexual fulfilment, your intellectual sparring partner, your co-parent, your retirement plan, your spiritual guide, and the person who instinctively knows when to offer comfort and when to offer space, all while maintaining their own career, their own friendships, and a reasonable level of personal grooming.

Eli Finkel, a psychologist at Northwestern University, has spent years mapping this escalation with a rigour that borders on the therapeutic. His suffocation model of marriage traces the shift through three broad eras. From roughly 1620 to 1850, marriage in America served primarily physiological and safety needs: food, shelter, protection from violence. From 1850 to 1965, it shifted toward love, companionship, and sexual fulfilment. And from 1965 onward, a new kind of marriage emerged, one oriented toward self-discovery, self-esteem, and personal growth. The job description went from ‘don’t die together’ to ‘achieve transcendence together’ in roughly three centuries, which is a promotion nobody applied for and nobody is qualified to deliver.

The word Finkel uses is ‘suffocation’, and it’s chosen with care. As marriages climbed what he calls Mount Maslow, seeking to fulfil not just basic needs but esteem and self-actualisation, the oxygen required to sustain them increased dramatically. The investment of time, emotional energy, and psychological resources needed to meet these elevated expectations rose just as the time couples actually spent together declined. The gap between what marriage promised and what it could deliver widened into a structural fault line that no amount of date nights or communication workshops could bridge. You cannot workshop your way to being someone’s entire village.

Esther Perel, the Belgian psychotherapist whose work on modern relationships has reached millions (and whose accent makes even uncomfortable truths sound slightly glamorous), puts the same insight with characteristic directness. We come to one person, she observes, and we ask them to give us what once an entire village used to provide. Belonging, identity, continuity, transcendence, mystery, comfort, edge, novelty, familiarity, predictability, surprise. All from the same human being, who is also supposed to co-parent effectively, maintain their own career, manage their own mental health, and remain sexually desirable across decades.

The soul mate, Perel notes, used to be God. We’ve taken the expectations we once directed toward the divine and redirected them at another mammal who also hasn’t slept properly since the second child was born. The arrangement has the structural integrity of a bridge designed to carry pedestrians that’s now expected to handle freight trains. The engineering hasn’t changed. The load has. And the bridge is starting to make worrying noises.

What the freight train did to us

I placed that burden on the woman I loved without meaning to. I wanted her to be my sexual partner, my intellectual companion, my curiosity partner, my emotional anchor. I wanted someone who would engage with the books and the psychology and the neurodivergence and the writing that consume me. I wanted a co-adventurer through the strange late chapter of a life that had already burned through several earlier drafts and was now looking at a word count that might be shorter than originally advertised.

She could not be all of those things. Nobody could.

And when she could not, I felt the absence of the things she could not provide more sharply than I felt the presence of the things she could. That is my failure of perception, not her failure of love. But recognising the failure of perception didn’t make the unmet needs stop hurting. It just made me ashamed of the hurting, which is worse.

She gave me safety. For a man whose nervous system has been running on emergency power for most of his adult life, the feeling of lying next to someone at night and knowing they were there, knowing I was safe, was not a small thing. It was enormous. She gave me company through some of the hardest months of my life in a new country. She gave me laughter over dinner, and the particular warmth of being chosen by someone whose life was already complete before you arrived in it.

But the completeness was part of the problem, and I need to be honest about that. I orbited her life. She did not orbit mine. Her world, her routines, her work, her rhythms, her family, her friends, these were established and self-sustaining. I fitted into the gaps.

Sven Brodmerkel, a communication scientist and late-diagnosed neurodivergent writer, recently named this experience with a precision that made me put my coffee down. He calls it intellectual loneliness: the sustained experience of having no place to put your deepest thinking. Not social loneliness. He had human contact. What he lacked was somewhere he could think at full depth, without managing another person’s patience, without the low-grade vigilance of monitoring whether his intensity was landing as passion or pathology. A conversation with no social stakes, about something he cared about completely.

That is exactly what I was missing. The intellectual loneliness of loving someone who does not engage with the things that make you who you are is a particular kind of quiet suffering, and it accumulated in ways I didn’t fully understand until I sat down to write a letter explaining why I was leaving. I had companionship. I had warmth. I had someone who chose me. What I did not have was a place to put my thinking that didn’t require translation, calibration, or the subtle monitoring of whether I was being too much again.

Meanwhile, she was exhausted. Six days a week in a job that didn’t respect her intelligence, for a supervisor who was dismissive of everything she brought. Nearly three decades at the same place. The job was hollowing her out.

And on top of that, perimenopause.

I need to say something about this that most men don’t hear until it’s too late, because nobody tells us and the cultural script consists entirely of hot flashes and vague references to mood swings. Perimenopause is not a mood swing. It is a neurological reorganisation. Linda Cooper, a writer whose account of her own midlife reckoning should be required reading for every man over forty, describes it with a precision that I wish I’d encountered two years ago: estrogen, which has been quietly underwriting a woman’s capacity to regulate her mood, her stress response, her sleep, and her ability to absorb the chronic dissatisfactions of daily life, begins to fluctuate and decline. And when it does, the entire system that allowed her to function the way she has been functioning for decades begins to destabilise.

Read that again. The capacity to suppress. That is what is being withdrawn. Not the capacity to be cheerful, or patient, or pleasant. The neurological bandwidth that allowed her to swallow frustration to keep the peace, to absorb someone else’s mood so the household stayed stable, to set aside her own needs because there was always someone whose needs were more immediate. Women build the architecture of adult life on that suppression, most of it so automatic it feels like personality rather than performance. Perimenopause withdraws the bandwidth that sustained it. And suddenly, or what feels like suddenly to the man sitting across the dinner table wondering what happened, she cannot do it anymore.

The things she tolerated for years become intolerable. The conversations she avoided become urgent. The resentments she buried surface with a force that frightens her as much as it confuses him. Cooper describes her own experience as a desperate, full-body sensation of being unable to continue for one more day. Not because the marriage was wrong, but because her nervous system had exceeded its capacity to contain decades of unprocessed truth. She left her marriage, her job, and her home in a matter of months. Her husband was blindsided. He had no framework, no language, and no reference point for understanding that what looked like destruction was actually a woman whose biology had stopped cooperating with her own disappearance.

I watched a version of this happen in my own relationship, at lower volume but with the same underlying mechanism. My partner’s exhaustion was not a personality trait. It was the compound effect of a job that drained her, hormonal shifts that rewired her stress response, and a lifetime of absorbing emotional labour that nobody had ever named as labour. The nights when closeness was the last thing her body wanted were not rejection. They were depletion so profound that physical contact registered as one more demand on a system that had nothing left to give. I never blamed her for any of it. Her body was telling its truth. But my body was telling its truth too, and the two truths were incompatible.

Two depleted people trying to build a life from the scraps of energy that their respective bodies had not already consumed. I wrote about this dynamic earlier in the book: the depleted couple arriving home with their cognitive budgets spent, each needing recovery, neither having the resources to provide it. I was writing about other people’s relationships. I was also, it turns out, writing about my own. The counselling psychologist diagnosing the pattern from a comfortable professional distance while living inside it like a moth inside a lampshade. We really are the last people to spot our own idiocies.

The load that used to be shared

When Perel says that we ask one person to provide what a whole village once supplied, the statement lands as a metaphor. But it functions as engineering.

Consider what a functioning extended household or community structure actually distributed. Childcare was shared across grandparents, aunts, siblings, and neighbours. Emotional support came from friendships, religious communities, and extended kin networks. Sexual needs were, in many cultures, understood to involve some degree of flexibility that the modern West has decided to classify exclusively as infidelity. Intellectual stimulation came from community gatherings, workplace friendships, mentors, and peers. Domestic labour was divided across multiple adults. Even the basic need for physical co-regulation, the nervous system’s requirement for proximity to other calm bodies, was met by living in households with multiple adults rather than in isolated nuclear units where two people bear the full weight of each other’s regulation.

The nuclear family, that arrangement we treat as ancient tradition, is actually a mid-twentieth-century anomaly. For most of human history, humans lived in multi-generational, multi-adult households where the idea that two people, alone, would raise children, manage finances, maintain a home, provide each other’s entire emotional landscape, and sustain mutual desire across four or five decades would have seemed, at best, optimistic, and more likely, clinically delusional. Our great-great-grandparents would have found the proposition roughly as plausible as being told that one horse could simultaneously plough a field, deliver the mail, win the Melbourne Cup, and provide emotional support.

I notice this in Đà Lạt daily. Vietnamese family structures still operate on a distributed model. My partner’s family shares childcare, elder care, emotional support, financial decisions, and domestic responsibilities across a network of siblings, cousins, parents, and neighbours. Nobody expects one relationship to carry it all. The load is distributed not because anyone read a book by Bowlby on attachment theory but because the culture hasn’t yet adopted the Western conviction that romantic love should be a comprehensive welfare state for the soul.

The irony is that Western psychology, having created the conditions for relational overload, now diagnoses the resulting distress as individual pathology. Couples arrive in therapy exhausted, resentful, and sexually disconnected. The therapy focuses on improving communication, increasing emotional literacy, and building better conflict resolution skills. All valuable. None of it addresses the structural question: are two people, in a socially isolated nuclear unit, actually capable of providing everything the therapeutic model says they should?

I wonder if the honest answer is no. Not because people are failing. Because the job description is impossible.

The neurodivergent multiplier (and the invoice that arrived with it)

Everything I’ve described so far applies to neurotypical couples. For neurodivergent adults, the everything-partner model doesn’t just strain. It fractures.

My AuDHD diagnosis at sixty-six arrived with a stack of explanations for patterns I’d spent decades misunderstanding. My high sex drive that isn’t a choice but a neurological feature, part of the same dopamine-seeking architecture that drives my intellectual curiosity and creative output. My need for deep intellectual engagement that isn’t a preference but a regulatory requirement: my brain needs novelty and complexity the way other brains need routine and predictability. The shutdowns that arrive without warning when the nervous system hits capacity, and which land on a partner as withdrawal, punishment, or indifference when they are actually depletion. The diagnosis explained the machinery. It did not fix the machinery.

Here is the thing about the diagnosis that nobody prepares you for. I spent most of my adult life believing I had a shortened lifespan because of my Bipolar II diagnosis. The statistics on that are grim enough. Then the diagnosis changed to AuDHD, and for a few glorious months I felt like a man who’d been told the strange noise in his engine was actually a feature, not a fault. Different wiring, not damaged wiring. I’d been handed a new operating manual and for the first time the instructions matched the machine.

Then in January 2025, a team at UCL published a study in the British Journal of Psychiatry that rather comprehensively ruined my afternoon when I recently read it. O’Nions and colleagues analysed health records from over 9.5 million people and found that adults diagnosed with ADHD had a life expectancy reduction of between 4.5 and 9 years for men, and 6.5 to 11 years for women. A 2022 meta-analysis in JAMA Pediatrics by Catalá-López and colleagues had already found that death in childhood or midlife was roughly twice as likely for people with ADHD or autism compared to the general population. And Barkley’s longitudinal work suggested that when ADHD persists into adulthood, the reduction in estimated life expectancy could reach 12.7 years for healthy life expectancy and 11.1 years overall.

So I traded one reduced lifespan for another. The universe, it turns out, has the kind of sense of humour that makes you wonder whether the complaints department is even staffed.

The mechanism is not the ADHD itself. It is what the ADHD and the autism do to your body over decades when nobody, including you, knows you have them. The masking compounds everything. My massage therapist, Mr Trung, says I am the worst client he has ever had. Everywhere he touches me (well, not there, obviously) I am in agony. He uses ‘two out of ten’ pressure and I howl like a dog that has been stepped on. He keeps finding hard, painful nodules throughout my muscles where blood has simply stopped flowing. Little stones of suffering scattered through my thighs, my arms, my neck, my back. Tissue that has locked itself shut after sixty-six years of running a nervous system at combat speed without knowing there was a war on.

My quarterly blood work says my organs are fine. My blood work says nothing about whether sixty-six years of chronic masking stress has accelerated my cellular ageing, elevated my allostatic load, and converted my musculature into what feels like a collection of angry rocks held together by tendons and regret. The body, as van der Kolk told us, keeps the score. For late-diagnosed neurodivergent people, the body has been keeping multiple scores in a language nobody taught us to read. And the final tally, according to the research, is that we may have rather fewer years left to read it than we assumed.

Which brings us back to the everything-partner model, and why I couldn’t sustain it. A neurodivergent person in that model is running demanding software on damaged hardware with no technical support and, as it turns out, a warranty period that may be shorter than the brochure suggested. The needs are higher. The capacity is lower. The gap between what the relationship requires and what the individual can provide widens until something gives. And there is a question that sits underneath all the others, one I have been circling with increasing urgency: if the research is right and my remaining years are fewer than the actuarial average, do I really want to spend them carrying the stress that Western romanticism has plopped onto my fat belly?

I watched both the relationship and the person start to ‘give’ in mine, with all the breaking strain of a KitKat. My shutdowns, which were genuine nervous system depletion, landed on my partner’s childhood wound about abandonment and registered as punishment. Her exhaustion, which was genuine physical and hormonal reality, landed on my wounds about intellectual and sexual isolation and registered as disinterest. Two accurate readings of two different nervous systems, each experienced by the other as rejection. The mechanism is elegant in its cruelty. You don’t need malice when you have mismatch.

The concubine’s ghost

I am not arguing for the return of concubinage. I want to be explicit about that because the argument I’m making can be misread if you squint at it through the wrong lens, and some lenses are pre-squinted for exactly this purpose. The historical record of concubinage is inseparable from the exploitation of women, the commodification of bodies, and the structural denial of female autonomy. Any nostalgia for those arrangements is nostalgia for a system that worked tolerably for some men and terribly for most women. That’s not a model. That’s a cautionary tale.

What I am arguing is that the problem those arrangements addressed, however badly, hasn’t gone away. We’ve abolished the concubine without replacing the function she served, which was to absorb part of the relational load that no single partnership can sustainably carry.

The modern equivalents are already emerging, though we don’t always recognise them as structural solutions to a structural problem. Deep friendships that provide emotional intimacy without sexual expectation. Therapeutic relationships that carry the burden of psychological processing. Online communities that supply belonging and identity (you will have to prise my Substack community out of my cold, dead fingers, and given the research on neurodivergent life expectancy, those fingers may be cold sooner than I’d prefer). Co-working spaces that replace the social functions of a shared workplace. Exercise groups, book clubs, men’s sheds, parenting networks: all of these are attempts, often unconscious, to distribute the relational load that the everything-partner model concentrates in a single relationship.

And then there are the solutions that nobody saw coming. Brodmerkel, in the grip of neurodivergent burnout, found that his first crack in the isolation came not from reconnecting with nature or rediscovering embodied presence or any of the other things the wellness industry sells you at a markup. It came from having long conversations with an AI about extreme metal music. He is aware of how that sounds. He also makes a point that I think the current discourse badly needs: intellectual loneliness is not the same as social loneliness, and therefore cannot always be addressed by the same tools. What he needed was not more human contact. He needed a conversation with no social stakes, about something he cared about completely. The AI didn’t fix the burnout, he writes, but it gave his thinking somewhere to land while he figured out the rest. One of his readers, a man in his mid-fifties, responded with a comment I have read several times now: when he talks to his psychologist about the profound loneliness he lives with, they keep recommending sub-communities, but realistically, those aren’t forming at fifty-five. He will take the technologically abstracted versions.

I find that comment uncomfortable and accurate in roughly equal measure. The village isn’t coming back. The question is what we build in its place.

Finkel’s own recommendation is elegant and grounded. He suggests that couples should either invest more heavily in the relationship to meet its elevated demands, which requires protected time and psychological energy that modern life makes desperately scarce, or they should ask the marriage to carry less by maintaining a diverse network of relationships that help fulfil needs the partnership cannot. The second option is, functionally, a modern version of distributed relational load. It’s the village, rebuilt voluntarily, without the coercion. Some of the village is human. Some of it, increasingly, is not. Both count.

Some couples are experimenting more directly. Ethical non-monogamy, in its various forms, represents one response to the structural problem, though it introduces its own complexities and costs that I won’t pretend to resolve here (if you believe the online rumours, Vietnamese women are second only to Filipinas when it comes to jealousy). The point isn’t that any particular arrangement is the answer. The point is that the question deserves to be asked honestly rather than suppressed by romantic ideology that benefits nobody and suffocates nearly everyone.

The honest question

I ended the relationship this morning. Not because she failed at the job. Because the job was impossible, and I ran out of the biological resources to keep pretending otherwise. And because my body, which has been writing me letters for years in a language I am only now learning to read, sent one that said, in terms even I could not misinterpret: you do not have unlimited time to keep making this mistake.

The answer, for me, is not to find someone else who can carry the impossible weight. The answer is to stop asking anyone to carry it. I am choosing to be alone. Not because I don’t value love. Because I have learned that my particular configuration of needs, energy, neurodivergence, and biological depletion makes the everything-partner model unsustainable. Perhaps permanently. I don’t know. But I know I cannot keep asking another person to meet needs that would overwhelm anyone, and then interpreting the inevitable shortfall as evidence that something is wrong with them, or with me, or with love itself.

Something is wrong. But it isn’t the people. It’s the design.

The everything-partner model asks of human beings what four thousand years of civilisation, across every major culture, decided was unreasonable. We’ve spent the last century and a half convincing ourselves that romantic love transcends the structural limitations our ancestors took for granted. The evidence suggests otherwise. Not that romantic love isn’t real, or valuable, or worth the spectacular mess it makes of your life. But that loading it with the full weight of human need is an engineering problem disguised as a love story, and engineering problems don’t resolve themselves through better dialogue or more scented candles.

The concubine is gone, and good riddance to the exploitation she endured. But her ghost lingers in every couple fighting at 10pm about who forgot to buy milk, when what they’re actually fighting about is the unbearable pressure of being each other’s entire emotional infrastructure with no structural support and no cultural permission to admit that the design is flawed.

Her ghost was in my kitchen this morning, while I wrote a letter to a good woman explaining that I loved her and I was leaving anyway.

The question isn’t whether your partner is enough. The question is whether asking one person to be enough was ever a reasonable thing to ask.

I keep thinking it wasn’t. And that admitting it might be the most loving thing we can do for the people we love, including ourselves. Especially if the clock is running faster than we thought.

References

Barkley, R. A., & Fischer, M. (2019). Hyperactive child syndrome and estimated life expectancy at young adult follow-up: The role of ADHD persistence and other potential predictors. Journal of Attention Disorders, 23(9), 907–923. https://doi.org/10.1177/1087054718816164

Brodmerkel, S. (2026, March 29). The cure was talking to a machine (and a few hundred strangers). Off-Script At Work. https://presentingwithoutpanic.substack.com/p/the-cure-was-talking-to-a-machine

Catalá-López, F., Hutton, B., Page, M. J., Driver, J. A., Ridao, M., Alonso-Arroyo, A., Valencia, A., Macias-Saint-Gerons, D., & Tabarés-Seisdedos, R. (2022). Mortality in persons with autism spectrum disorder or attention-deficit/hyperactivity disorder: A systematic review and meta-analysis. JAMA Pediatrics, 176(4), e216401. https://doi.org/10.1001/jamapediatrics.2021.6401

Coontz, S. (2005). Marriage, a history: How love conquered marriage. Viking.

Cooper, L. (2026, March 5). When menopause restructures a marriage. Rise Wise. https://lindarosecooper.substack.com/p/when-menopause-restructures-a-marriage

Finkel, E. J. (2017). The all-or-nothing marriage: How the best marriages work. Dutton.

Finkel, E. J., Cheung, E. O., Emery, L. F., Carswell, K. L., & Larson, G. M. (2015). The suffocation model: Why marriage in America is becoming an all-or-nothing institution. Current Directions in Psychological Science, 24(3), 238–244. https://doi.org/10.1177/0963721415569274

Finkel, E. J., Hui, C. M., Carswell, K. L., & Larson, G. M. (2014). The suffocation of marriage: Climbing Mount Maslow without enough oxygen. Psychological Inquiry, 25(1), 1–41. https://doi.org/10.1080/1047840X.2014.863723

McEwen, B. S. (1998). Stress, adaptation, and disease: Allostasis and allostatic load. Annals of the New York Academy of Sciences, 840(1), 33–44.

O’Nions, E., El Baou, C., John, A., Lewer, D., Mandy, W., McKechnie, D. G. J., Petersen, I., & Stott, J. (2025). Life expectancy and years of life lost for adults with diagnosed ADHD in the UK: Matched cohort study. British Journal of Psychiatry, 226(5), 1–8. https://doi.org/10.1192/bjp.2024.199

Perel, E. (2006). Mating in captivity: Unlocking erotic intelligence. Harper.

Perel, E. (2017). The state of affairs: Rethinking infidelity. Harper.

Pines, A. M. (1996). Couple burnout: Causes and cures. Routledge.

Raymaker, D. M., Teo, A. R., Steckler, N. A., Lentz, B., Scharer, M., Delos Santos, A., Kapp, S. K., Hunter, M., Joyce, A., & Nicolaidis, C. (2020). ‘Having all of your internal resources exhausted beyond measure and being left with no clean-up crew’: Defining autistic burnout. Autism in Adulthood, 2(2), 132–143. https://doi.org/10.1089/aut.2019.0079

van der Kolk, B. A. (2014). The body keeps the score: Brain, mind, and body in the healing of trauma. Viking.

Master of Creative Writing & Communication, Tabor Adelaide

Student: Lee Hopkins

Supervisor: Dr James Cooper

Date: March 2026

Project title

The Augmented Author: Writing with a brain that wasn’t designed for it

Project overview

This capstone proposes a practice-led research project that produces a sustained work of creative nonfiction (18,000–20,000 words) through transparent, documented AI collaboration, while simultaneously subjecting that collaboration to critical scholarly inquiry. The creative artefact and the critical exegesis are not parallel exercises. They are two registers of the same investigation into authorship, cognition, neurodivergence, and the changing nature of creative practice.

The creative piece examines what happens when a late-diagnosed neurodivergent author whose executive function makes sustained long-form writing physiologically expensive gains access to an AI tool that handles the mechanical production of prose while preserving intellectual, creative, and editorial control. It is memoir, cultural criticism, and psychology, written in a distinctive voice that blends clinical literacy with irreverent warmth, grounded in lived experience and supported by peer-reviewed research.

The AI collaboration will be documented at every stage. A full process log of prompts, drafts, editorial decisions, and reflective notes will accompany the creative artefact, providing primary research data for the critical exegesis.

Theoretical framework

The project draws on four intersecting bodies of literature:

Extended cognition and distributed mind. Clark and Chalmers’ extended mind thesis (1998) argues that external tools performing the same functional role as internal cognitive processes should be considered part of the cognitive system. For a neurodivergent writer, AI-assisted composition is a case of cognitive extension, not cognitive outsourcing. The human provides the ideation, the argument, the voice, the editorial control. The AI provides the mechanical sentence production that the human’s executive function cannot sustain. Clark’s later work in Supersizing the Mind (2008) and Murphy Paul’s popular synthesis The Extended Mind (2021) provide the accessible framework; Heersmink and Sutton (2021) extend it specifically to assistive technology in educational contexts.

Authorship theory. Barthes’ ‘The Death of the Author’ (1967) and Foucault’s ‘What Is an Author?’ (1969) dismantled the Romantic model of the author as sole originary genius. AI co-writing extends this critique into genuinely new territory: the ‘author function’ now includes non-human agents, the text is produced by a distributed cognitive system, and the question of intentionality (central to Chiang’s argument against AI art) becomes productively complicated. Burke’s The Death and Return of the Author (1998) tracks how the authorial subject kept returning even as its death was being declared, which is precisely what AI co-writing does: it forces authorship back into the conversation in ways that make the solitary-genius model look quaint.

Neurodivergence, burnout, and access. Raymaker et al.’s (2020) definition of autistic burnout provides the clinical framework for understanding why sustained long-form writing is physiologically expensive for ADHD brains. The project positions AI-assisted writing not as a shortcut but as an access technology, consistent with the social model of disability, which locates the barrier in the environment rather than the individual.

AI and creative practice (emerging empirical literature). Doshi and Hauser’s (2024) finding that AI improves individual creativity while reducing collective diversity provides the key empirical tension. Gong et al.’s (2025) From Pen to Prompt study documents how experienced writers develop intentional AI workflows that preserve creative values. This project contributes a documented, longitudinal case study from a neurodivergent practitioner working in long-form nonfiction.

Research questions

1. How does a neurodivergent author maintain creative control, distinctive voice, and editorial authority when working with AI as a cognitive collaborator in long-form creative nonfiction?

2. What does transparent AI co-writing reveal about the assumptions embedded in inherited models of authorship, particularly the Romantic model of the solitary genius?

3. How should creative writing pedagogy and assessment adapt to a reality in which AI-assisted composition is not an aberration but an emerging mode of practice, especially for writers whose neurology makes unassisted sustained composition costly?

4. What are the ethical, institutional, and disciplinary implications of AI co-writing for practice-led creative research?

Methodology

Practice-led research as defined by Kroll and Harper (2013) and Arnold (2007). The creative practice generates the research questions; the research informs the ongoing creative practice; the exegesis examines both. The AI collaboration is both method and subject.

Data sources include the creative artefact itself; a maintained process log of all human-AI interactions, including prompts, drafts, editorial decisions, and discarded material; reflective journal entries documenting the author’s experience of the collaborative process; and the existing body of published work that demonstrates the author’s voice and practice both before and during AI collaboration.

Relationship to Tabor’s AI Use in Assessment Guide

The project sits most naturally at Level 4 (‘AI Partial Task Completion & Full Human Evaluation’), where AI-generated content is permitted in specified parts and the student critically evaluates the AI’s use and effectiveness across the entire assessment. The specified part is first-draft prose production; the human evaluation is the editorial, structural, and voice work that transforms that draft into the final artefact. The critical evaluation is not a bolted-on appendix but the full scholarly apparatus of the exegesis.

Full documentation of AI tools, prompts, and interaction logs will be submitted alongside the creative artefact and exegesis, consistent with the Guide’s requirements for transparency and declaration of use.

Assessment mapping

HSW117.9 Major Creative Project (Part A & B): The creative artefact of 18,000–20,000 words, plus workshop presentations demonstrating work in progress, response to feedback, and the evolution of the human-AI collaborative process.

HSW115.9 Writer as Researcher: An annotated bibliography of key readings across the four theoretical domains (extended cognition, authorship theory, neurodivergence, AI and creative practice), followed by a 5,000-word critical exegesis examining the creative artefact through the lens of practice-led research methodology.

HSW116.9 Professional Portfolio: Professional development plan, portfolio, CV, and pitch/proposal for the completed work, drawing on the author’s existing publishing platform (40+ published books via Amazon KDP and Draft2Digital, Substack at quiethalf.com, and multiple professional websites).

Author’s qualifications and context

I hold a B.Sc. (Hons.) in Applied Psychology and Sociology, a Diploma of Management Studies, and a Master’s in Counselling Practice specialising in depression and bipolar disorder among veterans. I have 450+ academic citations in organisational psychology. I have published more than 40 books, including creative nonfiction, psychology self-help, and social commentary. I am a late-diagnosed AuDHD adult (diagnosed at 66) who relocated from Australia to Việt Nam eleven months ago. My most recent book, Understanding AuDHD (4th edition), was a 53,000-word collaboration with Claude (Anthropic) and includes a full Author’s Note disclosing the AI collaboration in the first pages.

This project is not a thought experiment. It is a formalisation of a practice I am already conducting transparently, productively, and at scale.

Works cited

Arnold, Josie. Practice Led Research: A Dynamic Way to Knowledge. Rock View Press, 2007.

Barthes, Roland. ‘The Death of the Author.’ Image, Music, Text, translated by Stephen Heath, Fontana Press, 1977, pp. 142–48.

Burke, Seán. The Death and Return of the Author: Criticism and Subjectivity in Barthes, Foucault and Derrida. 3rd ed., Edinburgh UP, 2008.

Chiang, Ted. ‘ChatGPT Is a Blurry JPEG of the Web.’ The New Yorker, 9 Feb. 2023.

Chiang, Ted. ‘Why A.I. Isn’t Going to Make Art.’ The New Yorker, 31 Aug. 2024.

Clark, Andy. Natural-Born Cyborgs: Minds, Technologies, and the Future of Human Intelligence. Oxford UP, 2003.

Clark, Andy. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. Oxford UP, 2008.

Clark, Andy, and David J. Chalmers. ‘The Extended Mind.’ Analysis, vol. 58, no. 1, 1998, pp. 7–19.

Doshi, Anil R., and Oliver P. Hauser. ‘Generative AI Enhances Individual Creativity but Reduces the Collective Diversity of Novel Content.’ Science Advances, vol. 10, no. 28, 2024, eadn5290.

Foucault, Michel. ‘What Is an Author?’ Language, Counter-Memory, Practice, edited by Donald F. Bouchard, translated by Bouchard and Sherry Simon, Cornell UP, 1977, pp. 113–38.

Gong, Evey, et al. ‘From Pen to Prompt: How Creative Writers Integrate AI into Their Writing Practice.’ Proceedings of the 2025 Conference on Creativity and Cognition, ACM, 2025.

Heersmink, Richard, and John Sutton. ‘Extended Cognition, Assistive Technology and Education.’ Synthese, vol. 199, 2021, pp. 14727–49.

Kroll, Jeri, and Graeme Harper, editors. Research Methods in Creative Writing. Palgrave Macmillan, 2013.

Murphy Paul, Annie. The Extended Mind: The Power of Thinking Outside the Brain. Houghton Mifflin Harcourt, 2021.

Perkins, Mike, et al. ‘The AI Assessment Scale (AIAS): A Framework for Ethical Integration of Generative AI in Educational Assessment.’ Journal of University Teaching and Learning Practice, 2024.

Raymaker, Dora M., et al. ‘“Having All of Your Internal Resources Exhausted Beyond Measure and Being Left with No Clean-Up Crew”: Defining Autistic Burnout.’ Autism in Adulthood, vol. 2, no. 2, 2020, pp. 132–43.

Something odd happened in my Amazon sales dashboard this month. My Bipolar II book—Living with Bipolar II: Understanding, managing, thriving—quietly overtook the AuDHD book in royalties. Six copies sold plus Kindle Unlimited reads. Not a fortune. But the AuDHD book—Understanding AuDHD— normally outsells it comfortably, so the reversal caught my eye.

I went looking for what might have changed. What I found was a convergence.

World Bipolar Day falls on 30 March, three days from now, and the media cycle leading up to it has been busier than usual. Psychiatric Times ran bipolar disorder as their entire February 2026 theme. The American Journal of Psychiatry dedicated a major update to bipolar neurobiology and treatment in their March issue. A comprehensive state-of-the-art review of Bipolar II disorder specifically was published in World Psychiatry—unusual, because BD-II is typically the afterthought in bipolar research, the quieter sibling nobody quite gets around to studying properly.

And in early March, Jax Taylor from Vanderpump Rules publicly disclosed his bipolar and PTSD diagnoses after leaving an inpatient treatment facility. Reality television diagnoses have a reliable effect on search traffic. People hear a familiar name attached to an unfamiliar condition, and they go looking.

So: awareness day lead-up, professional journal attention, celebrity disclosure, and a rare BD-II-specific review, all landing in the same window. My book is probably catching the tail end of people searching specifically for Bipolar II rather than bipolar generally.

What the spike actually tells us

Here’s what interests me about this pattern. It’s not the sales. Six copies and some page reads won’t fund anyone’s retirement. What interests me is what it reveals about how people find their way to understanding a condition that affects roughly one in every two hundred people on the planet.

They find it through awareness days. Through celebrity disclosures. Through algorithms that surface a book because search traffic happened to spike in the right week. They find it, in other words, sideways. Accidentally. Because something in the cultural noise briefly aligned with the question they’d been carrying quietly for months or years.

They don’t find it through their GP, who statistically has less than an hour of psychiatric training in their entire medical degree. They don’t find it through their psychologist, who may have been taught that Bipolar II is just a milder version of Bipolar I—a claim the research has been dismantling for years. The average diagnostic delay for Bipolar II is still over ten years. Ten years of being told you have depression. Ten years of antidepressants that don’t quite work, or work briefly and then stop, or trigger something that looks suspiciously like hypomania but nobody mentions it because nobody thought to ask.

The fact that it takes a reality TV star checking out of a treatment facility, or a date on a calendar, or a Substack algorithm to get someone to the information they need—that’s not a feel-good awareness story. That’s a diagnostic system failing in plain sight.

The quiet ones

I wrote Living with Bipolar II because I spent years inside that diagnostic gap myself. Misdiagnosed, medicated for the wrong thing, compensating in ways that looked like high functioning and felt like survival. The book exists for the people who suspect something doesn’t add up but can’t find language for what they’re experiencing, because the clinical system keeps handing them the wrong vocabulary.

If you’re one of the people who found this post through whatever algorithmic accident brought you here, I’m glad you did. And if you know someone who’s been cycling through depression treatments that never quite land, who has periods of sharp productivity followed by crashes they can’t explain, who has been told they’re just anxious, or stressed, or not trying hard enough—it might be worth a conversation.

World Bipolar Day is 30 March. Vincent van Gogh’s birthday, if you want something to mention at dinner. The awareness campaign is fine. I’d rather people didn’t need an awareness day to find accurate information about their own brain. But until the diagnostic system catches up, we work with what we’ve got.

Living with Bipolar II: Understanding, managing, thriving is available on all the usual book sites. And free for paid subscribers here on Substack: quiethalf.substack.com/subscribe

This is Part 2. If you missed Part 1, go back and read it first. Or don’t. You’re an adult. But you’ll be missing the bit about Jake the Muss, and that’s really the key to all of this.

In Part 1, we walked through what America actually is when you strip away the mythology: a country that can’t educate its teenagers, can’t keep its mothers alive, can’t protect its children from traffickers, locks up more of its own people than any democracy on Earth, and exports broken psychiatry, toxic social media platforms, and weaponised religion to every corner of the globe.

That was the domestic picture. This is the part where America’s dysfunction stops being an internal matter and starts landing on the rest of us. This is where I get genuinely angry, Karen. Because your country’s choices are now destroying things that belong to all of us.

Pull up a chair. Or don’t. The table’s about to flip either way.

Science: the crown jewel, smashed on the floor

If there was one area where American exceptionalism had a legitimate claim, it was science. Was. Past tense.

In 2024, China overtook the United States in publications in the world’s most elite scientific journals. Nature Index: China 37,273, America 31,900. The lead quadrupled in a single year. China now produces nearly twice the total research output with a 17% advantage in elite publications. The US share of highly cited articles has been declining since 2014. Physics publications down 31% since 2010. Materials science down 26%. These aren’t fringe fields. These are the foundations of technological supremacy, and America is watching them crumble like its bridges.

China produces 77,000 STEM PhDs a year versus America’s 40,000. China now employs more researchers than America and the entire EU combined.

And then you Americans voted for Trump. Again.

7,800 research grants cancelled or suspended. The NIH cut 2,100 grants worth US$9.5 billion. NSF new grants down 25%. The proposed 2026 budget: NSF cut 57%, NASA 24%, NIH 40%. Non-defence science spending slashed to 1991 levels. More than 25,000 people fled the science agencies. PhD enrolments flatlined.

America isn’t just falling behind. It is deliberately, systematically vandalising its own scientific infrastructure. It’s like watching someone burn their own library because they don’t like one of the books.

'There is a cult of ignorance in the United States, and there has always been. Anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that my ignorance is just as good as your knowledge.' — Isaac Asimov, 1980

Culture: the world did not begin with Hollywood

A brief palate cleanser, because we all need a breath.

The Iliad: 2,800 years old. The Epic of Gilgamesh: 4,000. Chinese poetry flourished when Rome was still a village of mud huts arguing about drainage. Islamic scholars were preserving and advancing Greek philosophy, mathematics, and medicine while Europe was busy dying of plague and burning people for knowing too much. Japanese woodblock printing was a sophisticated art form centuries before Gutenberg had his bright idea. Aboriginal Australians have the oldest continuous culture on Earth: 65,000 years. Sixty-five thousand years, Karen. The United States is 249 years old. It’s a teenager in civilisational terms, and it’s behaving like one.

Vietnamese culture spans millennia of poetry, philosophy, textile art, and architecture. I live here. I see it every day. It is rich, deep, layered, and entirely unconcerned with what Americans think about it.

Jazz, blues, hip-hop: genuine American cultural achievements. But they exist alongside, not above, the rest of human civilisation. The assumption otherwise is the arrogance of a country that has never looked over the fence. And the fence, Karen, is lower than you think.

Right. Palate cleansed. Back to the wreckage.

The opioid crisis: America’s pharmaceutical genocide

Here’s a uniquely American story, and by ‘uniquely American’ I mean no other wealthy country on Earth managed to produce anything this comprehensively evil.

A family of billionaires, the Sacklers, owned a company called Purdue Pharma. They manufactured a drug called OxyContin. They knew it was addictive. Their own testing in 1995 showed that 68% of the oxycodone could be extracted from a single pill when crushed. They marketed it as safe anyway. Their memo to the sales team read: ‘Your priority is to Sell, Sell, Sell OxyContin.’ Capitalisation theirs, by the way. Even the grammar was aggressive.

They hired 3,000 doctors for their speakers’ bureau. They paid pain specialists millions to contradict earlier warnings about addiction. They told physicians the risk of addiction was under one per cent. They grew sales from US$48 million in 1996 to US$1.1 billion by 2000. They generated US$35 billion in revenue by 2017. And they killed over 700,000 Americans.

Seven hundred thousand. More than every American combat death in every war since the Civil War combined. A family of billionaires killed more Americans than America’s enemies have managed in 160 years. And they’re still billionaires.

In nearly half of all cases, addiction began with a doctor’s prescription. Eighty per cent of people who ended up on heroin first misused a prescription opioid. Purdue pleaded guilty twice to federal charges about lying, in 2007 and again in 2020. The Sacklers withdrew more than US$10 billion from the company before filing for bankruptcy, most of it in offshore accounts. They’ve now agreed to a US$7.4 billion settlement. They admitted no wrongdoing. Their names have been scraped off museums and universities, but they keep their billions. Crime pays, Karen, if you do it from a boardroom.

In 2019, Rolling Stone called it ‘a uniquely American problem.’ Prescription rates for opioids were 40% higher than in other developed countries. As one professor put it: ‘Most insurance, especially for poor people, won’t pay for anything but a pill.’

Overdoses remain the leading cause of death for Americans aged 18 to 44. And the Trump administration’s response? Propose shuttering the Substance Abuse and Mental Health Administration while classifying fentanyl as a ‘weapon of mass destruction.’ When in doubt, militarise it. That’s the American way.

No other wealthy country has anything remotely like this. Because no other wealthy country allows pharmaceutical corporations to buy doctors, lie to patients, addict a nation, and then negotiate their way out while keeping the fortune. That’s not a healthcare system. That’s a business model built on corpses.

And speaking of things America puts in its people’s bodies that the rest of the world wouldn’t tolerate.

Food: eating things that are illegal in most of the civilised world

Americans are literally eating substances that are banned in the European Union, the United Kingdom, and Australia. Not occasionally. Not in trace amounts. Routinely, in everyday products on every supermarket shelf.

Titanium dioxide, a whitening agent linked to DNA damage, is banned in the EU but sits happily in American Skittles, coffee creamer, and salad dressing. Brominated vegetable oil, which leaves residue on body fat, was in Mountain Dew until recently. Potassium bromate, a probable carcinogen, is in American bread. Azodicarbonamide, a bleaching agent that produces carcinogens when baked, is in bread rolls. Artificial food dyes banned across Europe are in Froot Loops, Lucky Charms, Gatorade, Twinkies, and Little Debbie Swiss Rolls. Skittles had to be reformulated for Europe because the American version contained too many banned ingredients. Let that sink in. Skittles. Reformulated. Because the American version was too toxic for Europeans.

The US allows the food industry to ‘self-certify’ that a substance is safe under a loophole called GRAS, Generally Recognised as Safe. Nearly 99% of new chemicals introduced into the American food supply between 2000 and 2021 came through this self-certification, not FDA review. The foxes are guarding the henhouse, and the hens are full of brominated vegetable oil.

The EU: prove it’s safe before you put it in food. America: put it in food and wait for someone to prove it’s dangerous. And when they do, fight the ban for years.

You are what you eat, Karen. And what America eats is illegal in most of the developed world.

Labour rights: the only wealthy country that treats work as indentured servitude

The United States is the only wealthy nation that does not guarantee a single day of paid vacation. Not one. France: 30. Australia: 20 as a legal minimum. America: whatever your employer feels like giving you, which might be nothing, and you’ll say thank you because the alternative is unemployment in a country with no safety net.

No federal requirement for paid sick leave. At-will employment means your employer can fire you for any reason or no reason, with no notice. Union membership crushed from a third of workers to barely 10%. Amazon warehouse workers documented urinating in bottles because bathroom breaks risk their productivity targets. In the twenty-first century. In the richest country on Earth. People pissing in bottles so Jeff Bezos can build another rocket.

The federal minimum wage has been US$7.25 an hour since 2009. Sixteen years without a raise. A full-time minimum wage worker earns roughly US$15,000 a year. Below the poverty line. For working full-time. In America.

‘The problem is not that people are taxed too little. The problem is that government spends too much.’ — Ronald Reagan

Every other wealthy country decided workers deserved basic protections. America decided corporations deserved maximum flexibility. The result is a workforce that is overworked, underpaid, unprotected, and terrified of losing the health insurance tied to their employment, because in America, the right to not die of a treatable illness is a perk of employment rather than a feature of civilisation.

Student debt: punishing your young for the crime of learning

Total US student loan debt: approximately US$1.7 trillion. Owed by 43 million Americans. The average graduate carries about US$37,000. An eighteen-year-old can sign a loan document they barely understand, accumulate six figures of debt before they can legally drink a beer, and spend thirty years paying it off at interest rates that would make a loan shark wince.

Germany: university is essentially free. Australia: you don’t repay until you earn above a threshold, and the debt dies with you. Scandinavia: students are paid a stipend to attend. Scotland: free for Scottish students.

'If a nation expects to be ignorant and free, it expects what never was and never will be.' — Thomas Jefferson

America decided education should be a market commodity rather than a public good. The same country that spends US$886 billion on its military decided investing in the minds of its own citizens was someone else’s problem. Then it wonders why a third of its teenagers can’t do basic maths. Cause and effect, Karen. It’s not complicated.

Infrastructure: a first-world country with third-world pipes

In 2014, Flint, Michigan, switched its water supply to save money. Lead leached into the drinking water. For eighteen months, residents, predominantly Black and low-income, drank, cooked with, and bathed their children in lead-contaminated water while officials insisted it was safe. Children suffered irreversible neurological damage. A generation’s brain development, permanently compromised, to save a few dollars on water treatment.

'A nation that continues to spend more on military defence than on programmes of social uplift is approaching spiritual death.' — Martin Luther King Jr.

In February 2021, the Texas power grid collapsed in a winter storm. The grid had been deliberately kept separate from the national system to avoid federal regulation, because Texans would rather freeze to death than accept government oversight. At least 246 people died. The state blamed renewable energy, despite the failures being overwhelmingly in gas, coal, and nuclear infrastructure that hadn’t been winterised because regulation was considered an affront to freedom. Freedom from electricity, apparently.

In Jackson, Mississippi, 2022, the water treatment system failed completely. 150,000 people, in a city 82% Black, with no safe drinking water. Schools closed. Hospitals brought in tanker trucks.

The American Society of Civil Engineers gives the nation’s infrastructure a C-minus. Forty-two per cent of bridges are at least 50 years old. This is the richest country in the history of civilisation. It can fund 750 military bases in 80 countries but cannot provide clean drinking water to its own citizens or keep the lights on when it gets cold. But sure, tell me again about American greatness.

Democracy: for sale to the highest bidder

The 2024 election cycle: US$14 billion in total political spending. Fourteen billion dollars spent choosing leaders. That’s not democracy. That’s an auction with bunting.

Citizens United ruled that corporations are people and money is speech. Gerrymandering predetermines election outcomes in some states before a single vote is cast. Voter suppression laws target Black and minority communities. Twice in living memory, in 2000 and 2016, the popular vote winner lost the presidency through the Electoral College, a system designed partly to protect the interests of slave-owning states. A vote in Wyoming carries 3.6 times the weight of a vote in California.

The Economist Intelligence Unit ranks the United States 36th on its Democracy Index. Classified: ‘flawed democracy.’ Behind Uruguay, Costa Rica, and Taiwan. The country that lectures the world about democratic values can’t even run a clean election without billions of dollars of corporate money picking the winner in advance.

And what does this flawed democracy spend most of its money on? Oh, Karen. Glad you asked.

The military-industrial complex: the world’s most expensive hammer looking for nails

'In the councils of government, we must guard against the acquisition of unwarranted influence by the military-industrial complex. The potential for the disastrous rise of misplaced power exists and will persist.' — Dwight D. Eisenhower, Farewell Address, 1961

The United States spends more on its military than the next ten countries combined. US$886 billion in 2024. Over 750 bases in 80+ countries. The only country to have used nuclear weapons in war. Since 1945, it has bombed, invaded, or conducted military operations in Korea, Vietnam, Laos, Cambodia, Lebanon, Grenada, Libya, Panama, Iraq (twice), Somalia, Bosnia, Kosovo, Afghanistan, Yemen, Syria, Pakistan, and now Iran. The list is not exhaustive. I ran out of breath.

Eisenhower warned about the military-industrial complex in 1961. Sixty-five years later, it is the single most powerful economic force in American politics. The F-35 programme alone has cost over US$1.7 trillion across its lifetime. That’s the same as the entire student debt crisis. America could have educated every young person in the country for the cost of one fighter jet programme. It chose the jets.

The country cannot provide clean water to Flint. Cannot keep the power on in Texas. Cannot guarantee new mothers a day of paid leave. Cannot stop its pharmaceutical companies killing 700,000 citizens. But it can absolutely, definitely, without question, project military force onto every continent simultaneously.

Priorities.

The scorecard from hell: every time America ‘liberated’ someone

The military budget isn’t abstract. It produces outcomes. Let’s look at them. Because the record reads like a horror novel written by a drunk psychopath with a crayon.

Iran, 1953. Iran had a democratically elected Prime Minister, Mohammad Mossadegh. He nationalised Iran’s oil because he thought maybe Iranians should profit from Iranian oil. Radical concept. So the CIA sent Kermit Roosevelt (yes, Teddy’s grandson; yes, named after a Muppet) to Tehran with suitcases full of cash, bribed newspaper editors, hired people to riot, and pulled off a coup in four days. Around 300 people died. They installed the Shah, who signed over 40% of Iran’s oil fields to American companies as a thank-you card. The Shah then ruled for 26 years with the help of SAVAK, his secret police, trained by the CIA. And what grew in the shadow of that 26 years of Western-installed oppression? The Islamic Revolution of 1979. The regime everyone now wants America to go and remove exists because of the last time America went and removed someone. The CIA publicly admitted it in 2013.

Guatemala, 1954. One year after Iran. The CIA overthrew the democratically elected President Jacobo Arbenz because he wanted land reform that threatened the United Fruit Company’s profits. An American banana company. The United States overthrew a sovereign democracy so rich people could keep making money off bananas. What followed: decades of military juntas, a 40-year civil war, more than 200,000 people killed, mostly Indigenous civilians. Today, Guatemala is so destroyed by poverty and organised crime that its people risk their lives crossing deserts to reach the US border. And then Americans have the nerve to ask why there’s a migrant crisis. You made the migrant crisis. Over bananas.

Chile, 1973. Salvador Allende, democratically elected socialist president. Kissinger couldn’t stomach it. The CIA destabilised the country, backed a military coup, and in came Pinochet. Seventeen years. Over 3,000 killed. Tens of thousands tortured. Political opponents disappeared. Thrown from helicopters. But at least the socialists were gone.

Afghanistan, 1979-2021. In the 1980s, the US armed the mujahideen to fight the Soviets. Billions in weapons. Stinger missiles handed out like party favours. The Soviets left. America popped the champagne and walked away. Then came civil war, the Taliban, and Al Qaeda, all using the weapons, training, and networks America had paid for. Which led to September 11. Which led to Afghanistan Round Two. Twenty years. Over 2,000 American soldiers dead. Trillions of dollars. And the moment the US pulled out, the Taliban took the country back in days. Not months. Days. Afghan women today can’t speak in public. The most expensive own goal in human history.

Iraq, 2003. The magnum opus. They fabricated an entire war. Weapons of mass destruction that didn’t exist. Colin Powell stood at the United Nations with a vial of powder and lied to the entire planet. They removed Saddam, then some genius decided to disband the entire Iraqi military overnight. Hundreds of thousands of armed, trained, furious men suddenly unemployed. What grew out of that? ISIS. The most brutal terrorist organisation the modern world had seen. America spent trillions removing one dictator and accidentally created a death cult that made the dictator look like a school principal.

Libya, 2011. NATO forces helped overthrow Gaddafi. Obama later called it the biggest mistake of his presidency. Because there was no plan for what came next. Libya collapsed into a failed state. Open-air slave markets. People being sold at auction in a country the West ‘liberated.’

The pattern is the same every time. America can topple a government in days. Five stars. No notes. The part they have never once figured out is what comes next. The Cato Institute reviewed the entire history and concluded that regime change is more likely to fail than succeed, and the most common outcome is less democracy, not more. Countries are more likely to experience civil war or mass killings after a foreign-imposed overthrow than before one.

Name one that worked. Just one. Name one where the people of that country would look back and say, thank God America intervened, everything turned out great.

I’ll wait so long I’ll fossilise.

And now they’re doing it again. Iran. Same region. Same strategy. Same people who’ve never opened a history book. As Gman at I Fucking Love Australia put it: at some point, doing the same thing over and over and expecting a different result isn’t foreign policy. It’s insanity.

A note on the Americans I know

I need to say this before we go any further, because I meant what I said at the beginning: I don’t hate America.

I have American friends. Good ones. Decades-long friendships. Intelligent, articulate, educated people who have enriched my life and continue to challenge my thinking in ways I’m grateful for. They almost certainly didn’t vote for Trump the first time, and they definitely didn’t the second. They are horrified by what is happening to their country. They are grieving.

But individual virtue runs out of road. America voted for him. Twice. After the insurrection. After the convictions. After all of it. ‘I didn’t vote for him’ doesn’t absolve a nation. He got in because enough people wanted him or enough people couldn’t be bothered to stop him. We tried to warn you, voices inside and outside your country, and you wouldn’t listen.

The rest of the world is dealing with the consequences.

Captain Impulse Control and the coming dark age

The Mango Mussolini. Captain Impulse Control. This is who America chose to represent it on a world stage that, it bears repeating, exists outside of America. And now the consequences are everyone’s problem.

Trade war against the entire planet. Tariff rates past Great Depression levels. J.P. Morgan: 60% global recession probability. US GDP contracted in Q1 2025. Global growth below 3% for the first time since the pandemic. Even his own supporter Bill Ackman called it ‘economic nuclear war on every country in the world.’ When your own cheerleaders are reaching for nuclear metaphors, Karen, that’s not a good sign.

Then another Middle East war. Your country ran this scam before. Iraq. A million dead. At least Bush had the courtesy to fabricate evidence. He got Colin Powell to stand up at the UN with a vial and a PowerPoint. It was bullshit, but it was effort. There was a storyline. A narrative arc. Trump just bombed Iran like he was ordering room service. Operation Epic Fuckup.

Bush destabilised Iraq over years. Trump has torched the global oil supply in what feels like a long weekend. Somewhere in Texas, George W. Bush is painting a dog and whispering, ‘Thank fuck. I am no longer the worst US president in history.’ Twenty years of being the punchline, and now he’s not even in the conversation.

Destroying the machinery of civilisation

Day one: Trump declared an ‘energy emergency,’ gutted renewables. Blocked wind farms. Killed solar projects. Cancelled US$13 billion in clean energy. Instructed the Department of Energy to ban the words ‘climate change,’ ‘emissions,’ ‘green,’ and ‘clean energy.’ Orwell would have wept, then written a sequel. Pulled out of the Paris Agreement. Withdrew from 66 international organisations and treaties. The EPA now considers only costs to industry, not human lives. Oil and gas pollution kills 91,000 Americans a year. The EPA no longer cares.

The Iran war closed the Strait of Hormuz. Oil past US$108 a barrel. Gasoline up 93 cents in three weeks. Refineries across the Gulf states targeted and damaged. Captain Impulse Control’s response? Posted on Truth Social from his golf resort, on a Saturday night, threatening to blow up all 400 of Iran’s power plants. Signed off: ‘Thank you for your attention to this matter.’ Like a strata notice about bins.

Those 400 power plants run the refineries, the pumping stations, the entire production chain. He wants the Strait open so oil flows, and his plan is to destroy the infrastructure that produces the oil. That’s burning down the brewery because the pub’s closed. Rebuilding takes three to five years minimum. He’s baked in US$100-plus oil for a decade. Maybe permanently. But at least the sign-off was polite.

Cuba. Your country blockaded its fuel supply. The entire power grid collapsed. Eleven million people in total blackout. Hospitals dark. Children in houses they can’t cool eating food they can’t keep cold. And when the US Embassy in Havana ran out of diesel, because, you know, they’re on an island your country blockaded, they asked Cuba for permission to import two containers of fuel. Just for the Americans. And Cuba said no. Called it ‘shameless.’ You blockade an island. Your own embassy runs dry because of your own blockade. You ask the country you’re strangling to make an exception for you. They tell you to sit in the dark like everyone else. Magnificent.

When Trump begged Europe for help reopening the Strait, a French general went on live television and told him to go and get fucked. Really, those words. In French, obviously, which somehow makes it even more devastating. From the country that invented modern diplomacy. Zero fucks given, zero fucks implied, zero fucks available upon request. This is what happens when you spend years calling allies freeloaders, threatening to invade Greenland, slapping tariffs on their exports, then ringing them when the oil stops. They didn’t forget. They never will.

He gutted renewable energy, the thing that insulates against fossil fuel supply shocks. Then created the supply shock. There are no price spikes for sunlight, Karen. No embargoes on the wind. But Trump killed the wind farms and now the oil isn’t flowing either.

Australia has 50 days of fuel reserves. We refine almost nothing. Our economy runs on imported diesel. And here’s an irony that should keep American strategists awake: you can’t build bombs without our dirt. Australia supplies the critical minerals the US military depends on. Lithium, rare earths, cobalt. Every missile, every jet. Trump tariffed our exports while starting a war that needs our resources. He’s demanded the US manufacture more weapons, more bombs, more everything, immediately. With what minerals, exactly? Strategic genius.

This world runs on energy. Every factory, hospital, transport network, cold chain, heating system. All of it being destroyed from both ends: renewables gutted by policy, fossil fuels destroyed by war. The world’s ability to feed, clothe, move, and medicate eight billion people is being degraded by one man, in one country, that the rest of us never voted for.

We are sliding, with terrifying speed, towards an economic meltdown of a scale the modern world has never seen. Not a recession. A systemic unwinding. And the man at the controls formats threats against civilian infrastructure like a letter from his accountant.

So about THAT expression

‘Calm your tits’ is an Australian colloquialism. It means ‘settle down.’ It’s gender-neutral. Informal, cheeky, and about as offensive in Australian English as telling someone to keep their shirt on.

You don’t have to like it, Karen. But the assumption that your discomfort makes it objectively wrong is the problem I’ve been describing for the last twelve thousand words. Your cultural lens is not the only lens. And right now, your country’s offences against the rest of the world are so vast, so reckless, so catastrophically stupid, that getting upset about an Australian idiom is like complaining about the in-flight music while the plane is on fire.

Number one in healthcare spending (all going to the companies, by the way, not the people). Number one in incarceration. Number one in gun deaths among wealthy nations. Number one in maternal mortality among developed countries. The only wealthy nation that doesn’t guarantee new mothers a single day of paid leave. A foster care system that feeds children into the sex trade. A psychiatric framework that misdiagnoses millions worldwide. Social media platforms that have done more damage to global mental health than any drug cartel. A pharmaceutical industry that killed 700,000 of its own citizens and kept the profits. Food full of substances banned in the rest of the civilised world. Workers pissing in bottles. Students drowning in debt. Bridges falling down. Water full of lead. Democracy for sale. And a military budget larger than civilisation’s capacity to comprehend it.

Now actively dismantling its own science, torching its alliances, and dragging the rest of the world towards a catastrophe none of us asked for.

But sure. Greatest country in the world.

A final word, for Karen from Ohio

The world may eventually forgive you, Karen. Forgive you and the millions like you who voted this man in, or couldn’t be bothered to vote at all, or shrugged and said ‘both sides are the same’ while the rest of us watched in horror. We tried to warn you, voices both inside and outside your country. You wouldn’t listen.

It may take decades. A generation. Longer than any of us will be alive to see.

Your country will not be welcome on the world stage for a very long time. Your passport will carry a different weight. Your diplomats will enter rooms with a deficit of trust that no handshake will overcome. Your alliances, the ones your grandparents built and your parents maintained and you destroyed, will take decades to rebuild. If they can be rebuilt at all. Some bridges, once burned, stay burned. And given the state of your infrastructure, you’re not great at building bridges anyway.

The world will recover. It always does. Trade routes will reroute around you. Energy systems will be rebuilt without you. Scientific partnerships will form without you. The rest of us will find our way through the wreckage your country created, because we have no other choice.

But the world will never forget.

It will never forget that the United States of America, the self-proclaimed leader of the free world, voted twice for a man who launched trade wars on 100 countries simultaneously, started a Middle East war without a reason, threatened to destroy civilian infrastructure from a golf resort, blockaded an island and then asked for its diesel, gutted its own science, poisoned global discourse through its social media platforms, exported a broken psychiatric framework to every country on Earth, allowed pharmaceutical corporations to kill 700,000 of its own citizens, fed its children food that’s illegal elsewhere, abandoned its own mothers, caged more of its own people than any democracy in history, and let its foster care system feed children to sex traffickers.

That’s your legacy now, Karen. Not jazz. Not the moon landing. Not the Marshall Plan. This. This is what the world will remember about your generation of Americans.

So when an Australian tells you to calm your tits, take a breath. Look around at what your country has done to the rest of us. Consider the possibility that the problem here is not the idiom.

The problem is you.

Remember Jake the Muss, Karen. Remember the quiet Kiwi at the bar. The jaw setting. The light flickering behind the eyes. The table about to flip.

The whole world is standing up now. Slowly.

Calm your tits.

In a recent Note I told the AI-haters to ‘calm your tits’.

Karen from Ohio took offence at an Australian expression. An American, offended by something someone from another culture said. Truly unprecedented. Alert the media. Oh wait, your media’s busy covering whatever Captain Impulse Control tweeted from the golf course at 3am. Never mind.

And like all the best keyboard warriors, you didn’t have the guts to say it in public. You slid into my DMs with your snottogram, Karen from Ohio, because apparently confrontation is only acceptable when you’re doing it from behind a screen where nobody can see you. Very brave. Very on-brand.

Let me tell you something about Australians. We have non-biological cousins. We call them Kiwis. New Zealanders are some of the nicest, politest, most generous people on this Earth. They will sit there at the bar while you rip into them, take the piss out of them, tell them why they’re wrong about everything you know is right. They’ll smile. They’ll nod. They’ll buy you another round.

But everyone at the bar recognises when our Kiwi cousin has had enough. It’s not like they stop smiling. It’s more the way their jaw sets. A light flickers, so subtly, behind their eyes that the hapless newcomer doesn’t see it happen. But the locals know. And the locals leave the bar via the nearest door or window, because they know what is coming next.

Our cousins will come over and upturn your table, then give you a pummelling so intense your embassy will find it hard to recognise you, and you will never, ever forget it. If you’ve seen Once Were Warriors, you know Jake the Muss. Sitting quietly in the pub, minding his own business, until some dickhead says the wrong thing and suddenly furniture is airborne, teeth are on the floor, and grown men are climbing out bathroom windows.

I am an Australian Jake the Muss.

You sent me your little DM. You poked the Aussie. And now you’re going to hear what the rest of the world actually thinks about your country. Not because I hate America. I have dear American friends, and I’ll get to that. But because someone needs to say it with the data to back it up, and I am entirely out of patience.

Your offence doesn’t interest me, Karen. What interests me is what it reveals: the reflexive assumption that American cultural norms are universal norms. That the way things are done in the United States of America is the way things are done, full stop, everyone else can adjust.

They’re not. There is a vast, ancient, complex world outside of your borders. A world with millennia of culture, art, language, philosophy, science, and humour that was already old when your country was still a tax dispute with the British. A world that, by virtually every measurable metric of human flourishing, is doing a damn sight better than America right now. And a world that is increasingly furious about being dragged into the consequences of American stupidity.

'It is not that I am not a fan of American exceptionalism. That is like saying I am not a fan of the moon being made out of green cheese. It does not exist.' — Noam Chomsky

Don’t take my word for it. I’m just an Australian psychologist sitting in a café in the Vietnamese highlands, drinking cà phê sữa đá and wondering how we got here. But the data doesn’t care about my feelings or yours. Let’s walk through it. Every number sourced, verified, and damning.

Education: a nation of people who can’t do basic maths

Let’s start where it all begins. The classroom. Because the ignorance has to come from somewhere, and Karen, I have a theory about where.

The PISA assessment tests 15-year-olds across 81 countries. In 2022, the United States ranked 26th in mathematics. Twenty-sixth. Singapore topped all categories, in case you’re wondering what competence looks like. American maths scores were the lowest in the assessment’s entire history. More than a third of American teenagers could not compare distances between two routes or convert prices into a different currency. A third of your teenagers can’t manage the basics of functioning in a world that uses numbers. No wonder you elected a man who thinks tariffs are paid by other countries.

And here’s the part that should make you spit your coffee out: 39% of American students who sat the test were among the most socioeconomically advantaged students globally. They had every material advantage money could buy. Private schools. Tutors. Laptops. The works. They still couldn’t keep up with kids in Estonia. You’re not failing because you’re poor. You’re failing because your system is broken and you refuse to see it.

'A lot of the educational system is designed to make people as stupid and ignorant as it is possible for a human being to be.' — Noam Chomsky, paraphrasing Adam Smith

But education is only the first domino. Because when you don’t educate your population, the consequences cascade through everything else. Starting with what happens when those poorly educated people get sick.

Healthcare: spending the most, dying the youngest

The United States spent US$13,432 per person on healthcare in 2023. Over US$3,500 more than the next highest country, Switzerland. Nearly double the average of comparable wealthy nations. For that kind of money, you’d expect Americans to live forever. Or at least longer than the Poles.

You don’t. The result? The lowest life expectancy among large, wealthy countries. 78.4 years, ranking 30th out of 38 OECD countries. On par with Poland. Now, Poland is a perfectly fine country with excellent pierogis, but it’s probably not the benchmark the Greatest Nation on Earth was aiming for.

'Of all the forms of inequality, injustice in healthcare is the most shocking and inhumane.' — Martin Luther King Jr., 1966

Australia: 84.1 years. The US: outside the top 40 globally. And getting worse. Forecast to drop from 49th to 66th by 2050. Drug use, obesity, high blood sugar, high blood pressure. Americans are dying younger while their healthcare system hoovers up more money per capita than any nation on Earth and delivers it straight to insurance company shareholders.

You spend the most. You get the least. You die the youngest. And you call it the greatest healthcare system in the world. If that’s not a symptom of a country that can’t think straight, I don’t know what is.

And nowhere is the human cost more obscene than in what America does to its mothers.

Maternal mortality: killing your own mothers

The US maternal mortality rate: 22.3 deaths per 100,000 live births in 2022. More than 55% higher than Chile, the nation with the second-highest rate among high-income countries. Norway: zero. Zero, Karen. As in, none. Switzerland: 1.2. Sweden: about 3. The US ratio is three times higher than Sweden, Japan, the Netherlands, Germany, the UK, and France.

‘The true measure of any society can be found in how it treats its most vulnerable members.’ — Mahatma Gandhi (attributed)

Among Black women in America: nearly 50 deaths per 100,000. In the richest country on Earth. In the twenty-first century. Women are dying in American hospitals at rates that would shame most of the developed world, and the country can’t even be bothered to send someone round afterwards to check on the ones who survive.

The United States is the only high-income country that does not guarantee provider home visits or paid parental leave after birth. The only one. Every other wealthy nation looked at new mothers and said, ‘You need support.’ America looked at new mothers and said, ‘Good luck. Don’t forget to fill out your insurance paperwork on the way out.’

Which brings us to the broader question of whether America cares about its people at all. Spoiler: the evidence is not encouraging.

Paid parental leave: the only wealthy nation that doesn’t give a damn

The United States is the only country among 41 OECD nations that does not mandate any paid leave for new parents. Zero at the federal level. Estonia offers more than a year and a half. Thirty-seven of 38 OECD countries offer paid maternity leave, averaging 17.3 weeks.

The richest country on Earth wraps itself in the flag, weeps about family values every Sunday and election year, and then tells a woman who’s just given birth to get back to work or starve. The cognitive dissonance would be impressive if it weren’t so lethal.

Child poverty: a deliberate cruelty

Relative child poverty exceeds 20%. Norway: 5%. The US rate is twice as high as the UK, Sweden, or France. The US spends about 1% of GDP on family benefits. The UK and Ireland spend 4%.

This is not an accident. This is not a funding shortfall. This is not ‘we’d love to help but the money just isn’t there.’ This is a policy choice. America chooses, actively and repeatedly, to let its children grow up in poverty while spending more on its military than the next ten countries combined. Priorities, Karen. Your country has them exactly backwards.

And what happens to those children when the system is done failing them? Brace yourself, Karen, because the answer is worse than you think.

The foster care pipeline: what America does to its own children

I have written a book called The Convenient Monster. It examines the global child sexual exploitation trade from the demand side: the market economics, the systemic failures, the questions nobody wants to ask. And one of the most damning findings concerns the United States itself.

The Polaris Project and multiple state-level investigations have identified the American foster care system as a direct pipeline to sex trafficking. Children age out at eighteen with no support, no housing, no money. Traffickers know this. They wait. In some documented cases, they recruit inside the care facilities themselves, building relationships with children months before they age out. The underage-sex market doesn’t need to cross an ocean. It’s waiting at the care home door.

The National Center for Missing & Exploited Children reported over 32 million reports of suspected child sexual exploitation material online in a single year. Thirty-two million. The number is so large the brain can’t hold it. It becomes noise. Which is, I suspect, partly why nothing proportionate to the scale has been done about it.

The mechanism is identical to what operates in Thailand, the Philippines, or Cambodia. Vulnerability plus economic desperation plus inadequate protection equals supply. The only difference is that when it happens in Ohio, nobody makes a slow-motion documentary with obligatory melancholic, sparse piano. Because the comfortable narrative requires the problem to be over there. Among them. Not us.

'The test of the morality of a society is what it does for its children.' — Dietrich Bonhoeffer

A country with these numbers has no moral authority to lecture anyone about protecting children. None whatsoever. Does that stop you? No, you love pontificating about how badly Johnny Foreigner treats their children. You have reports and publications and conferences about it. Entire careers built on pointing fingers at the developing world while your own foster care system operates as a recruitment agency for predators.

But America’s talent for exporting dysfunction doesn’t stop at its borders. It also exports its psychiatric framework. And that framework is doing damage worldwide.

Psychiatric hegemony: exporting broken thinking

Full disclosure: this one’s personal.

I am a counselling psychologist. I hold a Master’s in Counselling Practice and my academic research has generated over 450 citations across 25 years. (That means it’s still being cited a quarter of a century later, which is indicative of something, but I’m not entirely sure what. Proof that I am both smart and that I also know nothing, I guess.) At the age of 66, I was diagnosed with AuDHD—autism and ADHD occurring together—after spending decades being medicated for bipolar II disorder that I almost certainly never had. Thirty years of wrong medication, wrong diagnosis, wrong framework. My book Misdiagnosed documents this in full, and believe me, it’s not a fun read. Enlightening, yes. Fun, no.

A psychiatrist who oversaw my care in Australia, Dr Amel Hmam, abandoned me as a patient mid-treatment, a patient with documented recent suicidal ideation, because I had the audacity to lodge a written complaint about her chronic lateness and missed appointments. She fired me by deleting every future appointment without consultation. That’s not a personality conflict. That’s a systematic professional failure with potentially lethal consequences. I filed a formal complaint with the Australian Health Practitioner Regulation Agency. Their response: they’d look into it, but I’d never know the outcome. Regulatory opacity protecting incompetent practitioners while patients remain in the dark. Outstanding.

But here’s where it becomes an American export problem. The diagnostic framework that enabled my misdiagnosis, the DSM—the Diagnostic and Statistical Manual of Mental Disorders—is an American product. Published by the American Psychiatric Association. It is the dominant diagnostic framework used by psychiatrists worldwide. Its categories, its assumptions, its pharmaceutical entanglements shape how mental illness is understood, diagnosed, and treated in countries that had no say in writing it.

'It is no measure of health to be well adjusted to a profoundly sick society.' — Jiddu Krishnamurti

The chemical imbalance theory of depression, the one used to justify prescribing SSRIs to hundreds of millions of people globally, was an American export. The research evidence doesn’t support the theory. At all. But it was phenomenally profitable for American pharmaceutical companies, and so it travelled. It crossed every border. It shaped psychiatric practice in Australia, in Europe, in Asia, in countries whose own research traditions and cultural understandings of mental distress were bulldozed by American diagnostic hegemony.

When American psychiatry gets it wrong, the whole world gets it wrong. Because American psychiatry, like American foreign policy, assumes its framework is universal. It isn’t. And the bodies, the decades of wrong medication, the misdiagnoses, the lives diminished by a framework that mistakes neurodivergence for mental illness, those are on America’s tab too.

The DSM isn’t the only American product that’s damaged the rest of the world, of course. The technology sector has managed something equally destructive, with added surveillance capitalism and slightly better graphic design.

Social media: America’s gift to global mental health destruction

I was in the vanguard of social media for business, back when it first started getting public attention in 2005. By 2009 I was being flown around the world to lecture on how social media was an incredibly powerful tool for human connection, for business transformation, for democratic participation. I was a true evangelist. I believed in it.

I was wrong. Or rather, I was right about the potential and catastrophically wrong about the people who would control it. Which is, now I think about it, a fairly succinct summary of the American experience generally.

Every major social media platform that has shaped global discourse is an American product. Facebook. Instagram. Twitter. YouTube. All of them. Built in Silicon Valley. Governed by American corporate ethics, which is to say governed by shareholder extraction and almost nothing else.

Meta, the company behind Facebook, Instagram, and WhatsApp, is built on surveillance capitalism, profiling users with a granularity that would make the Stasi blush. It enables, amplifies, and profits from political manipulation, ethnic violence, and disinformation on a planetary scale. It ran secret psychological experiments on 700,000 users without consent. Its own internal research showed Instagram was harmful to teenage girls’ mental health, and it did nothing because fixing it would hurt user growth. It paid a US$5 billion fine, the largest privacy settlement in history, admitted nothing, and carried on. It has been fined €405 million for violating children’s data protections. It builds shadow profiles on people who don’t even have accounts. If you’re reading this and you’ve never had a Facebook account, congratulations: Facebook has a file on you anyway.

Twitter, before Elon Musk bought it and turned it into his personal ideological playground, was something genuinely remarkable. It was imperfect, chaotic, sometimes vicious. But it was also the platform where the Arab Spring found its voice. Where dissidents in authoritarian regimes communicated in real time. Where passengers live-tweeted from plane crashes and the world responded. Where movements like Black Lives Matter organised and amplified. It was the closest thing the internet had produced to a genuine global public square.

Musk bought it and torched it. Gutted the safety teams. Turned verification into a paid attention-ranking scheme. Platformed extremism. Suppressed dissent. Algorithmic preference for the loudest, angriest, least thoughtful voices. An American billionaire took a globally significant communication tool and turned it into a rage-click carnival because he could. Because in America, if you’re rich enough, you can buy the town square and set it on fire and call it free speech. America does love a bully.

‘If you’re not paying for the product, you are the product.’ — commonly attributed to Andrew Lewis, 2010

The void Twitter left has been filled, imperfectly but meaningfully, by Substack. A platform built on direct relationships between writers and readers. Little algorithmic manipulation. No ad economy. Long-form thinking, depth, nuance, and conversation over performance. Ironically, closer to what social media was supposed to be than anything Silicon Valley ever produced. And it is where the writers, thinkers, and citizens driven off other platforms by American corporate enshittification have landed. For now, anyway. We don’t trust it, because you taught us not to.

And then there’s TED, which went from a global platform for innovative, informed thinking to a platform where someone will confidently tell you how to live your life and then coincidentally sell you a book and a course about how to live your life. TED was LinkedIn for the intelligentsia and intelligently gullible. Many of my incoming DMs on Substack begin with words to the effect of, ‘I trust this DM finds you well.’ It does not.

The pattern is always the same. America builds something with genuine potential for human connection, then its corporate culture extracts every dollar of value from it until nothing remains but a behavioural-modification farm. It did it with social media. It did it with healthcare. It did it with education. It does it with everything. It’s like watching someone buy a beautiful old house, gut it for copper wire, and then wonder why the roof fell in.

But social media and Silicon Valley are relatively recent American exports. There’s an older one. One that’s been doing damage for far longer, and far more insidiously.

Religion: God’s own country exports God’s own bigotry

The United States is the most religious wealthy nation on Earth. Not in a quiet, contemplative, tend-your-own-garden kind of way. In a loud, organised, politically weaponised, globally exported kind of way. American evangelical Christianity is not just a domestic phenomenon. It is a multinational operation with a budget, a strategy, and a body count.

The prosperity gospel, the doctrine that God rewards faith with material wealth and that poverty is essentially a spiritual failing, is an American invention. Born in megachurches and televangelism studios, perfected by preachers with private jets and stadium-sized congregations, and then exported to every corner of the developing world. It is now, by many accounts, the most common form of Christianity in sub-Saharan Africa, Latin America, and parts of Asia. It tells some of the poorest people on the planet that if they’re poor, it’s because they don’t believe hard enough. And then it passes the collection plate. Which is, when you think about it, quite the business model.

'When fascism comes to America, it will be wrapped in the flag and carrying a cross.' — commonly attributed to Sinclair Lewis (disputed, but widely cited)

But the prosperity gospel is merely cynical. The real damage is political.

When American evangelicals lost the fight against gay marriage at home in the early 2000s, they didn’t accept defeat. They exported the war. They shifted focus to Africa, particularly Uganda, which they saw as fertile ground due to its conservative Christian majority and young population. In 2009, top American evangelical leaders headlined a three-day conference in Kampala on ‘exposing the homosexuals’ agenda.’ They told audiences that gay Westerners were trying to corrupt and recruit African children. Shortly after, Ugandan politician David Bahati drafted the first version of the Anti-Homosexuality Act.

That law, signed in 2023, imposes life imprisonment for same-sex relations and the death penalty for ‘aggravated homosexuality.’ It makes it a crime to simply identify as LGBTQ+. More than 20 US-based conservative groups have funnelled at least US$54 million into anti-LGBTQ campaigns across Africa since 2007. The Billy Graham Evangelistic Association: US$7.6 million. Family Watch International, an Arizona-based hate group, co-sponsored an inter-parliamentary conference on ‘family values’ in Uganda just weeks before 444 of 557 parliamentarians voted for the bill. The law was partly drafted with the assistance of Jay Sekulow, a personal attorney to Donald Trump.

The grotesque irony: many pre-colonial African belief systems tolerated homosexuality and, in some cases, incorporated LGBTQ+ people into society with named roles and identities. The homophobia now being presented as ‘African values’ is itself an American and colonial import. The evangelicals framed their campaign as resistance to Western imperialism, when it was Western imperialism. American bigotry, dressed up in African clothing, funded by American dollars, killing African people.

Similar bills are being pushed in Ghana, Kenya, Nigeria, Tanzania, and Zambia. The pattern is coordinated. The money is American. The ideology is American. The suffering is everyone else’s.

Meanwhile, at home, the same evangelical movement that exports death-penalty homophobia to Africa constitutes the single most reliable voting bloc for Donald Trump. Eighty-one per cent of white evangelicals voted for him in 2016. They came back for more in 2020 and again in 2024. The movement that tells Ugandans God wants them to execute gay people also told Americans God wanted them to elect a man who raw-dogged a porn star while his wife was home with a newborn. Moral clarity, American style.

The rest of the world has religions too, Karen. Ancient ones. Complex ones. Religions that predate Christianity by millennia. Hinduism, Buddhism, Islam, Indigenous spiritual traditions that stretch back to the dawn of human consciousness. None of them needed American televangelists to explain God to them. None of them asked for the prosperity gospel. None of them invited American hate groups to write their laws.

But they got them anyway. Because America doesn’t just export its products, its pharmaceuticals, its diagnostic manuals, and its social media platforms. It exports its God. And its God, like everything else America exports, arrives with a price tag and an agenda.

And what has all this moral certainty, all this God-given exceptionalism, produced at home? Let’s see.

The children America already stole

During Trump’s first term, the ‘zero tolerance’ policy separated more than 5,000 children from their parents at the US-Mexico border. Babies. Toddlers. Children as young as eight months old, flown to foster facilities in the middle of the night, in states thousands of miles from where their parents were being held, without being told where they were going.

The policy was deliberate. This was not bureaucratic incompetence. This was cruelty by design. Attorney General Jeff Sessions told federal prosecutors in May 2018, in words that became the title of a Human Rights Watch report: ‘We need to take away children.’ When a senior ICE official discovered that parents were being reunited with their kids too quickly after court appearances, he wrote to confirm that ‘the expectation is that we are NOT to reunite the families,’ because reunification ‘obviously undermines the entire effort.’

Read that again. A senior official of the United States government explicitly instructed his colleagues to prevent parents from being returned to their children, because giving children back to their mothers and fathers would undermine the policy objective of using stolen children as a deterrent.

The director of the Office of Refugee Resettlement instructed his staff not to maintain a list of which children had been separated from which parents. A federal judge later observed that the government kept better records of seized property than of the children in its care. Let that land. Better records of confiscated suitcases than of confiscated children. There was no tracking system. No database linking children to their parents. No plan for reunification. Because reunification was never the intention.

'The moral arc of the universe is long, but it bends toward justice.' — Martin Luther King Jr., paraphrasing Theodore Parker

Children were held in cages. Literal cages. Chain-link enclosures in converted warehouses and detention centres. A five-year-old boy who would not speak to anyone. A twelve-year-old who developed suicidal ideation. A one-year-old returned to his mother after three months, dirty and infested with lice. Infants taken at legal points of entry, not from people sneaking across borders in the dead of night, but from families who presented themselves at official crossings and asked for asylum, as international law permits them to do.

Hundreds of parents were rapidly deported without their children. Sent back to Central America while their kids remained in US government custody, with no mechanism to find them and no legal pathway to get them back. Mothers were given the choice between abandoning their asylum claim or abandoning their child. In the United States of America. In the twenty-first century.

By March 2024, the ACLU estimated that approximately 2,000 children had still not been reunited with their parents. Human Rights Watch put the number at 1,360 children still unaccounted for, six years after the policy was supposedly ended. Nearly 30% of all separated children. Gone. Disappeared into a system designed, from the start, not to find them.

Human Rights Watch concluded that the government’s conduct met the legal definition of enforced disappearance. They further concluded the forcible separations may have constituted torture: the intentional infliction of severe suffering for an improper purpose by a state agent.

Enforced disappearance. Torture. These are defined terms in international law. And they are being applied to the actions of the United States government against children.

The architects were not held to account. They were promoted. Thomas Homan became Trump’s ‘border czar.’ Stephen Miller became deputy chief of staff. Matthew Whitaker was nominated as ambassador to NATO. The people who stole children from their parents and then lost them were rewarded with more power.

That’s not immigration enforcement. That’s state-sponsored child abuse. And no amount of flag-waving will ever wash it off.

Happiness: a nation eating itself alive

You’d think all of this would make Americans unhappy. You’d be right.

The 2025 World Happiness Report ranked the United States 24th. Its lowest ever. Down from 16th just two years ago. Finland has topped the rankings since 2018, because apparently universal healthcare, functional education, clean air, and not starting wars is good for morale. Who knew.

'America is the only country that went from barbarism to decadence without civilisation in between.' — Oscar Wilde (attributed)

The researchers noted America’s decline was driven by political polarisation and votes against ‘the system.’ Since 2000, deaths of despair have decreased worldwide, with two notable exceptions: the United States and the Republic of Korea.

Your people are miserable, Karen. Your young people are the most miserable in the English-speaking world. And your collective response was to vote for a man who promised to make it all worse. Twice.

Which raises the question of what America does with its miserable citizens. Answer: it locks them up.

Incarceration: land of the free, home of the caged

The United States has the highest incarceration rate of any independent democracy on Earth. 583 per 100,000. One in five of the world’s prisoners is American. Massachusetts, the most progressive state, still locks up more people per capita than Iran. The land of the free has more people in cages than any other free country on the planet. So naturally you lecture others about liberty.

'No one truly knows a nation until one has been inside its jails.' — Nelson Mandela

Xenophobia: a nation built by immigrants that despises immigrants

Here is a country whose founding mythology is literally immigration. Give me your tired, your poor, your huddled masses. It’s inscribed on the Statue of Liberty, for God’s sake. And here is that same country, in 2026, deporting people to prison camps in third countries where they have no ties, no language, no legal protections, and no way home.

'In the first place, we should insist that if the immigrant who comes here in good faith becomes an American and assimilates himself to us, he shall be treated on an exact equality with everyone else.' — Theodore Roosevelt, 1907

The Trump administration has made ICE the highest-funded federal law enforcement agency in American history, backed by nearly US$170 billion in new enforcement spending. That’s more than some countries’ entire military budgets, all directed at rounding up people whose primary crime is wanting a better life in the country that told the world it was the place to come for exactly that.

ICE arrests of Asian immigrants quadrupled under Trump, from around 2,000 in 2024 to more than 7,700 in 2025. Vietnamese immigrants, Karen. The people who built your nail salons, your phở restaurants, your tech workforce. ICE arrested 930 Vietnamese nationals in the first ten months of 2025. Some have been deported to third countries they’ve never set foot in. I live in Vietnam. I know these people. They are not your enemy.

The administration deported over 260 people, mostly Venezuelans, to El Salvador’s Terrorism Confinement Centre. A 60 Minutes investigation found no criminal charges against 179 of them. A Cato Institute investigation determined more than 50 had entered the United States legally. Deported anyway. To a prison. In a country they’d never been to. Without charges. Others have been sent to Eswatini, a tiny African kingdom that reportedly accepted deportees for US$5.1 million. To Sudan, which is in the middle of a civil war. These are not deportations. These are disappearances with paperwork.

ICE agents have been authorised to raid schools, hospitals, and churches. Forty-one people died in ICE custody in 2025, the deadliest year in two decades. A Hmong American man was asked by a national park attendant whether everyone in his car was a US citizen, for no reason other than the colour of their faces.

Trump called immigrants an ‘invasion.’ When asked about using the military against civilians, he said, ‘These aren’t civilians.’ The President of the United States looked at human beings and said they are not civilians. That is dehumanisation as state doctrine.

Might want to update the Statue of Liberty inscription. ‘Give me your tired, your poor, your huddled masses, so ICE can deport them to a prison in Eswatini.’ Pithier than the original, but more honest.

Gun violence: an American speciality

Among comparable wealthy countries: highest firearms per capita, highest homicides, lowest safety ranking. 22,830 homicides in 2023. Every other comparable country looked at the problem and did something about it. Australia did. Britain did. Canada did. New Zealand did. America bought more guns and buried more children and sent more thoughts and prayers and changed absolutely nothing.

'We learned that we can bury children and go back to normal.' — Amanda Gorman, after Uvalde

I know, I know. Second Amendment. Shall not be infringed. The founding fathers definitely intended for an eighteen-year-old to walk into a shop and buy a weapon of war to take to a primary school. Obviously. That’s clearly what they had in mind when they were writing with quill pens about muskets.

The failure compounds. The ignorance, the ill health, the dead mothers, the stolen children, the caged citizens, the buried schoolkids. None of it is accidental. All of it is the output of a system that has stopped working. A system that doesn’t educate its people, doesn’t keep them healthy, doesn’t protect their children, doesn’t support their mothers, cages the desperate, and then wonders why nobody’s happy.

But we’re only halfway through, Karen. So far we’ve only talked about what America is. In Part 2, we talk about what America does to the rest of us. And that, I promise you, is where the table gets flipped.

Part 2: ‘The world is standing up’ — coming next.

I wrote the first edition of Understanding AuDHD because nobody had written it yet and I was tired of waiting. Three editions later, I’m publishing the fourth, and the reason is simple: the science moved, the community moved, and I moved with it.

The fourth edition is a different book. Same title, same stubbornness, entirely new guts. Fifteen chapters, six appendices, and a reference list that would make a doctoral committee nod approvingly before asking why it took so long. Every chapter has been rewritten from the ground up against the latest research, the emerging voices in the neurodivergent community, and, frankly, the things I got wrong or didn’t go far enough on in earlier editions.

Here is what changed and why it matters.

The core argument is sharper. AuDHD is not autism plus ADHD. It is an emergent neurological profile, a compound rather than a mixture, with its own internal logic, its own presentation patterns, and its own clinical needs that neither condition’s literature captures on its own. The DSM still files it as a comorbidity. The DSM is wrong, and the research published since the third edition makes that case more convincingly than I ever could alone.

The lived experience sections are more honest. I was diagnosed with AuDHD at 66 after decades of being told I was bipolar. That story is in here, not for sympathy, but because it is the story of an entire generation of adults whose neurology was misread by confident professionals using confident tools that were confidently inadequate. If you recognised yourself in the earlier editions, this one goes further. If you are newly wondering whether the label fits, this one meets you earlier in the process.

The women and girls chapter is new. It should have existed sooner.

The healthcare chapter will make you angry and then give you something useful to do with the anger. The systems chapter names the institutions. The practical chapters are woven through the whole book rather than bolted on at the end, because that is how AuDHD actually works; you need the strategy while you are inside the experience, not three chapters later.

Listen to the DeepDive

I put the book through a DeepDive podcast conversation so you can hear the argument before you read it. The episode is embedded below. If you are the kind of person who processes better through audio, or if you want to know whether this book is worth your time before committing to 53,000 words, start here.

For paid subscribers

If you are a paid subscriber to Letters from the Quiet Half, you can download Understanding AuDHD (4th edition) for free. No hoops, no funnels, no seven-email nurture sequence. Just the book.

You also have access to a curated library of my other titles, available for download from the paid subscriber area. The library grows as I publish. If you have been a paid subscriber for a while, there will be new things in there since you last looked.

If you are not yet a paid subscriber and you want to be, the button is below. The price of a coffee once a month gets you every book I publish, the full library, and the knowledge that you are directly funding independent, ad-free writing about neurodivergence by someone who actually lives inside the experience.

quiethalf.substack.com/subscribe

For everyone

Whether you are paid or free, the DeepDive episode above is yours. Listen, share it, send it to someone who needs it. The best thing you can do for the neurodivergent community is not buy my book. It is make sure the people around you understand that AuDHD exists, that it is distinct, and that the systems meant to help are still catching up.

The book will find its readers. The conversation needs everyone.

Neil, I’m going to be straight with you because you’ve always said you prefer that. I’d rather type this than try to get it right over a beer, because there’s a lot to say and I want to say it clearly.

I’ve been sending you ‘too busy’ messages for a while now. You deserve the honest reason.

Before you stop reading, or read half of this and tell me I’ve got it wrong or you never said any of it: I have screenshots of our entire Zalo conversation. Every message, every link, every claim you’ve made. Everything I’m about to describe is documented.

The reciprocity problem

Over the past year you’ve sent me a relentless stream of YouTube videos, health advice, political content, and Facebook links, always with the expectation that I watch, listen, and engage. When I share my work—my published articles, my research, the books I’m writing, the things I’m actually building my life and career around—the answer is ‘too busy.’ Every time.

I’ve got a Masters degree in progress. Books in production. A business I’m building from scratch in a foreign country with limited language. 450 academic citations that are still being referenced decades later, and I’m still publishing. I am not sitting on my arse. But somewhere along the line, ‘busy’ became something only you get to be. Your time is valuable. Mine is apparently available on demand.

You want me to watch your YouTube links but you won’t read a five-minute article I wrote. You want me to take your health advice but you dismiss my professional training. You tell me to stay out of politics then send me political videos the next day. The exchange rate has been broken for a long time, mate, and I’ve run out of ways to pretend otherwise.

Your health advice

I need to say this plainly: you don’t know what you’re talking about.

You told me to take Alaxan regularly before bed. Alaxan contains ibuprofen. I have documented calcified plaques in my carotid arteries and atherosclerosis confirmed by Doppler ultrasound. The cardiovascular research is unambiguous: regular NSAID use in someone with active atherosclerosis increases the risk of serious cardiovascular events. The prescribing information says do not use for more than 10 days without a doctor. My cardiologist, Dr Diem, a German-trained specialist who ran two full days of diagnostics on me, would be horrified by that recommendation.

You told me to do static push-ups and called me a lazy prik for not doing them. I have L5 spinal issues, chronic pain, PTSD, and documented cardiovascular disease. Unsupervised isometric exercises can spike blood pressure dangerously in someone with my profile.

You told me to go zero carbohydrates. I have Type 2 diabetes controlled with medication. Eliminating carbohydrates without medical supervision in a medicated diabetic risks hypoglycaemic episodes. Your source was Paul Mason and Thomas DeLauer on YouTube. Mine is an actual endocrinologist.

You ‘officially diagnosed’ me with altitude sickness. Dalat sits at about 1,500 metres. Clinical altitude sickness typically begins above 2,500 metres. My symptoms had identified medical causes: low testosterone, low vitamin D, chronic inflammation, low iron stores, and suspected long COVID neuroinflammation. All confirmed by actual diagnostic equipment, not a YouTube video.

You told me you could ‘reprogram’ me based on a video about Dr Michael Levin’s bioelectricity research. Levin’s work is legitimate developmental biology about bioelectrical signals in tissue regeneration in laboratory settings. It has nothing to do with treating humans. Levin himself would be appalled at the extrapolation.

You pushed Wim Hof ice baths on me so relentlessly that I actually did what you never do: I read the research. All of it. The Radboud endotoxin study, the 2024 PLOS ONE systematic review, the RCTs. And then I wrote you a 3,000-word evidence-based analysis and published it on my website. Addressed to you. With full citations. ‘Wim Hof—for Neil.’ You didn’t read it. Of course you didn’t. Because your mind was already made up. You never wanted evidence. You wanted agreement.

I also sent you my article about what 66 years of undiagnosed neurodivergence and long COVID actually look like from the inside—the chronic pain, the fascial tissue damage from decades of autonomic hypervigilance, the way masking activates the same stress pathways as sustained threat, the HPA axis staying switched on for decades until muscles harden and blood stops flowing properly. Mr Trung, my massage therapist, says I’m the worst client he’s ever had. Everywhere he touches me I’m in agony. That’s not hypochondria. That’s the physiological cost of sixty-six years of performing neurotypicality, compounded by long COVID autonomic dysfunction. I wrote about it in detail, cited the research, explained the mechanisms. I sent it to you. You didn’t read that either.

That article represents more clinical understanding of my own body than everything you’ve sent me in a year combined. And you couldn’t be bothered to read it. But you expect me to watch a 45-minute YouTube video from some bloke flogging supplements.

You said you were ‘10 years ahead’ of me and that I needed to ‘update my knowledge to 2026.’ Neil, your knowledge hasn’t left the 1980s. You get your health information from YouTube influencers and your own self-experimentation, and you’ve extrapolated it to a 67-year-old with AuDHD, PTSD, atherosclerosis, Type 2 diabetes, chronic neuroinflammation, and a late-life neurodivergent diagnosis that fundamentally changes the clinical picture. I have a Masters in Counselling Practice. I have clinical training. I have a medical team managing a complex, multi-system profile. I don’t need a medically-untrained bloke telling me to do push-ups and pop Alaxan.

And when I set that boundary, you told me I ‘wasn’t a good listener.’ That’s not listening, Neil. That’s compliance.

Your ADHD

You’ve told me you have ADHD and that you know a lot about it. I’ll be honest with you: I don’t think you know what it’s actually doing to you.

I recently sent you my article ‘The research nobody’s talking about,’ about AuDHD and allostatic load—the cumulative biological cost of decades of unmanaged neurodivergence. The research in that article is from the last two years. This is breaking science, Neil. A 2025 UK mortality study of 30,000 adults with ADHD found life expectancy reductions of up to 9 years in men. A 2023 Columbia study found that chronic stress hormone exposure increases cellular energy expenditure by 60% and accelerates biological ageing at the DNA level. The emerging MCAS-neurodivergence connection is showing that what looks like emotional dysregulation may in part be an inflammatory process treatable with antihistamines. None of this existed in the literature five years ago.

I asked you to forward it to your anaesthesiologist sister. You probably thought I was being a hypochondriac, catastrophising about my own health as usual.

I wasn’t.

I was worried about you.

Your sister is retired. She may not be across this research because it’s so new. That’s why I wanted her to see it—not as general interest, but because of what I’ve been observing in you.

Neil, I’m a trained counselling psychologist. I notice things. Over the last few months I’ve noticed changes in your cognitive patterns, your conversational coherence, your flexibility of thinking. The rigidity, the repetitive loops, the inability to update information, the increasing retreat into a closed media ecosystem—these aren’t just stubbornness. From where I’m sitting, they warrant a proper assessment. You’re in your 70s, you have self-reported ADHD, and you’re showing a pattern that a neuropsychologist should look at. That’s not an insult. It’s a clinical observation from someone who cares enough to say it when nobody else will.

I wanted your sister to contact me so we could have that conversation. Not so she could read about my health problems. About yours.

Your politics

I find your politics morally repugnant. I’ve held back on saying this for a long time.

When Charlie Kirk was killed, you seized on the trans angle before any evidence was in. You said trans people were to blame. You said they were sick and should be killed. You sent me a video of Laura Ingraham or whoever it was from Fox News saying how disgusting these people were. When I asked you the obvious question—why would a hated minority do something guaranteed to make their lives even worse—you refused to answer.

The investigation was a mess of competing narratives. Groypers, leftists, gamers, foreign bots, everyone trying to claim the shooter for the other side. What actually emerged from the charging documents was a young man from a conservative family who had shifted left, had a transgender partner, and appears to have acted alone because he believed Kirk was spreading hate. Not ‘the trans community.’ One person. With complicated, contradictory motivations that don’t fit neatly into anyone’s talking points.

But you didn’t wait for any of that. You had your conclusion before the body was cold. And when the picture became clearer and the trans community conspiracy theory fell apart, you refused to update. Not a word, even when I asked you on a number of different occasions what your views were in light of the actual evidence. Because the correction didn’t fit your narrative. You didn’t care what actually happened. You cared what confirmed what you already believed.

That’s the pattern, Neil. That’s the whole problem in one incident.

You cite Joe Rogan, Russell Brand, Andrew Tate, and others as people who ‘know what’s really happening.’ These are reactionary media entertainers. They are not journalists. They are not researchers. They are not thinkers. They are content creators who have worked out that outrage is profitable.

You watch YouTube videos and treat them as fact. You don’t read. You don’t interrogate sources. You don’t update your position when new information arrives. My mother had a phrase for it: ‘My mind’s made up, stop confusing me with facts.’

I’m not asking you to agree with me politically. I’m telling you that your wilful refusal to engage critically with information, combined with your expectation that I engage with everything you send, has made these conversations exhausting.

What I’m doing

I’m stepping back from the daily chat. I haven’t got the bandwidth for a friendship that only flows one direction, that expects me to be your audience while dismissing everything I do, that gives me medical advice that could genuinely harm me, and that asks me to sit quietly through political views I find unconscionable.

I’m not disappearing. If you’re ever in a bind or need something real, I’m here. But I’m done being the audience.

Cheers,
Lee

References

Hopkins, L. (2025, December 6). Wim Hof. MindBlown Psychology. https://mindblownpsychology.com/wim-hof-for-neil/

Hopkins, L. (2026, February 28). Your body has been writing you letters for years. You just couldn’t read them. MindBlown Psychology. https://mindblownpsychology.com/late-diagnosed-neurodivergent-long-covid-body-pain/

Hopkins, L. (2026, March 17). The research nobody’s talking about: AuDHD and allostatic load. Letters from the Quiet Half. https://www.quiethalf.com/p/the-research-nobodys-talking-about

Hopkins, L. (2026, March 22). When the body is part of the emotional problem. Letters from the Quiet Half. https://www.quiethalf.com/p/when-the-body-is-part-of-the-emotional

Then Vera Hart landed in my inbox and pulled the whole thing into focus.

Dr Vera Hart is a trauma psychiatrist and neurologist. Her essay ‘When Safety Becomes a Cage’ is one of the most quietly devastating things I’ve read this year. Not because it’s dramatic. Because it’s precise. She writes from inside her own nervous system’s history with the clinical language she earned afterwards, and the combination is something most academic writing never achieves.

I want to share what she’s saying, because it connects to a much larger shift happening in the neurodivergence conversation right now. And because I suspect more than a few of you will recognise yourselves in it.

The trade

Hart’s essay begins with a sentence she carried for twenty-two years like a private verdict. She’d internalised Benjamin Franklin’s line about liberty and safety, not as political philosophy, but as personal law: if you chose safety over liberty, you deserve whatever follows. At nineteen she left the person she loved and married someone her family approved of, because her father told her a young woman moving to another country should be married. The body knew immediately. She came home crying during the courtship. But she told herself the adults were right, that love at nineteen wasn’t real love, that stability would prove the decision correct.

What followed was not stability. It was a marriage in which her fear was consistently weighed against someone else’s convenience, and convenience won. A winter night stranded at an abandoned train station. A husband who told her to get in a car with men whose eyes she didn’t trust, or stay in the cold. A year of sexual assault by a supervisor with power over her career, endured in silence beside a husband who slept through it all. An overdose that changed nothing in the room around her. Nobody came. Nobody asked.

Hart doesn’t tell this story for sympathy. She tells it to map a pattern. And the pattern is this: when distress is consistently met with non-response, the nervous system stops sending distress signals. It builds something else instead.

The box of achievement

She calls it the box of achievement. I think most neurodivergent people will feel a jolt of recognition.

The box is what happens when you can’t trust that your feelings will be held by anyone, so you make your competence undeniable instead. You freeze the part of yourself that wants to be witnessed and develop the part that produces. You cross oceans, pass exams in languages that aren’t yours, build careers, collect credentials. From the outside it looks admirable. From the inside it’s airless. The body never truly comes down because rest requires believing the world won’t punish you for stopping.

Hart names this not as pathology but as adaptation. The brain is trying to control what it can control. Under chronic relational threat, the stress system stays permanently on. Achievement briefly quiets the fear of being disposable. The box gives the alarm a socially acceptable outlet.

I wonder how many of us built our boxes before we had language for what we were building them against.

The body’s refusal

The essay’s turning point is literal and devastating. After years of institutional whistleblowing, after choosing to protect patients over her own comfort, Hart’s visual field fractured under stress. Half her world went hazy. She drove herself to hospital. Her husband didn’t offer to drive her there or pick her up. She drove herself home after midnight with her vision still not fully restored.

The irony she names is almost too precise: the year she lost part of her literal vision was the year she finally began to see her life clearly.

This is where Hart’s clinical training and lived experience converge into something genuinely useful. She’s not romanticising the symptom. She’s saying the body eventually refuses to cooperate with a story that’s been damaging you for years. The symptom interrupts the bargain. It writes in capital letters what the mouth has been refusing to say.

What this has to do with neurodivergence

Here’s why I’m sharing this on a psychology Substack rather than just bookmarking it privately.

Hart’s essay maps a pattern that runs through almost every late-diagnosed neurodivergent life I’ve encountered, including my own. The constrained choice at a young age. The compliance mistaken for safety. The decades of masking that look like competence. The box of achievement built from fear rather than freedom. The body’s eventual insistence on truth, whether that arrives as burnout, breakdown, or some other refusal the mind wasn’t planning.

What Hart calls ‘the box of achievement’, the neurodivergence conversation is starting to call ‘masking collapse’ or ‘autistic burnout’. What she describes as a nervous system learning that distress won’t bring help, we’re beginning to understand as the mechanism behind decades of undiagnosed AuDHD in adults.

The language differs. The pattern doesn’t.

This week’s Substack landscape has been converging on exactly this territory:

Bridgette Hamstead is challenging the entire evidence base for neurodivergent interventions, arguing that the research has been measuring institutional goals rather than neurodivergent wellbeing. She’s also mapping the perimenopause-to-late-diagnosis pipeline, where oestrogen decline removes the hormonal scaffolding that made decades of masking possible.

Lindsey Mack is asking whether undiagnosed Binocular Vision Dysfunction, invisible to standard eye tests, has been adding a sensory processing burden to neurodivergent nervous systems since childhood, changing the entire arithmetic of burnout.

The writer behind Inside Attunement is drawing a distinction between dysregulation and depletion that reframes the whole therapeutic conversation.

Sher Griffin is arguing that neurodivergent people who transparently use AI as an access tool are being penalised for honesty while those who conceal it are rewarded.

And now Hart, from trauma psychiatry, is saying what all of these writers are circling from their own angles: the nervous system is not broken. It’s been carrying a load nobody thought to measure, in conditions nobody thought to question, for longer than anybody thought to ask about.

The question that matters

Hart reframes Franklin’s quote at the end of her essay in a way I haven’t been able to stop thinking about. She’s not interested in liberty as political abstraction. She’s interested in liberty as the right to trust your own perception when something is wrong, and to act on that perception without treating yourself as a criminal for noticing.

For those of us who spent decades masking, achieving, enduring, and calling it maturity, that reframe lands hard. Could it be that what we called discipline was actually captivity dressed in respectable clothes? Could it be that the box we built to survive was the thing slowly destroying us?

I wonder.

Hart’s essay deserves to be read in full. It’s long. It earns every word.

You can find it here: When Safety Becomes a Cage by Dr Vera Hart, MD, PhD. Follow her work on Substack at Vera Hart MD PhD, and on Instagram at @verahartmdphd.

-----

If you’re interested in the other writers I mentioned: Bridgette Hamstead writes at bridgettehamstead.substack.com, Lindsey Mack at lindseymackereth.substack.com, Sher Griffin (The Cognitive Ecologist) at shergriffin.substack.com, and Inside Attunement at wonder2wisdom.substack.com. All worth your time.

There is an emerging connection between neurodivergent emotional experience and the immune system that the clinical mainstream has barely begun to acknowledge.

Mast cell activation syndrome—MCAS—is a condition in which mast cells, part of the immune system, release excessive inflammatory mediators inappropriately. It produces a bewildering constellation of symptoms including fatigue, brain fog, headaches, gastrointestinal distress, and skin reactions. It is common, rarely diagnosed, and increasingly associated with both neurodivergent conditions and connective tissue disorders like hypermobile Ehlers-Danlos syndrome (hEDS).

A 2025 study by Weinstock and colleagues in Brain, Behavior, and Immunity—Health surveyed 553 MCAS patients and 558 controls and found dramatically elevated rates of neuropsychiatric conditions in the MCAS group (Weinstock et al., 2025). Female MCAS patients showed significantly higher rates of ADHD, OCD, bipolar disorder, PTSD, suicidality, anxiety, and depression compared to controls. The odds ratios were striking—and critically, many patients reported symptom improvement after mast-cell-directed treatment, including antihistamines.

Read that again. Psychiatric symptoms improving after antihistamine treatment.

The implications are uncomfortable for a field that has been treating these symptoms as primarily psychological. If immune dysregulation is producing or amplifying the emotional intensity, the sensory sensitivity, the fatigue, and the cognitive fog that AuDHD people experience, then treating the emotional symptoms without investigating the immune system is like mopping the floor while the tap is running. The emerging MCAS-hEDS-neurodivergence connection suggests that for some AuDHD people, what looks like emotional dysregulation may be, in part, an inflammatory process. This doesn’t mean everyone with AuDHD has MCAS. It means the question should be asked, and it almost never is.

From my forthcoming book, Understanding AuDHD (4th edition).

——-

Addendum:

I’m going to resist the temptation to end this chapter with advice about how to manage AuDHD emotions better, because the whole argument of the chapter has been that managing the emotions is the wrong starting point. The question is not ‘how do I regulate my feelings?’ The question is: are the circumstances producing those feelings as reasonable as my feelings say they are? And if so, which needs changing—my emotional response, or the circumstances?

The next chapter moves from the internal world to the external one: education and academic success, where the AuDHD brain meets institutional structures that were designed by people who have never, as far as I can tell, met a neurodivergent child. The emotional regulation challenges described here don’t stay at home. They walk through the school gates every morning, and what happens next is the subject of Chapter 7.

Paid subscribers will be able to read the whole book once it is published. quiethalf.substack.com/subscribe

Forget Mossad (well-meaning work experience students at best).

Ignore the CIA (sugar-rushed kindergarteners playing dress-up).

Don’t even think about MI6 (earnest middle managers with PowerPoint presentations) or the FSB (your uncle’s mate who ‘knows a bloke’).

The world’s most formidable intelligence network?

Vietnamese women (aka ‘VWN’, a named whispered fearfully in quiet corridors around the globe).

The Vietnamese government’s surveillance apparatus is impressive, sure. But the aunties, sisters, mothers and wives have built something that makes the NSA look like two tin cans and a piece of string.

You bought new shoes yesterday? Your wife’s cousin in Hanoi already knows. You had coffee with a female colleague? Your mother-in-law in Huế received a full briefing including the colour of her dress before you’d finished your cà phê sữa đá. You told your wife you were working late? Seventeen women across three provinces are currently cross-referencing your motorbike’s location with CCTV footage from outside the bia hơi.

I once bought Huong flowers on a Tuesday. A Tuesday. Within four hours, her sister called from Saigon wanting to know what I’d done wrong. I hadn’t done anything wrong. The flowers were just… flowers.

Turns out random Tuesday flowers are a ‘guilty conscience indicator’ flagged in the network’s database somewhere around 1987.

The system is flawless. The reach is global. The response time would make CIA operatives weep into their classified documents.

And the terrifying part? There’s no central command. It’s all organic. Distributed intelligence that would make blockchain developers feel inadequate.

Gentlemen living in Vietnam: you are not unobserved. You have never been unobserved. You will never be unobserved.

Make peace with it.

(Currently posting this while Huong is at the market, which means I have approximately eleven minutes before she knows what I’ve written, despite her not being a member of this private tribe)

In early March 2026, I sat in a consulting room at FV Hospital in Ho Chi Minh City and watched a doctor flip through test results that said, essentially, nothing to see here.

I’d travelled six hours from Đà Lạt to get those results. I’d spent months preparing the case. I’d written to the hospital citing van der Kolk, McEwen, and Raymaker in my referral letter, because I know from decades of clinical practice that if you don’t frame the question precisely, the system will answer a different one.

Here is what my body was doing while my quarterly blood panels kept returning normal.

• A chronic daily headache, present almost every morning for ten months, with characteristics resembling a hangover despite no alcohol consumption. Standard analgesics, ibuprofen, paracetamol, couldn’t touch it. Previous CT and MRI scans at the same hospital had found no structural cause.

• Severe, persistent fatigue characterised by approximately one hour of normal energy followed by prolonged exhaustion lasting several hours. Sleep provided no restoration. Constant yawning throughout the day.

• Widespread body pain with myofascial trigger points so severe that even light pressure caused extreme pain; my massage therapist, who has worked on hundreds of bodies, described mine as the worst case he had encountered.

• Brain fog.

• Reduced concentration.

• Balance difficulties.

• Previously documented episodes of unexplained bruising.

• Blood pressure recorded at 170/89 during an emergency visit to my local hospital in Đà Lạt, where the doctor glanced at my Vietnamese-language medical summary, laughed, and walked away.

I want you to sit with that inventory for a moment. Because it is not unusual. If you’re reading this and you have AuDHD, I suspect at least three of those symptoms are familiar. The headache that doesn’t respond. The fatigue that sleep doesn’t fix. The pain nobody can explain. The blood tests that keep telling you you’re fine.

You’re not imagining it. Your tests are answering the wrong question.

In my referral letter to FV, I framed the clinical reasoning around allostatic load. I explained that sixty-six years of undiagnosed neurodivergence had required chronic masking and autonomic hypervigilance, resulting in sustained HPA axis activation and cumulative physiological damage. I requested cortisol rhythm testing, extended thyroid panels, inflammatory markers, iron studies, coagulation profiles, and autonomic assessment. I also flagged a drug interaction between my clopidogrel and escitalopram that increases bleeding risk and may have been contributing to the unexplained bruising nobody had connected to the medication combination. I’d been on both for months. Nobody had flagged it.

The comprehensive results confirmed no acute pathology. A medication review and adjustment resolved the headache that had tormented me for ten months. The new regime is working. I feel healthier than I have in decades.

But something about that outcome kept me thinking. The headache wasn’t structural. It wasn’t a tumour. It wasn’t vascular. It was, as best anyone can determine, my nervous system’s response to cumulative, unrelieved strain. The body pain wasn’t orthopaedic. It was fascial: the physical residue of decades of bracing, hypervigilance, and chronic stress held in tissue that nobody had thought to examine through a neurodivergent lens. The fatigue wasn’t laziness. It was a system running vastly above normal energy expenditure without anyone accounting for the metabolic cost of performing neurotypicality for six decades.

My quarterly blood panels were normal because quarterly blood panels don’t ask the right questions about a neurodivergent body.

I went looking for the right questions. What I found kept me awake at 4am.

The chapter this is building toward

I’m currently writing the fourth edition of Understanding AuDHD. Chapter 3 is called ‘The invisible toll: AuDHD, allostatic load, and what it costs the body.’ Its argument is this: decades of sensory overload, masking exhaustion, executive function taxation, chronic social stress, and financial precarity produce measurable biological damage that accelerates aging at the cellular level. AuDHD is not just a neurodevelopmental profile. It may be an unrecognised risk factor for premature physical decline and, emerging evidence suggests, for neurodegeneration.

I want to share the research backbone of this chapter with you, because I think you deserve to see it. Not filtered through a clinician’s reassuring summary. Not softened for comfort. Just the studies, what they found, and the picture they form when placed side by side.

The life expectancy evidence

A 2025 UK study used mortality data from 30,000 adults with diagnosed ADHD and found life expectancy reductions of 4.5 to 9 years for men and 6.5 to 11 years for women (O’Nions et al., 2025). A landmark Swedish study of 27,000 autistic individuals found premature mortality was markedly increased across a multitude of medical conditions, with the strongest effects among women and those with intellectual disabilities (Hirvikoski et al., 2016). The AuDHD-specific life expectancy gap, the one that would tell me what having both conditions together actually costs, has never been studied.

Nobody has done that research.

I wonder why.

The allostatic load evidence

This is where my own experience starts to make biological sense.

A 2023 Columbia University study found that chronic stress hormone exposure increases cellular energy expenditure by approximately 60% and accelerates biological aging as measured by DNA methylation clocks and telomere shortening (Bobba-Alves et al., 2023). Sixty percent. That’s the metabolic surcharge of a body running on chronic stress. Nobody gets a bill for it. The body pays anyway.

A 2026 study confirmed that financial strain and everyday discrimination are associated with accelerated epigenetic aging across the lifespan (Ciciurkaite et al., 2026). A 2025 neuroimaging study found that allostatic load is linked to structural brain changes in regions sensitive to aging (Palix et al., 2025).

Now map those findings onto a life spent navigating neurotypical systems with an AuDHD brain. The masking. The sensory management. The executive function taxation. The social performance. The financial instability that follows executive function differences through every career decision. Every one of those experiences generates exactly the kind of chronic, unrelieved stress that the allostatic load research identifies as biologically damaging. It is the mechanism behind symptoms like mine: the headache that imaging can’t find, the pain that blood tests can’t explain, the fatigue that sleep can’t repair.

A real-world study of 2.3 million insurance members found that co-occurring autism and ADHD synergistically increases the odds of additional comorbid conditions beyond what either condition produces alone (Thomas Craig et al., 2025). Not additively. Synergistically. The combination is worse than the sum of its parts.

And Raymaker’s foundational 2020 study defined what the autistic community had been describing for years: autistic burnout as a distinct syndrome resulting from chronic mismatch between expectations and capacity, characterised by pervasive exhaustion, loss of previously reliable skills, and reduced sensory tolerance (Raymaker et al., 2020). That’s not a metaphor. That’s a clinical description of what happens when a system exceeds its load-bearing capacity.

The dementia evidence

The allostatic load research was sobering. What follows is worse.

A 2023 Israeli cohort study following 109,218 people over 17 years found that adults with ADHD had nearly three times the risk of developing dementia, with a hazard ratio of 2.77. Notably, psychostimulant medication appeared to reduce the association (Levine et al., 2023). A Swedish registry study of 3.59 million individuals found an ADHD dementia hazard ratio of 2.92, attenuating to 1.62 after adjusting for psychiatric comorbidities but remaining significant (Dobrosavljevic et al., 2022).

A 2024 Argentine study, small but mechanistically striking, found a hazard ratio of 54.54 for Lewy body dementia specifically in the ADHD cohort, with 95% of ADHD participants who developed dementia developing the Lewy body variant (Golimstok et al., 2024). That confidence interval is wide enough to drive a truck through, but the dopaminergic pathway makes biological sense.

The autism evidence is equally sobering. A 2021 Medicaid study found autistic adults were approximately 2.6 times more likely to be diagnosed with early-onset dementia than the general population (Vivanti et al., 2021). A 2025 follow-up using linked Medicare and Medicaid data from over 114,000 autistic adults found that among those over 64, 35.1% carried a dementia diagnosis, versus approximately 10 to 11% in the general population (Vivanti et al., 2025). A Swedish family study found elevated dementia risk not only in autistic people but in their biological relatives, suggesting shared genetic pathways (Zhang et al., 2025).

A 2025 Geneva University study identified the first plausible neurobiological mechanism: altered iron distribution in the brains of adults with ADHD, correlated with elevated neurofilament light chain protein, a marker of neuronal damage seen in early-stage dementias (Berberat et al., 2025).

And the 2023 International Summit on Intellectual Disability and Dementia produced a consensus document identifying overlapping symptoms of autism and dementia as a critical barrier to accurate diagnosis (Janicki et al., 2025). Executive function decline, increased rigidity, memory difficulties, anxiety. These are features of both conditions. Clinicians may interpret early dementia in an autistic person as ‘just their autism.’

I wonder how many people that’s already happened to.

What connects my body to these numbers

I want to be careful here. I’m not arguing that AuDHD causes dementia. The research doesn’t support that claim and neither do I.

What I’m arguing is that the conditions of life, the decades of environmental mismatch, the masking, the chronic stress, the inadequate healthcare, the financial precarity, produce allostatic load that may accelerate biological aging and elevate neurodegenerative risk. And that this load produces real physical symptoms along the way. Symptoms like mine. Symptoms that routine blood panels don’t capture because they were designed to measure acute pathology, not the slow accumulation of physiological debt from sixty-six years of running the wrong operating system in the wrong environment.

The good news, and there is good news, is that allostatic load is modifiable.

Environmental change reduces allostatic load. I moved to Vietnam and my depression vanished within months. Same brain. Different environment. Same pension. Different cost of living.

Appropriate medical investigation reduces allostatic load. Once the right questions were asked at FV Hospital, the right medications were found, and a headache that had tortured me for ten months resolved.

Accurate diagnosis reduces allostatic load. Understanding at sixty-six that I was neurodivergent, not broken, reframed decades of suffering from personal failure to systemic mismatch.

Stopping the performance of neurotypicality reduces allostatic load. Learning to unmask, imperfectly, gradually, is the most important thing I’ve done for my body since arriving in this country.

The research suggests that the dementia risk associated with AuDHD may be driven substantially by these modifiable factors. That is not a reason for complacency. It is a reason for urgency. Because the generation of late-diagnosed AuDHD adults, the ones who spent their entire working lives undiagnosed and unsupported, are entering the age range where the consequences become visible. Their bodies have been paying the allostatic surcharge for decades. The question is not whether damage has occurred. The question is how much can be recovered, and how quickly the systems that produced the damage can be changed.

I wonder what it costs a body to spend fifty years pretending to be someone else.

I wonder whether the medical establishment will take this seriously before the answers are written in dementia statistics.

I wonder whether destigmatising neurodivergence and taking its physical health consequences seriously might turn out to be the same project, not competing ones.

Why I’m showing you this

Chapter 3 of Understanding AuDHD, fourth edition, will explore all of this with the appropriate caveats, the honest uncertainties, and the gentle contrarian position that asking these questions is itself an act of care, not an act of alarm.

But I wanted you to see the studies now. Because you deserve to make your own assessment of what the evidence is saying, without anyone, including me, telling you what to conclude.

The references are below. Every one has been independently verified against the original publication.

Lee
Đà Lạt, Vietnam, 17 March 2026

References

Berberat, J., Kagerer, S. M., Späni, C., Hua, J., Bavato, F., Gruber, P., van Zijl, P. C., Perroud, N., Li, X., Stämpfli, P., Seifritz, E., Lövblad, K.-O., Quednow, B. B., & Unschuld, P. G. (2025). Brain iron load and neuroaxonal vulnerability in adult attention-deficit hyperactivity disorder. Psychiatry and Clinical Neurosciences, 79(5), 282–289. https://doi.org/10.1111/pcn.13806

Bobba-Alves, N., Sturm, G., Lin, J., Ware, S. A., Karan, K. R., Monzel, A. S., Bris, C., Procaccio, V., Lenaers, G., Higgins-Chen, A., Levine, M., Horvath, S., Santhanam, B. S., Kaufman, B. A., Hirano, M., Epel, E., & Picard, M. (2023). Cellular allostatic load is linked to increased energy expenditure and accelerated biological aging. Psychoneuroendocrinology, 155, 106322. https://doi.org/10.1016/j.psyneuen.2023.106322

Ciciurkaite, G., Peng, S., Mitchell, C., Perry, B. L., & Lee, B. (2026). Psychosocial stressors, accelerated biological aging, and multiple morbidities: Evidence from an age-diverse sample. PLOS ONE. https://doi.org/10.1371/journal.pone.0343987

Dobrosavljevic, M., Solares, C., Cortese, S., Andershed, H., & Larsson, H. (2022). Attention-deficit/hyperactivity disorder as a risk factor for dementia and mild cognitive impairment: A population-based register study. European Psychiatry, 65(1), e2. https://doi.org/10.1192/j.eurpsy.2021.2261

Golimstok, Á., García Basalo, M. J., Majul, M., Buonanotte, W., Rojas, M., Campos Fein, M., & Eichel, R. (2024). Adult attention deficit-hyperactivity disorder is associated with Lewy body disease and cognitive impairment: A prospective cohort study with 15-year follow-up. The American Journal of Geriatric Psychiatry, 32(10), 1220–1230. https://doi.org/10.1016/j.jagp.2024.04.016

Hirvikoski, T., Mittendorfer-Rutz, E., Boman, M., Larsson, H., Lichtenstein, P., & Bölte, S. (2016). Premature mortality in autism spectrum disorder. The British Journal of Psychiatry, 208(3), 232–238. https://doi.org/10.1192/bjp.bp.114.160192

Janicki, M. P., McCallion, P., Jokinen, N., Service, K. P., Mughal, D. T., Watchman, K., Gomiero, T., Larsen, F. K., & Keller, S. M. (2025). Autism, diagnostics, and dementia: A consensus report from the 2nd International Summit on Intellectual Disabilities and Dementia. International Journal of Geriatric Psychiatry, 40(6), e70110. https://doi.org/10.1002/gps.70110

Levine, S. Z., Rotstein, A., Kodesh, A., Sandin, S., Lee, B. K., Weinstein, G., Schnaider Beeri, M., & Reichenberg, A. (2023). Adult attention-deficit/hyperactivity disorder and the risk of dementia. JAMA Network Open, 6(10), e2338088. https://doi.org/10.1001/jamanetworkopen.2023.38088

O’Nions, E., El Baou, C., John, A., Lewer, D., Mandy, W., McKechnie, D. G. J., Petersen, I., & Stott, J. (2025). Life expectancy and years of life lost for adults with diagnosed ADHD in the UK: Matched cohort study. The British Journal of Psychiatry, 226(5), 261–268. https://doi.org/10.1192/bjp.2024.199

Palix, C., Chauveau, L., Felisatti, F., Chocat, A., Coulbault, L., Hébert, O., Mézenge, F., Landeau, B., Haudry, S., Fauvel, S., Collette, F., Klimecki, O., Marchant, N. L., De La Sayette, V., Vivien, D., Chételat, G., Poisnel, G., & the Medit-Ageing Research Group. (2025). Allostatic load, a measure of cumulative physiological stress, impairs brain structure but not β-accumulation in older adults: An exploratory study. Frontiers in Aging Neuroscience, 17, 1508677. https://doi.org/10.3389/fnagi.2025.1508677

Raymaker, D. M., Teo, A. R., Steckler, N. A., Lentz, B., Scharer, M., Delos Santos, A., Kapp, S. K., Hunter, M., Joyce, A., & Nicolaidis, C. (2020). “Having all of your internal resources exhausted beyond measure and being left with no clean-up crew”: Defining autistic burnout. Autism in Adulthood, 2(2), 132–143. https://doi.org/10.1089/aut.2019.0079

Thomas Craig, K. J., Khan, R., Hariharaputran, S., Gao, J., & Haber, M. (2025). Real-world evaluation of prevalence, cohort characteristics, and healthcare utilization and expenditures among adults and children with autism spectrum disorder, attention-deficit hyperactivity disorder, or both. BMC Health Services Research, 25, 1048. https://doi.org/10.1186/s12913-025-13296-2

Vivanti, G., Tao, S., Lyall, K., Robins, D. L., & Shea, L. L. (2021). The prevalence and incidence of early-onset dementia among adults with autism spectrum disorder. Autism Research, 14(10), 2189–2199. https://doi.org/10.1002/aur.2590

Vivanti, G., Lee, W. L., Ventimiglia, J., Tao, S., Lyall, K., & Shea, L. L. (2025). Prevalence of dementia among US adults with autism spectrum disorder. JAMA Network Open, 8(1), e2453691. https://doi.org/10.1001/jamanetworkopen.2024.53691

Zhang, L., et al. (2025). Association between autism and dementia across generations: Evidence from a family study of the Swedish population. Molecular Psychiatry. https://doi.org/10.1038/s41380-025-03045-6

Once upon a time, I believed in facts the way other people believe in gravity. Utterly. Unconsciously. With the profound confidence of someone who has never been weightless.

Then gravity stopped working.

Not all at once. Not in some dramatic collapse that would make for a satisfying origin story. More like discovering, one Tuesday morning over coffee, that the floor you’d been standing on for forty years was actually a very convincing hologram projected by people who’d figured out that confidence and competence look identical from a distance.

Which, if you think about it, is considerably worse.

The book nobody asked me to write

The Collapse of Knowledge started as a question I couldn’t stop asking myself: what happens when someone trained to spot bullshit discovers their bullshit detector has been turned into a weapon against them?

I spent decades as a scientist. Proper peer-review-and-replication scientist. I could identify pseudoscience from three suburbs away. Unfalsifiable claims made my left eye twitch. I had Karl Popper’s philosophy of science tattooed on my professional identity, and I wore it like armour.

That armour, it turns out, had holes I couldn’t see. Holes that were put there by people who understood exactly how it was built.

This book is about those holes.

What you’ll find inside

Twelve chapters. No easy answers. A fair amount of self-mockery, because if you can’t laugh at the absurdity of a trained epistemologist who got fooled by YouTube’s algorithm, you probably shouldn’t be writing about knowledge at all.

I trace the whole arc, from my upbringing in scientific certainty through the seduction of being right (it’s basically a dopamine addiction, and I was hooked), through the first cracks that appeared when five different psychiatric medications failed to treat a condition I didn’t actually have.

There’s a chapter about moving to Vietnam and discovering that Eastern approaches to knowledge had been comfortable with paradox for centuries while Western thinking was still demanding that contradictions pick a side.

There’s a chapter about YouTube becoming a counterfeit university, where confidence has been successfully rebranded as competence and the algorithm has appointed itself Dean of Everything.

There’s one about micro-truths and macro-lies, which is the sophisticated new technique of building cathedrals of misinformation using nothing but verified facts. Every brick is real. The building is a hallucination.

And yes, there’s a chapter about AI evangelism. Because I did it again. Fell for the same pattern I fell for with social media. Stood on the same stage, different technology, same breathless excitement, same failure to check for exit signs.

I should have known better. Chapter ten is called ‘How I became John the Baptist again, and yes, I should have known better.’

Why this book exists

Because I keep meeting intelligent people who feel like they’re losing their minds. Not because they’ve become stupid, but because the systems we built to separate truth from nonsense have been compromised in ways that make individual verification nearly impossible.

I started noticing a new kind of anxiety in my psychology practice. Epistemic anxiety, I call it. The distress that comes from not knowing what’s true anymore, especially about things that directly affect your wellbeing.

Traditional anxiety responds to traditional interventions. Epistemic anxiety is different. It’s not irrational fear. It’s the perfectly rational recognition that nobody, including the experts, can reliably tell you what’s real anymore.

That’s not a mental health problem. That’s an accurate reading of the situation.

And if your accurate reading of the situation looks like a mental health problem, then perhaps the situation, not your reading of it, is what needs treatment.

What this book isn’t

It isn’t a guide to thinking more clearly. There are enough of those, and most of them are written by people who haven’t noticed that their own critical thinking frameworks are part of the problem.

It isn’t a manifesto against science. Science remains the best method we have for understanding reality. The issue isn’t the method. The issue is what happens when the method gets captured by incentive structures that reward continued confusion over actual understanding.

And it isn’t optimistic in the usual sense. I don’t promise you’ll emerge from these pages with better tools for navigating the information landscape. I promise you’ll feel less alone in finding it bewildering.

Sometimes that’s enough.

The ending might surprise you

The final chapter asks a question I spent eleven chapters avoiding: what if the solution isn’t better frameworks or more sophisticated analysis? What if it’s just stopping for a bit?

Going for a walk. Taking photographs. Having coffee without checking whether the health claims about coffee have updated since breakfast.

The universe spent billions of years evolving consciousness. The least we can do is use it thoughtfully rather than turning it into an optimisation project.

You’re allowed to just be here. The collapse of knowledge isn’t something you solve. It’s something you survive. And survival doesn’t require winning every epistemological battle.

It requires knowing which battles to fight, which to walk away from, and which to acknowledge you’re too tired for today.

Read it on your device

The Collapse of Knowledge is available to read on your phone, tablet, or computer as part of a paid subscription to Quiet Half. No extra purchase needed. Just subscribe, and the book is yours to read whenever you like, wherever you like.

If you’re already a paid subscriber, thank you. The book is waiting for you.

If you’re not yet, well. I wonder if a book about the collapse of everything we thought we knew might be worth the price of a decent coffee.

Subscribe to Quiet Half

The book argues that focusing on villains can actually obscure the systems that sustain harm. When outrage concentrates on monsters, the economic and institutional structures that produce exploitation often remain untouched.

Paid subscribers can read the book for free.
quiethalf.substack.com/subscribe