Between having my brain zapped 25 times, the induced seizures, and maybe a little sprinkle of trauma...

I literally don’t remember 🤷

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. The comments are open too; I read and answer everything.

Share

Leave a comment

I write about mental health, recovery, depression and ADHD, sometimes through film, sometimes through logic, and occasionally through whatever strange connection my brain has made that week. New essays most weeks.

Free subscribers: come read.

Paid subscribers: get everything I publish here, including paid essays, bonus pieces, practical resources, downloadable tools and the full paid archive.

Patrons: get everything paid subscribers do, and help me keep making it.

If you’d like to support Lucid Intervals, a paid subscription is the best way. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

I spend an unreasonable amount of time reading mental-health research, government announcements and studies with titles designed to make sure nobody outside a university ever opens them.

Some of that is for school. Some of it is, apparently, for fun.

So I’m going to try to make myself useful.

Each week, The Lucid Briefing will pull out three things:

Something important.

Something interesting.

Something I’m suspicious of.

I’m obviously not covering everything. Nobody has time for that, and I’m not an actual journalist, so I get to pick and choose.

Just three things I found interesting enough, important enough or suspicious enough to spend far too long reading about.

This week: WA has opened a different front door for people needing urgent mental-health care, researchers asked what we actually mean when we say a depression treatment “worked”, and nine people with anxiety got considerably better after taking a CBD oil.

Let’s start in Fremantle.

Something important

Western Australia has opened a different front door for a mental-health crisis

First up, something local and very relevant to me.

I’ve been through the WA mental-health system, and I’m now preparing to work in it, so I read this wondering two things: would this have helped me, and could I end up working there?

WA has just opened its first dedicated 24-hour Mental Health Crisis Centre at Fremantle Hospital.

The idea is fairly simple. If you’re aged 18 to 65 and live within the Fiona Stanley Fremantle Hospital Group catchment, you can contact the centre directly by phone or show up in person. Family and friends can refer someone too, as can GPs, hospitals and other services.

A mental-health clinician assesses what’s happening and works out what comes next. The centre can connect people with community services, follow-up care, their GP or another service better suited to what they need.

The team includes psychiatry, nursing, allied health, peer workers and Aboriginal mental-health liaison staff. There are also pathways involving police and ambulance.

Basically, WA has built a mental-health-specific entry point into acute care rather than sending everyone through a general emergency department first.

There is one slightly confusing detail in some of the coverage.

The wider Fremantle mental-health redevelopment cost $88.1 million and added 40 inpatient beds: 20 secure adult beds, 10 open adult beds and 10 beds for older adults with neurobehavioural needs.

Those beds and the crisis centre are part of the same redevelopment. They are separate services.

The government hopes the centre will get people assessed by mental-health clinicians earlier and take some pressure off emergency departments.

Sounds good.

Now we get to find out whether it actually works.

There is some reason for optimism. A 2022 systematic review of hospital-based short-stay mental-health crisis units found shorter emergency-department stays and fewer psychiatric admissions. ED time dropped by about 164 minutes on average.

Pretty impressive number.

Unfortunately, the units in that review aren’t a clean match for Fremantle. They were short-stay services where people could remain overnight for assessment and stabilisation. Fremantle looks more like a central point for access, assessment and referral.

Australian evidence is fairly thin too. A review published this year found only 15 studies across several different crisis-care models. Results were promising, but the services varied a lot and most of the research was observational.

So 164 minutes is interesting background. It isn’t Fremantle’s result.

We don’t have Fremantle’s results yet.

What I want to know now is fairly basic.

How long are people waiting?

How many still end up in an ED or inpatient ward?

How many come back soon afterwards?

Are people actually getting connected with ongoing community care?

And what do patients and carers think of the experience?

There’s also an odd bit of public messaging to sort out. The centre is promoted as another route into urgent mental-health care, while its own webpage still tells somebody experiencing an emergency or crisis to call 000 or attend their nearest ED.

Clearly there are boundaries around who the centre is designed for and when. I’d like those boundaries to become clearer as the service beds in.

For now, Fremantle has a genuinely different way into acute mental-health care. Some people can ask for help, be assessed by mental-health clinicians and get directed towards care without automatically starting in a general emergency department.

The next useful announcement will hopefully contain less ribbon-cutting and more numbers from the people walking through the door.

Read more

I don’t remember exactly when I came up with the name Lucid Intervals.

There was no notebook full of alternatives. No brand strategy. No dramatic moment where two vaguely literary words descended from the heavens and demanded a Substack.

I remember the thought behind it, though.

I was feeling okay.

Actually okay.

Clear enough to read. Clear enough to think about things for longer than five minutes. Clear enough to have ideas and occasionally do something with them.

After the previous year, that still felt temporary enough that I didn’t entirely trust it.

So I remember thinking that if I wanted to write, I should probably start now.

While I could.

This was a lucid interval.

So I opened Substack.

How I got here

The short version is that I spent about seven months in psychiatric hospitals across two stays.

The slightly longer version involves running coffee shops, burning myself into the ground, major depression, a lot of medication, twenty-five rounds of ECT, a suicide attempt, more hospital, coming home, and eventually trying to work out what exactly I was supposed to do with myself afterward.

I haven’t returned to work since.

For a while there wasn’t really a question of what came next. Getting through days occupied enough of the schedule.

Eventually I went back to study. I’m doing a Certificate IV in Mental Health now, partly because spending that much time on the receiving end of the system made me curious about the other side of it, and partly because I think lived experience might actually be useful if I can pair it with some proper training.

Somewhere around the same time, I started reading a little book called Being Logical by D.Q. McInerny.

Which is probably an odd choice of reading material after a psychiatric admission.

I liked it.

More importantly, I kept finding myself holding the ideas in the book up against things that had happened to me.

If logic is partly about getting from premises to conclusions without fooling yourself, what happens when depression starts supplying the premises?

What does clear thinking even mean when your own mind is unreliable?

Why can something feel completely, obviously true and still be wrong?

That became my first real piece here.

I wrote about walking out of one of my coffee shops, going to an emergency department and saying, “Help me. I can’t take it anymore.”

At the time it felt like collapse.

Looking back, it might have been one of the clearest things I’d said in years.

That was roughly the idea behind Lucid Intervals when I started.

Mental health. Logic. Recovery. Trying to understand what had happened to me.

Then, predictably, I wandered off.

It got a little broader

Since then I’ve written about recovery feeling like brushing your teeth.

I’ve written about lying to your own mood tracker and somehow involved a British economist from the 1970s.

I wrote about having twenty-five medically induced seizures and why this has resulted in an entirely unreasonable refusal to watch season four of Slow Horses.

I wrote about mental-health advice having minimum system requirements after a throwaway Note about having 4% battery ended up reaching far more people than I expected.

I’ve written about psychiatric language, medication supply, ADHD, comparison, failure, pessimism, and learning mindfulness by falling off a skateboard.

That last one wasn’t part of the business plan.

There wasn’t much of a business plan.

And somewhere in all of that, Lucid Intervals started making more sense to me.

I don’t think I want this to become a publication where every Tuesday I select a mental-health problem from a spreadsheet and explain it.

I’m interested in mental health because I’ve lived a lot of it. I’m studying it. I’m still dealing with it.

So it will keep appearing.

But I’m also interested in logic, systems, films, weird bits of research, things we believe about ourselves, the ways people measure progress, why sensible advice becomes useless in the wrong circumstances, and apparently skateboards.

Mostly I like taking something I can’t quite leave alone and looking at it until I understand it a little better.

Sometimes that produces an answer.

Sometimes it produces another question.

Sometimes it produces 1,500 words about refusing to watch Gary Oldman.

All three seem to belong here.

Somehow there are 320 of you

When I started this three months ago, around 30 people subscribed.

Most of them, I assume, because they knew me and clicking the button seemed polite.

Now there are 320 of you.

That number is small by internet standards and enormous by mine.

I’ve had Notes reach thousands of people. I’ve had articles disappear almost completely. I’ve published things I thought would do well that nobody read, then watched something I typed into my phone travel further than work I spent days on.

I have learned absolutely nothing reliable about predicting what people on the internet will like.

I have, however, met a lot of people.

Other writers. People in recovery. Clinicians. People with ADHD. People going through things they don’t have language for yet. People who disagree with me intelligently. People who make very good jokes in the comments.

That part has surprised me more than the numbers.

I expected writing to be fairly solitary.

Substack apparently disagreed.

The slightly awkward money section

There’s another part of why I started this that I’ve talked about less.

I haven’t worked since I went into hospital.

I’m studying now and getting closer to placement, but paid employment has been absent from my life for a while.

Writing became one of the first things I could do consistently again that felt like mine.

And, yes, I hoped there might eventually be some money in it.

I didn’t expect to start a Substack and immediately become the primary financial threat to Stephen King.

But I did hope that if people liked the work, some might eventually decide to pay for it.

Three months in, I’ve made about US$100.

Stephen remains safe.

I’m thrilled with what the publication has become. Going from essentially nobody reading to 320 people voluntarily allowing me into their inbox is still strange to me.

I also spent most of these first three months figuring out what Lucid Intervals actually was.

The paid side has mostly been something I occasionally remember exists.

I want to change that.

What happens next

I’m going to put more effort into the paid side of Lucid Intervals.

The main essays will keep being free. If you want to come here, read them, comment occasionally and disappear back into the internet, please do.

Paid subscribers get everything I publish beyond that: paid essays, bonus pieces, practical resources, downloadable tools, deeper research and the full paid archive.

I’ve already started experimenting with what that can look like.

The 4% battery piece became a low-power-mode toolkit because there was something practical sitting underneath the essay. Other pieces will lead somewhere completely different.

Some won’t lead anywhere at all.

I don’t want to manufacture a seven-page worksheet every time I have a thought. Nobody needs a PDF called Processing the Emotional Implications of Slow Horses Season Four.

Probably.

My birthday is also coming up, so I’m using that as an excuse to make paid subscriptions much cheaper for a while.

For at least the next month, an annual subscription is US$30.

If you’ve been considering supporting Lucid Intervals, this is probably the best time to do it.

Subscribe now

And if you haven’t, genuinely, come read.

Reply to things. Send me something you’ve written. Lurk quietly for six months and occasionally open an email at 2am.

You’re still part of this place.

The people paying make it easier for me to spend more time writing, researching and making the things that live around the essays.

The interval

The funny thing is that when I named this publication, I was thinking about the possibility that the interval might end.

I felt okay enough to write, and some part of me thought I should make use of that before it disappeared.

Three months later, I still don’t know what happens next.

I’m close to finishing a course for the first time in my life.

I’m about to go on placement.

I’m writing regularly.

There are 320 people here.

Some weeks I feel good. Some weeks I don’t. I still watch myself more closely than I’d like. I still know how quickly a life can change direction.

I’ve stopped requiring certainty before I do anything with the time in front of me.

I’m here.

I can think.

I can write.

Apparently some of you want to read it.

Seems like a decent use of the interval.

I have never completed a course. In fact, I technically don’t have a high school equivalent education here in Australia…

Through other pathways, I managed to get in to University when I just turned 18, a year before all of my friends that I left back in the UK.

First was Engineering, at the most prestigious university in the state. I’d never considered studying engineering, I was pushed into it by my parents because of where we lived and the job opportunities. Made it a year. Failed hard at maths and physics. I don’t regret it though, as I found my love of computing and also the love of my partner (13 years ago wow).

Over the next decade, I studied Computer Science (made it the furthest), Cyber Security, and Film, across two other institutions.

Now here I am studying Mental Health, doing well, praise from my lecturers, top marks for everything, about to go on placement. In fact, I’m like three months from being qualified.

And this whole time, I’ve had a very tidy argument for why this one will end up the same way.

I’ve never finished a course before, therefore I will not finish this one. An argument with evidence.

I’m told to believe in myself but why should I when I have a stack of evidence to the contrary. I have the transcripts and contacts and dates of years lost spent studying. Hell I made it so far into the compsci degree that I tutored first year classes and became the vice president of the student society.

Still didn’t finish it.

So when that thought shows up, “you won’t finish this one either”, I can’t just dismiss it as some depressive invention.

My résumé literally proves otherwise.

how far can a fact travel?

My résumé proves four courses ended without qualifications. That’s all it can prove. It knows absolutely nothing about what happens in December, because December hasn’t happened yet.

Which feels stupidly obvious written down. I know how time works. Yet somewhere in my head those two things have been flattened together for months:

I haven’t finished a course before.

I won’t finish this one.

There are actually a few steps missing between those sentences.

My history probably does make me more likely to drop out again. I’m not going to pretend four unfinished courses are irrelevant because positive thinking demands it. If somebody came to me with the same history, I’d probably tell them to pay attention to it.

There’s a pattern there. The problem is how far I let the pattern travel.

Four unfinished courses becomes I don’t finish courses.

Then I don’t finish things.

Then I won’t finish this.

Give it long enough and I get all the way to I won’t make anything of my life.

Quite an achievement for four university enrolments.

This is part of why I became interested in the whole idea of “distorted thinking” in the first place. The popular version is very simple: depression tells you something awful, you challenge the thought, discover it isn’t true, replace it with something more reasonable.

Except researchers have spent decades arguing about whether depressed people are actually less accurate about reality, and the answer appears to be a deeply unsatisfying sometimes.

A 2026 meta-analysis pulled together 130 articles and more than 33,000 people looking at depression and accuracy of judgement. Overall, people with depression were slightly less accurate, but the difference was tiny. In some kinds of complex, self-referential judgements, the depressed groups were actually a little more accurate. In severe clinical depression and some ambiguous social situations, distortion showed up more clearly. There doesn’t seem to be one switch labelled DEPRESSION that simply makes everything you think wrong.

Which makes sense to me. Some of the worst things I think about myself are ridiculous.

Others come with references.

when the event becomes the person

Psychology has a slightly less catchy way of describing part of this. Researchers talk about whether we explain bad events as stable and global.

Stable means something along the lines of: this will keep happening.

Global means: this says something about everything else too.

A meta-analysis of 86 studies found both kinds of explanation were associated with depression. Again, association. We can’t neatly point an arrow from one to the other and call the case solved. I can see both happening in my own argument.

Computer science wasn’t:

I left a computer science degree unfinished.

It became:

I’m someone who doesn’t finish things.

That’s a much bigger sentence.

I do it with parenting too.

I have spent a lot of time worrying that I won’t be a good parent because I didn’t exactly grow up with the blueprint I’d want to copy.

There is real history underneath that fear as well.

I learnt plenty about parenting by negative example. Things I noticed, things I remember, things I decided very early I wanted to do differently if I ever had children of my own.

Then my brain does its little administrative trick.

I didn’t grow up with the kind of parenting I want to give my kids.

That shaped me.

Therefore I’m probably going to repeat it anyway.

One generation becomes evidence for the next.

And again, I don’t want to throw the history away. Of course it matters. If I notice myself repeating something I hated growing up, I should probably pay attention.

But “this shaped me” has somehow become “this decides me.”

Those are very different jobs for the same evidence.

the prediction that gets to keep moving the deadline

The course thought has another useful trick.

It can’t really lose.

Say I finish my next assessment.

Does that prove I’ll finish the course?

Nope.

I finish the next unit.

Still plenty of time.

I make it through placement.

Cool. I made it 2.5 years into computer science. We’ve got precedent for a late collapse.

I could be standing there three days before graduation and some part of my brain would probably still have a folder open marked SEE, TOLD YOU SO, ready for the opportunity.

Every success is temporary evidence. One failure would be final evidence. That is a spectacular arrangement if your goal is to keep believing the same thing forever.

There is some research that looks at something close to this.

A 2025 review combined 29 studies and more than 3,000 people looking at how depression relates to changing beliefs when new information comes in. Depressive symptoms were associated with being slightly less likely to update beliefs in a positive direction when the new information contradicted what someone already believed.

The effect was small. Depression doesn’t install some force field around negative beliefs where compliments and good exam results bounce off harmlessly.

But the researchers didn’t find the same association when people were updating toward negative information.

That interests me because it looks a lot like the accounting system I use on myself.

Passed another assessment?

Pending.

Lecturer says I’m doing well?

Nice of them.

Top marks?

Good week.

Quit the course?

THERE IT IS.

One piece of negative evidence gets entered into the permanent record while the positive evidence spends six months on probation.

I’m not saying that study explains why I do this. It doesn’t.

I’m saying I recognise the shape.

some thoughts have much better evidence

University is an easy example because the worst possible outcome is fairly boring.

I quit another course.

Would suck. Life would continue.

There’s another prediction I make where I have much more trouble accepting uncertainty.

I will end up back in hospital.

I don’t mean this in an abstract “everyone gets sad sometimes” sense.

I have had severe depression.

I’ve spent around seven months in psychiatric hospitals across multiple stays.

I’ve had medication, therapy, DBT, ECT, and other treatments. Some helped for a while. Nothing so far has given me anything resembling permanent immunity.

I have been discharged before and become seriously unwell again.

So when my brain tells me I’m going back, this one has a fucking bibliography.

And unfortunately the actual research isn’t particularly useful if what I want is reassurance.

Previous episodes of depression really are one of the stronger known indicators that somebody may experience another episode. A systematic review of relapse and recurrence research found that a history of previous episodes, residual symptoms and childhood maltreatment were among the factors with the strongest evidence behind them.

So I have no intention of telling myself:

You’re never going back to hospital.

I don’t know that.

Another episode of depression is possible.

Another severe episode is possible.

I may need intensive treatment again.

All of those sentences are horrible enough without improving them.

And yet I regularly do.

Another episode is possible.

Becomes:

Another episode will happen.

Which becomes:

It will get as bad as before.

Then:

Treatment won’t work.

Then:

I’ll end up in hospital.

Then:

I’ll inevitably end up back in hospital.

Look at how much future I managed to manufacture from the word possible.

I’ve predicted the depression, the severity, the response to treatment and the admission before any of them have happened.

There’s an old piece of research I love mostly because of the phrase it gave the problem: depressive predictive certainty.

Psychologist Susan Andersen was interested in the difference between expecting bad things and being certain those bad things would happen. In two studies, increasing depressive symptoms were associated with both more pessimistic predictions and more certainty in those predictions.

The studies were small and included college students, so I’m not hanging a grand theory of depression on research from 1990.

But the distinction is useful.

Pessimism still leaves probability.

Certainty finishes the story for you.

And I’m extremely good at finishing stories about myself.

Courses I haven’t left.

Parenting mistakes I haven’t made.

Hospital admissions that haven’t happened.

Apparently I only struggle to finish things when they exist in reality.

how much does the evidence actually earn?

So I’ve stopped trying to win the argument by proving the original thought false.

Sometimes I can’t.

I really haven’t completed a course.

I really did grow up with plenty of examples of the parent I don’t want to be.

I really have tried a lot of things and failed at plenty of them.

I really have been severely depressed more than once.

Those can stay.

The more useful question for me is:

How far does that evidence actually let me go?

My education history lets me say I have a pattern of leaving courses unfinished.

Fine.

It probably lets me say completion is something I should take seriously rather than assume will happen because this time feels different.

It might even tell me I should look at why the other courses ended and make sure I’m paying attention if the same things start happening again.

Then I hit the edge.

The evidence cannot tell me whether I finish this course.

My history of depression lets me say recurrence is a real possibility.

That is enough reason to watch my warning signs. Enough reason to take sleep changes seriously. Enough reason to keep support around me when things are going well rather than waiting until they aren’t.

Then I hit the edge again.

None of my previous hospital admissions contains information from the future.

Even the parenting one looks different when I force myself to stay with the evidence I actually have.

My childhood matters.

So does the fact that I’m 31 now.

So does everything I’ve learnt since.

So does what I actually do with my own children.

My childhood is part of the evidence. It doesn’t get to be the entire dataset.

There is one answer I keep arriving at through all of this, and I hate it.

I don’t know.

I don’t know if I’ll finish my course.

I don’t know if I’ll ever end up back in hospital.

I definitely don’t know what my children will think of my parenting when they’re old enough to write essays about it.

“I don’t know” feels weak compared with certainty.

There’s nothing to grab onto.

No motivational ending where I decide that this time will be different because I’ve grown and learnt and finally found my purpose.

Maybe I have.

Still don’t know.

And I think that’s where I kept going wrong.

I thought the alternative to a terrible prediction had to be a good one.

I won’t finish.

Okay, then say:

I will.

Believe in yourself. Manifest. Put something inspirational on the fridge.

But I have exactly as much access to that future as I did the depressing version.

Zero.

I know what happened in engineering.

I know what happened in computer science.

I know what happened in cyber security.

I know what happened in film.

I can use all of that.

I’m still studying mental health.

I’m about three months from finishing.

The course is currently sitting there in my history without an ending.

For once, I’m going to leave it that way.

references and further reading

Yeung, J. C., Lam, H., & Chan, D. K. S. (2026). “Depression and accuracy of judgment: A meta-analysis.” Clinical Psychology Review, 128, 102752. DOI: 10.1016/j.cpr.2026.102752.

The most current overview of the depressive-realism debate. Across 130 articles and more than 33,000 participants, depression was associated with only a very small overall reduction in judgment accuracy. The direction changed depending on the task: some complex, self-referential judgments showed depressive realism, while severe clinical depression and ambiguous social judgments were more associated with distortion.

Read the paper on ScienceDirect

Zabag, R., Vander Zwalmen, Y., Kube, T., Hoorelbeke, K., Koster, E. H. W., & Joormann, J. (2025). “Belief Updating Deficits in Depression: A Systematic Review and Meta-Analysis.” Clinical Psychology Review, 121, 102649. DOI: 10.1016/j.cpr.2025.102649.

A preregistered meta-analysis of 29 papers involving 3,130 participants. Depression was associated with a small reduction in updating beliefs toward positive information, while no corresponding association was found for updating toward negative information.

Read the paper on PubMed

Hu, T., Zhang, D., & Yang, Z. (2015). “The Relationship Between Attributional Style for Negative Outcomes and Depression: A Meta-Analysis.” Journal of Social and Clinical Psychology, 34(4), 304–321. DOI: 10.1521/jscp.2015.34.4.304.

This review combined 86 studies and more than 51,000 participants. Depression was associated with explaining negative outcomes in more internal, stable and global ways. In the language of this essay, an event can become evidence about me, then evidence about what is always true, then evidence about other parts of my life. The findings are correlational.

Read the paper via the DOI

Andersen, S. M. (1990). “The Inevitability of Future Suffering: The Role of Depressive Predictive Certainty in Depression.” Social Cognition, 8(2), 203–228. DOI: 10.1521/soco.1990.8.2.203.

Andersen introduced the idea of depressive predictive certainty: pessimistic predictions becoming highly certain rather than merely negative. Across two studies, greater depressive symptoms were associated with both more pessimistic forecasts and stronger certainty in those forecasts. The original samples were limited, so I’d treat this as a useful concept rather than a settled account of depression.

Read the paper via Guilford Journals

Beck, A. T. (1963). “Thinking and Depression: I. Idiosyncratic Content and Cognitive Distortions.” Archives of General Psychiatry, 9(4), 324–333. DOI: 10.1001/archpsyc.1963.01720160014002.

One of the foundational papers behind the cognitive model of depression. Beck describes distortions including selective abstraction, arbitrary inference and overgeneralisation. It is useful here because a thought does not need to begin with a fabricated fact to become distorted; the problem can appear in what is selected or inferred from it.

Open the paper via DOI

Buckman, J. E. J., Underwood, A., Clarke, K., Saunders, R., Hollon, S. D., Fearon, P., & Pilling, S. (2018). “Risk factors for relapse and recurrence of depression in adults and how they operate: A four-phase systematic review and meta-synthesis.” Clinical Psychology Review, 64, 13–38. DOI: 10.1016/j.cpr.2018.07.005.

This review is useful for the hospitalisation section because it stops the essay from offering false reassurance. Previous episodes and residual symptoms were among the better-supported prognostic indicators of relapse or recurrence. That establishes genuine risk, while still leaving individual outcomes uncertain.

Read the full open-access paper on PubMed Central

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and try to answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

My son is five, which seems relevant because five-year-olds have not yet been issued the full administrative department responsible for worrying about things.

He moves from one skateboard to another, different sizes, different feel, gets on, has a go, gets off, gets on something else.

Meanwhile I’m standing nearby with the entire bureaucratic machinery of adulthood online.

Risk assessment.

Future projections.

Imaginary X-rays.

A small incline that barely deserves the dignity of being called an incline becomes a proposal requiring committee approval and a six-week cooling-off period.

He’s five.

I have functioning knowledge of gravity.

This feels unfair.

Then I get on.

And the world gets very small.

There’s my front foot and my back foot and the board underneath them and my weight, which is apparently a thing I’ve been carrying around my entire life without ever needing to know exactly where I put it.

There’s where I’m looking.

There’s whatever I’m trying to do next.

There’s the ground nearby with its extremely consistent policy on what happens when I get it wrong.

There isn’t much room left for anything else.

The conversation from yesterday. The thing I need to do tomorrow. Whether somebody is annoyed at me. The seventeen browser tabs apparently running somewhere behind my eyes at all times.

Quiet.

For a few seconds, sometimes longer, my mind shuts up.

And somewhere inside that quiet, a concept I have been taught over and over for years finally started making sense.

Mindfulness.

Which is irritating.

I’ve done therapy.

I’ve sat through group DBT sessions in hospital.

I’ve read the books.

I’ve tried the apps, many of which appear to have concluded that enlightenment is most likely to occur somewhere between pale blue and slightly darker pale blue.

I understood mindfulness.

At least intellectually.

Pay attention to what’s happening now. Notice thoughts as thoughts. Notice feelings as feelings. Let them come. Let them go. Return your attention.

I could explain it.

I struggled to actually do it.

Apparently what several qualified professionals, multiple books and the entire soft-gradient sector of the App Store could not quite get through my skull required a plank of wood and four wheels.

Fair enough.

The present moment has wheels

One of the more recent attempts to pin mindfulness down came from a 2025 review of definitions used in psychological science.

The researchers thought the present moment was still too abstract. Telling somebody to pay attention to the here and now raises a fairly obvious question: which bit of it?

Their proposed definition gives attention actual places to land: sensations in your body, whether something feels pleasant or unpleasant, thoughts and emotions, and what’s happening around you. Then there’s how you meet those things. They use the terms allowing and equanimity, roughly openness to what turns up and an even-mindedness as it changes. A 2026 systematic review of mindfulness measures has since used that newer definition as its conceptual starting point.

I like allowing.

Acceptance has always sounded slightly too much like signing for a parcel.

Allowing feels closer to what’s been happening on the skateboard.

Because I am scared quite a lot.

I’d love to tell you skating has made me fearless, partly because that would give this essay a convenient arc and partly because fearless sounds considerably better than adult man intimidated by tiny slope.

The fear keeps turning up.

Before I try something new there’s often that tightening first, then the thought:

No.

Bad idea.

We have survived this long. Why introduce wheels?

And now, increasingly, I notice it.

That has become the interesting part.

I can feel the anxiety gathering. I can hear the thoughts trying to turn it into a prediction about what happens next.

Then I leave them alone.

I might wait a bit. I might decide today isn’t the day. Sometimes I put my feet where they need to go and try while still scared.

The fear can sit there submitting its objections.

It doesn’t automatically get the deciding vote.

I already knew this

This is where DBT becomes slightly annoying too.

DBT breaks mindfulness into very practical skills. The three “what” skills are observe, describe and participate. The “how” skills are nonjudgmentally, one-mindfully and effectively.

I learned these in hospital.

I understood them.

The skateboard gave them somewhere to go.

Observe.

I’m scared.

Describe.

There’s the fear. There’s the thought that I’m going to fall.

Participate.

I’ve decided to try, so come back to what I’m actually doing.

One-mindfully.

Feet.

Weight.

Board.

The present moment becomes considerably easier to locate when leaving it has immediate consequences.

You cannot really be on a skateboard three Thursdays ago.

You cannot move your centre of weight next Tuesday.

The board is deeply unreasonable about scheduling.

It requires you now.

And then sometimes I fall over.

Bruises lack ambition

I have become quite good at this part.

By which I mean falling over, obviously.

There are bruises. Small ones. Ordinary ones.

What’s surprised me is how often they stay close to their original size.

I fall.

It hurts.

There is occasionally a bruise.

I get back on.

The bruise does not submit a written assessment of my character.

It doesn’t know whether I have natural talent. It has no opinion on whether I should give up.

It is just sitting there being purple.

I like this about bruises.

They lack ambition.

My mind can be much more entrepreneurial.

Give it one unpleasant thing and it can open three new franchises before lunch.

A wobble becomes fear. Fear starts predicting. A prediction hangs around long enough to acquire the authority of fact. Eventually the thing never gets attempted, which very helpfully preserves the prediction forever.

Skating keeps interrupting that process.

Sometimes the feedback is too immediate.

My weight was wrong.

I hesitated.

The board went one way and I went another.

Then another attempt is sitting there waiting.

So I get on.

Feet.

Board.

Weight.

Go.

And every now and then there is this thin clean interval where nothing is being rehearsed or predicted or replayed, where the past stops demanding an appeal and the future stops sending advance copies of disasters that haven’t happened yet.

Everything narrows down to the small patch of reality underneath me.

A board under my feet.

Fear somewhere in my chest.

Concrete moving past.

The next second arriving exactly when it is supposed to.

Then my son gets scared.

He’s five

Until this point, I’ve mostly been watching him.

Watching him jump between different boards and get straight into it. Watching somebody who has only existed for five years approach a new physical problem without first requesting its complete psychological risk profile.

Then we get to a small incline.

And he stops.

He doesn’t want to do it.

He’s scared.

Suddenly I can see the hesitation from the outside.

The distance between where he is and where he wants to go.

The little internal no.

So I do it.

He watches me.

Then he tries.

That’s all.

There wasn’t a speech.

I didn’t crouch beside him and explain recent developments in the psychological definition of mindfulness.

He’s five. He would have been entirely justified in abandoning me halfway through the phrase present-centred awareness.

He saw me do it.

Then he had a go.

Did mindfulness ever suddenly make sense
somewhere you weren’t expecting it to?

Leave a comment

I keep thinking about that because until then I’d been borrowing something from him.

His willingness to climb onto another board without ceremony. His readiness to try. The way a failed attempt seemed to be followed, quite often, by another attempt.

Then he found the thing that scared him.

And for one tiny incline, he watched me go first.

I was scared enough to understand why he had stopped.

I did it anyway.

Then he did too.

I have no idea whether he’ll remember any of this.

Probably not. He has a lot going on.

But I will.

Because somewhere between watching him skate and having him watch me, all those mindfulness instructions I’d collected over the years stopped behaving like instructions.

I had spent a long time trying to understand them from inside my own head.

Now my body keeps explaining them back to me.

The fear arrives.

There it is.

My mind produces its objections.

They can stay.

My feet still have a job.

Then, when I decide to try, the world narrows again, and all that useless distance my brain is so good at creating starts to disappear.

Yesterday loosens its grip.

Tomorrow can wait its turn.

Four wheels roll underneath me.

And for a few seconds, the whole world fits between my feet.

References & further reading

Chems-Maarif, R., Cavanagh, K., Baer, R., Gu, J., & Strauss, C. (2025). “Defining Mindfulness: A Review of Existing Definitions and Suggested Refinements.” Mindfulness, 16, 1–20.
A recent review of how mindfulness is defined across psychological science. The authors propose a more concrete definition centred on awareness of bodily sensations, thoughts, emotions and the external environment, combined with an allowing and equanimous attitude. This is the basis for the definition used in this essay.
Read the open-access paper

Linehan, M. M. (2025). DBT Skills Training Manual: Revised Edition. Guilford Press.
The current edition of Marsha Linehan’s DBT skills manual. It covers the core mindfulness “what” skills, observe, describe and participate, and the “how” skills, nonjudgmentally, one-mindfully and effectively, which I was originally taught in DBT groups.
DBT Skills Training Manual

Paechter, C., Keenan, M., Stoodley, L., & Lawton, C. (2025). “‘Free therapy’: Young woman skateboarders, mental health and body self-compassion.” International Review for the Sociology of Sport, 60(2), 362–380.
A qualitative study of young women skateboarders. Participants described the deep concentration involved in skating, the sense of flow it could produce, and experiences of their minds becoming quieter while on the board. It doesn’t establish skateboarding as a mindfulness intervention, but the experiences reported are strikingly similar to some of what I describe here.
Read the paper

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

1. My hands started hurting halfway through legal and ethical class.

2. The material is dry and there is a lot of it, pages of legislation and responsibilities and standards and notes about what you should do and what you absolutely should not do, and by some point in the morning every line I wrote was sending this dull ache through both palms and into my wrists, just enough that I kept shifting the pen around in my fingers and opening and closing my hands under the desk.

3. I knew why they hurt.

4. The day before, my five-year-old son was sitting on the couch in our living room with his legs stretched out, pushing his feet underneath the rug.

5. Not really doing anything wrong (he’s 5), but annoying.

6. I told him not to. He moved them. I carried on.

7. Then as I got up to go to the kitchen I walked past the couch and he put them out again.

8. I stepped on his leg.

9. I could have walked another way, out of reach, barely any effort on my part.

10. I could have avoided him completely. I knew he’d been putting his legs there, because I’d already told him, I saw him move them and assumed that was the end of it, that he’d listened.

11. I told him not to put his legs there.

12. He’s five.

13. He listened.

14. But he’s five.

15. He put them back out almost immediately.

16. I should have known better.

17. He’s five.

18. Then he screamed.

19. A proper scream, sharp enough that for a second there was nothing else happening in the room, the house, the street,

20. He cried and collapsed on the floor and called for mum who ws right there and saw everything.

21. I went towards him. He didn’t want me.

22. I tried to talk to him. He wouldn’t let me.

23. He wanted mum, and I stood there looking at my child in pain because of something I had done while my brain started producing reasons, defenses, accusations, anything it could find quickly enough to make sese of what was happening.

24. I’d warned him.

25. (but he’s five).

26. He shouldn’t have put them back.

27. (but he’s five).

28. I couldn’t have known he’s do it exactly at that moment.

29. I knew what he’d been doing.

30. I didn’t mean to hurt him.

31. I hurt him.

32. I could have walked around.

33. I didn’t.

34. All of those things were true at the same time and none of them changed the fact that he was on the floor screaming because I had stepped on him.

35. I’ve been intentionally hurt by my parents before.

36. I know what it feels like to have a parent hurt you and mean it, and somewhere a long time a go I made a promise that probably sounds a lot simpler before you actually have children.

37. I would never do that to mine.

38. I would never hurt them on purpose. I would never make pain part of discipline. I would never use fear because I was bigger or angry because I could. I would never make the person who was supposed to be safe into the person they needed safety from.

39. Then my son was on the floor screaming.

40. And for a few seconds that promise lost all its qualifications.

41. I will never hurt my children.

42. I had hurt my child.

43. The word accidentally was still available. So was intention. So was context. I knew he had moved his leg back underneath me. I knew I hadn't chosen to hurt him. I knew, somewhere underneath everything else, that stepping on your child's leg by mistake and deliberately hurting your child are morally very different things.

44. The scream didn’t know, didn’t care.

45. I was furious with myself.

46. I couldn't think of anything worse than hurting my child, even unintentionally.

47. So I hit myself in the head.

48. Hard.

49. Then again.

50. I haven't done that in a long time. Since hospital, I think. Maybe before.

51. There wasn't much thought between the anger and doing it. I didn't decide that punishment was appropriate or work through some internal argument about what I deserved.

52. I was angry at myself, then I hurt myself.

53. Afterwards I left the room to go lie in bed.

54. I stayed there until dinner.

55. My son was fine within a few minutes.

56. He's sensitive, and I don't mean that in the way people sometimes use sensitive to explain why a child is being difficult or dramatic.

57. It's one of the things I love most about him.

58. He's caring. He notices when other people are upset. Things matter to him and people matter to him, and his feelings can be huge because he seems to experience them without much of the filtering adults gradually learn to put between themselves and everybody else.

59. I hope he keeps that.

60. I hope nobody teaches him that the easier way to be a boy, and eventually a man, is to care less or at least get better at pretending he does.

61. The difficult side is that big feelings sometimes arrive very loudly.

62. Pain does.

63. Shock does.

64. Fear does.

65. He’s five.

66. Something can become his whole world for five minutes and then five minutes later there is another whole world waiting.

67. So he screamed for mum.

68. He didn’t want me near him.

69. I knew all this about him already. I should have known that too.

70. Should.

71. I should have walked another way.

72. I should have remembered he'd probably put his legs straight back.

73. I should have known how frightened he'd be.

74. I should have stayed calm.

75. I should have waited.

76. I should have done something useful.

77. There are a lot of ways to put yourself on trial after something happens, especially when you already know what verdict you're afraid of.

78. Mine had been waiting for years.

79. What kind of father hurts his child?

80. I only knew one answer.

81. My son was allowed to be upset with me.

82. I want him to be able to scream when something hurts.

83. I want him to be able to call for mum when mum is who he wants.

84. I want him to be able to tell me no, to refuse comfort from me, to be angry with me, to decide that for the next few minutes I am absolutely the last person he wants anywhere near him.

85. I don't want my children learning that Dad's feelings have to be protected whenever Dad has done something that upset them.

86. Which means there will be times when I have to stand there while one of them is angry with me and do nothing about it yet.

87. He’s five.

88. I’m thirty-one.

89. He gets to be overwhelmed by the moment.

90. I have to learn how to stay outside it long enough to be useful.

91. I didn't manage that very well this time.

92. He wouldn't let me repair anything.

93. I couldn't pick him up. I couldn't comfort him. I couldn't make his leg stop hurting. I couldn't explain that I'd warned him without sounding like I was blaming a five-year-old for being five, and I couldn't explain that it was an accident because at that moment he didn't need a philosophical distinction between intention and outcome.

94. His leg hurt, I had done it, he wanted mum.

95. There was nothing for me to fix yet.

96. And guilt really wants a job.

97. Fix it. Undo it. Apologise. Explain. Pay something. Make the other person stop hurting so you can know you're forgiven and move on.

98. He wasn’t available for any of that (remember, he’s five).

99. Repair had to wait, but punishment didn’t

100. So I hit myself. I left. I went to bed and stayed there until dinner because I was so upset about failing, for a few seconds, to be the father I wanted to be that I removed that father from the house for the rest of the afternoon.

101. My son had already recovered and I was still in bed.

102. I haven't hit myself in ages, and when you stop doing something for long enough you start putting it safely in the past tense.

103. I used to do that.

104. I did that when I was unwell.

105. That was before.

106. Then one afternoon that changes.

107. I hit myself.

108. Again.

109. I don't know yet how much meaning to attach to that.

110. I would like to.

111. I'd like to know whether it was one terrible moment, whether it was an old reflex under enough pressure, whether it's a warning sign, whether I'm slipping, whether asking whether I'm slipping is itself reassuring or completely irrelevant.

112. I don’t know.

113. That uncertainty is probably why we haven't talked about it yet.

114. She was there. She saw everything.

115. She saw me step on him. She saw him scream. She saw me hit myself.

116. And then somehow we moved on with the afternoon without discussing the part that both of us know carries more history than the accident itself.

117. She worries about me going backwards.

118. I worry about me going backwards.

119. We both know what hospital looked like.

120. We both know how long it has taken to get from there to here.

121. We both know what an old behaviour returning might mean.

122. Or might not mean.

123. Probably doesn’t mean.

124. She reads everything I publish, so hopefully this will suffice. I don’t think I’m slipping.

125. There is something slightly absurd about studying to work in mental health while occasionally requiring a Substack essay to process my own thoughts or begin a difficult conversation in my own house.

126. The next morning I went to class.

127. Legal and ethical.

128. My son was fine.

129. Had been after just a few minutes.

130. The house was fine, relationships fine, family fine.

131. Nothing had collapsed simply because one terrible minute had happened in our living room.

132. I sat down, opened my notebook and started writing.

133. That’s when I noticed my hands.

134. Every time I pressed the pen into the page, both palms hurt.

135. Not badly.

136. Enough.

137. So I kept shifting my grip and taking notes about responsibilities and rights and how people are supposed to behave when somebody trusts them to care for them, while the previous afternoon sat there quietly in my hands.

138. I kept thinking about the promise.

139. I will never hurt my children.

140. I still mean it.

141. I also think I understand it differently now.

142. He's five.

143. I’m going to get things wrong.

144. I am going to misunderstand him. I am going to lose patience sometimes. I am going to say something I regret, miss something I should have noticed, make decisions that upset him and occasionally, apparently, stand somewhere I should have walked around.

145. Those moments will have different weights.

146. Some will need an apology.

147. Some will need me to change.

148. Some will need a conversation.

149. Some will need me to give him five minutes to scream for Mum because his leg hurts and he's five.

150. The promise can't be that I will somehow get through the rest of his childhood without ever being the cause of pain.

151. I can’t keep that promise.

152. But I can stay.

153. I can be there when he is ready.

154. I can let him be angry without making his anger about whether I am a good person.

155. I can apologise without needing him to make me feel better.

156. I can notice when something old and frightening shows up in me and tell the people who need to know.

157. And I can try, next time, to remember which one of us is five.

158. The next morning his leg didn't hurt.

159. My hands did.

160. I moved the pen slightly in my grip and kept taking notes.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

A bad morning has an extraordinary ability to produce theories about your personality, career, relationships and future before lunch.

You sleep badly. Something small goes wrong. A task you should have finished yesterday is still sitting there. Someone replies to a message in a way you read three times and like less each time.

By 11am, maybe you’re getting worse again.

Maybe you’ve chosen the wrong career.

Maybe people are getting tired of you.

Maybe the good stretch you’ve been having wasn’t progress after all.

Maybe this is simply who you are.

An impressive amount of ground for one morning to cover.

My own biography is very cooperative with this sort of thing.

I’ve dropped out of university multiple times. I studied computer science for years, did well, became a tutor, got close to finishing, then left. Cyber security lasted six months. Film followed. I’ve started drawing and stopped. Tried writing novels and stopped. There are old app ideas, game ideas and half-built things scattered across hard drives.

I once wrote an entire essay called Thirty-one years of quitting.

So when the theory arrives that I’m fundamentally someone who doesn’t finish things, my brain doesn’t have to make anything up.

It has receipts.

That’s what makes these theories so convincing.

Sometimes the facts are true.

one fact starts recruiting

Psychologists have studied a version of this under the fairly self-explanatory name overgeneralisation: taking one negative experience and drawing much broader conclusions from it.

In a 2017 experiment, people with depression showed a greater increase in general negative beliefs about themselves after a failure experience than a non-clinical control group.

The movement interests me more than the terminology.

I didn’t finish this.

I keep doing this.

I never finish anything.

I’m the kind of person who never finishes anything.

Once you’ve reached that sentence, the archive opens.

Computer science gets called in.

Cyber security.

Film.

The novels.

The abandoned drawing phase.

Everything unfinished gets admitted as evidence.

The completed things become surprisingly hard to find.

You started with something that happened this week and somehow ended up explaining your character.

Depression research keeps finding versions of this move from specific to general.

A large 2023 review of autobiographical memory research found that people with depression tended to recall fewer specific events and more general categories of events.

The difference looks roughly like this:

I abandoned that project after six weeks.

I always abandon projects.

That research doesn’t mean one terrible Tuesday rewrites your memory into categories. Human beings are inconveniently more complicated than that.

I still recognise the direction.

That happened.

This happens.

This is me.

then you start investigating yourself

Rumination is extremely good at keeping the process moving.

Why did I do that?

Why do I always do that?

Why am I like this?

Where else have I done this?

What if this explains everything?

Given enough time, the investigation can cover most of your known history.

Researchers have found that the level of those questions matters.

In one experiment, depressed participants were pushed toward different styles of self-focused thinking. More analytical, evaluative thinking increased broad negative judgements about themselves, including judgements around worthlessness and incompetence.

The useful distinction is much less technical.

Compare:

Why am I always like this?

What happened this morning?

The first requires a working theory of Jacob Chapman.

The second requires an account of a morning.

I know which one I’d rather attempt while tired, low, embarrassed, angry, or already convinced I’ve ruined everything.

The larger question can wait.

It rarely volunteers to.

unfortunately, you might be right

Here’s the awkward part.

Your relationship might actually be struggling.

Your workload might genuinely be too much.

Three missed deadlines could mean something needs changing.

A worsening mood could be an early warning.

You can notice an unpleasant truth on a bad day.

There has been an argument in psychology for decades around depressive realism, the possibility that depression can sometimes produce more accurate judgements rather than simply making everything look worse.

A large 2026 meta-analysis looked at more than 33,000 participants. Across the research, depression was associated with only a very small overall distortion in judgement. The result changed depending on what people were being asked to judge. In some circumstances, people with depressive symptoms were more accurate.

Which is inconvenient.

I’d quite like a rule saying every terrible conclusion I reach while depressed can be thrown straight in the bin.

I don’t have one.

Sometimes I’m wrong.

Sometimes I’m partly right.

Sometimes I’ve noticed something I should have noticed weeks ago.

Mood doesn’t provide the answer key.

That changes the question.

Say I’ve missed three deadlines.

Those are real.

Maybe I’ve taken on too much. Maybe my concentration has dropped. Maybe I need help. Maybe I need to change something.

Then comes:

I am incapable of following through on anything.

Look how much larger that sentence is.

Three deadlines are now being asked to establish something about my character.

Then:

I’ve always been this way.

Now they’ve been promoted again. They’re explaining the past.

Then:

I’m going to quit this too.

Apparently they can see the future as well.

The evidence stayed the same.

The conclusion kept getting bigger.

check the size

I’ve become suspicious of explanations that suddenly account for my entire life.

Especially when they arrive on days when everything already feels harder.

I used to think the job was working out whether the thought was true.

I’m increasingly more interested in another question:

How much does this evidence actually prove?

Three missed deadlines might prove I’m struggling with deadlines right now.

They might tell me I’ve taken on too much.

They might tell me something has changed.

They don’t automatically explain every unfinished thing I’ve ever left behind.

A conclusion about this week needs some evidence.

A conclusion about this course needs more.

A conclusion about every course I’ll ever attempt needs quite a lot.

A complete theory of my personality probably needs more than a difficult Thursday and an unanswered message.

Logic has a boring answer to this. A conclusion only gets what its premises can support.

I keep forgetting that my own conclusions are subject to the rule.

no biographies on bad days

My actual habit is fairly crude.

When everything feels unusually hard, I get suspicious of any explanation that becomes unusually comprehensive.

The cleaner version is:

No biographies on bad days.

If I missed something important, I deal with the missed thing.

If I upset someone, that might need fixing.

If I notice something that still bothers me tomorrow, or next week, I can look at it properly.

Some theories survive.

That’s useful information too.

The point of waiting is to see how much of the theory survives once one bad morning isn’t supplying all the evidence.

My history of quitting things is real.

I left those degrees.

The unfinished projects exist.

There are folders on old hard drives that have been waiting for me longer than some friendships survive.

I’m also still studying now.

I’m still writing.

I’ve stayed with things I once expected myself to leave.

The biography built entirely from the abandoned things contained plenty of facts. It still got the story wrong.

A bad day is good at that.

It finds whatever fits the morning, arranges it into a plot, then starts writing an ending before lunch.

Tuesday can tell me something went badly on Tuesday.

Anything larger can come back tomorrow and make its case.

For now, Tuesday doesn’t get to write the book.

If you want to take this a step further, I’ve made a paid companion called The Bad-Day Theory Test: a printable worksheet and quick-reference card for coming back to those enormous conclusions once the day itself has passed. It helps you separate what actually happened from the theory it produced, check how far the evidence really reaches, and work out what still deserves your attention later. Paid subscribers can download it from the perks page (or below), and also get the full research and further reading behind this piece.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

There’s a ten-year-old boy called Mason in this new heavily commented on Channel 4 documentary about ADHD.

He’s been taking lisdexamfetamine for a couple of years and apparently it works pretty well, at least by the usual standards we use for these things.

He can concentrate better at school.

His grades improved.

His mum talks about the difference in his school reports.

Mason himself puts it more simply.

“It takes my fun-ness away. But I have to, and it works.”

It works.

He doesn’t like what it does to him.

Both can apparently be true at the same time.

Which is inconvenient, because the documentary is called The Great ADHD Myth? and it has much bigger questions it wants to answer.

Is ADHD actually a neurodevelopmental condition?

Are we diagnosing normal childhood behaviour?

Is it phones?

Is it ultra-processed food?

Is it schools making children sit still all day?

Are we medicating children because society is unwilling to accommodate them?

All fair things to ask, to varying degrees. Channel 4 itself promoted the documentary around the question of whether ADHD is genuinely neurodevelopmental or something created by modern life.

I have ADHD, so obviously I have some skin in this particular argument, but I’m also probably less interested than most people in defending the psychiatric system as some flawless machine that hands down objectively correct diagnoses.

I’ve had a psychiatrist decide I was bipolar, stop my ADHD medication and create a mess I’m still trying to completely undo while later doctors basically went, yeah, we don’t think you’re bipolar.

So.

Doctors can be wrong.

Diagnoses can be wrong.

Medication can be prescribed badly.

None of those ideas make me particularly uncomfortable.

What I struggle with is how the documentary tries to find out.

Six changes, one child

Mason comes off his medication for six weeks.

Only he doesn’t really just come off his medication.

His screens and gaming go too.

His diet changes. More protein, less sugary and ultra-processed food.

Supplements.

Yoga.

More time outside.

More one-to-one time with his family.

There’s a coach involved, a nutritionist and, because this is television, a camera crew documenting the whole thing.

Then he gets happier.

Apparently more social too. More expressive. His mum sees more of him again.

The fun-ness comes back.

School goes the other way.

By week five, his teacher is reporting more fidgeting, talking and disruption. Work he could previously finish becomes harder. He gets frustrated.

The documentary’s end credits later reveal that after poor exam results, Mason went back on medication.

Researchers were fairly brutal about this after the programme aired. Professor Katya Rubia, who appeared in the documentary, said her contribution had been cherry-picked and misrepresented. Other ADHD experts pointed out the slightly obvious problem with calling Mason’s six weeks an experiment: half his life changed at once.

If Mason gets happier, which thing did it?

Stopping the medication?

More time outside?

Less gaming?

More attention from his family?

A different diet?

Yoga?

Maybe yoga is secretly the answer to everything and every annoying person on Instagram has been right all along.

We can’t tell.

And even if we could, I think there’s another jump happening that bothers me more.

Say taking Mason outside every afternoon genuinely improves his concentration.

Great.

That tells us taking Mason outside every afternoon helps his concentration.

Somewhere along the way that gets turned into evidence about what caused his ADHD in the first place.

I don’t see how we got there.

Exercise can help depression. Nobody goes for a run, feels slightly better and announces we have finally discovered that depression is caused by insufficient jogging.

Something helping with a condition and something causing the condition are two different questions.

Apparently that still needs saying.

The environmental argument is barely an argument

There’s another thing I find strange about all of this.

Mainstream ADHD treatment already agrees that environment matters.

NICE, the UK body responsible for clinical guidelines, recommends reducing distractions, changing seating and noise where useful, breaking up long periods of concentration, allowing movement breaks and backing verbal instructions up with written ones.

For children aged five and over, NICE recommends medication when significant impairment remains after environmental changes have been tried and reviewed.

The Australian ADHD guideline takes a similar approach. It recommends multimodal treatment, with medication generally doing more for core ADHD symptoms while other approaches can help reduce the impact ADHD has on everyday functioning.

So “what if we changed the environment around the child?” is hardly a dangerous idea medicine has been desperately trying to suppress.

Yes.

Please change it.

I would have quite liked someone to change a few environments around me before I spent thirty-one years assuming the problem was entirely that I was shit at being a person.

I function very differently depending on where you put me.

Give me something interesting, some urgency, freedom to move around and a deadline close enough that I can hear it breathing and I can get an absurd amount done.

Give me a boring task with no deadline and tell me I should probably get around to it sometime this month and I will apparently leave it untouched until the sun burns out.

The environment changes what I can do.

That still leaves the question of why the same problems have followed me through wildly different environments for most of my life.

And before this starts sounding like I’m blindly defending every ADHD diagnosis, there are serious questions there too.

An Australian-led review in JAMA Network Open looked at 334 studies and found evidence of ADHD overdiagnosis and overtreatment in children and adolescents. Twenty-five studies suggested that additional diagnoses were increasingly capturing milder cases, while very few had actually examined whether the benefits outweighed the harms for those children.

That deserves attention.

A diagnosis can describe something real and still be given to the wrong person.

Medication can help and still be prescribed when the benefit isn’t worth the cost.

Psychiatry being fallible does not require psychiatry to be fiction.

We seem to have become really bad at holding more than one of these ideas at once.

Which brings me back to Mason.

“It works”

“It takes my fun-ness away. But I have to, and it works.”

What does “works” mean there?

Better grades?

Less fidgeting?

Finishing worksheets?

Being able to sit through a lesson?

Those are real things.

I don’t particularly buy the argument that school performance is some meaningless demand imposed by adults and children should simply be left alone to be their authentic selves.

Repeatedly failing at things everyone around you can do is miserable.

I know.

Being unable to start or finish work matters. Falling behind matters. Spending every school day collecting evidence that everyone else seems able to do things you cannot manage matters.

Medication can help with that.

A large 2025 BMJ review pulled together 221 meta-analyses of randomised trials of ADHD treatments. For children and adolescents, several medications had moderate-to-high certainty evidence for reducing ADHD symptoms in the short term. The picture became much less certain over longer follow-up, where the researchers found no moderate or high certainty evidence for sustained benefits from any of the drug or non-drug interventions they examined. That is a limit in the evidence we currently have, rather than proof that treatment suddenly stops helping after a particular date.

Then there’s fun-ness.

A 2023 review looked at studies where children and adolescents taking methylphenidate talked about their own experience of it.

Some described better control, better school performance and better relationships.

Others talked about pressure to become “normal” or behave the way adults wanted. Questions about self-esteem and control kept appearing even when researchers had not specifically asked about them. The authors argued that these experiences should actually be part of the decision about treatment.

Which seems obvious once you say it out loud.

The adults can see whether Mason finishes the worksheet.

Mason is the only person who can tell them what being Mason feels like while he does it.

And I think this is where I’ve ended up after reading far too much about a Channel 4 documentary I originally thought was just going to annoy me.

Maybe “does it work?” is an unfinished question.

Treatment has a bunch of different scoreboards running at once.

Symptoms.

Functioning.

Side effects.

Quality of life.

How everyone around you experiences you.

Then somewhere in the middle of all that there’s you, having to actually be the person taking the tablet.

Those scoreboards can disagree.

Mason comes off his medication.

He seems happier.

School gets worse.

Which result counts?

Both, surely.

Which Mason counts?

This seems much more interesting to me than whether his six weeks somehow disprove ADHD.

If my doctor thought something was working because I was completing more tasks while I felt miserable taking it, I’d want the misery included.

And if I felt fantastic after stopping something while the rest of my life started falling apart again, I’d probably need somebody to tell me that too.

Neither person gets automatic ownership of the answer.

Maybe this is also where my own experience makes me suspicious of clean medical stories.

I’ve had treatments help one thing and make another harder.

I’ve had doctors looking at me from one side of a desk while I was experiencing myself from the other.

I’ve had words like “recovered” mean something completely different depending on who was saying them.

Mental health seems full of this.

Several descriptions of the same person can be true at once, and eventually somebody has to make an actual decision.

Which is harder.

And nowhere near as good a documentary title.

The Great ADHD Myth? went looking for an answer about whether modern life created ADHD and whether children really need medication.

It accidentally found a child whose medication seemed to work and also seemed to cost him something.

Then it removed the medication, changed a pile of other things around him, and found a child who seemed happier and also struggled more at school.

Eventually he went back on the tablets.

I don’t think that gives us a winner.

Maybe the dose changes. Maybe the medication changes. Maybe they keep the parts of the environmental changes that helped. Maybe he makes a different decision again when he’s older.

I have absolutely no idea what is best for one ten-year-old I have never met, and I’m fairly sure nobody should be using him as the deciding vote on whether ADHD exists.

But I do think he said something worth hanging onto before the whole experiment started.

It works.

It takes my fun-ness away.

I think you probably have to keep both.

Sources and further reading

The Great ADHD Myth? — Channel 4, 2026
The documentary discussed throughout this article, presented by psychiatrist Max Pemberton and first broadcast on 18 August 2026.
Watch The Great ADHD Myth? on Channel 4

Science Media Centre: “Expert reaction to the TV show ‘The Great ADHD Myth?’” — 19 August 2026
Responses from ADHD researchers and clinicians following the broadcast. Particularly useful for the criticism of Mason’s six-week experiment, the multiple variables changed at once, and Professor Katya Rubia’s account of how her interview was presented.
Read the expert responses

NICE: Attention deficit hyperactivity disorder: diagnosis and management (NG87)
The UK clinical guideline for ADHD. I used this for its recommendations on environmental modifications, medication, shared treatment decisions and reviewing whether medication should be continued.
Read the NICE recommendations

Australian Evidence-Based Clinical Practice Guideline for ADHD — Australasian ADHD Professionals Association
The Australian clinical guideline. Its treatment recommendations are explicitly multimodal: medication primarily targets core ADHD symptoms, while non-pharmacological support can help with functioning and the day-to-day impact of ADHD.
Read the Australian ADHD guideline

Kazda et al. (2021), “Overdiagnosis of Attention-Deficit/Hyperactivity Disorder in Children and Adolescents: A Systematic Scoping Review” — JAMA Network Open
A systematic scoping review of 334 studies. It found evidence of overdiagnosis and overtreatment, particularly concerns around additional diagnoses among children with milder symptoms, while also identifying major gaps in evidence about the balance of benefits and harms in those cases.
Read the study

Gosling et al. (2025), “Benefits and harms of ADHD interventions: umbrella review and platform for shared decision making” — The BMJ
A large umbrella review reanalysing 221 meta-analyses of randomised trials. It found moderate-to-high certainty evidence for short-term symptom benefits from several ADHD medications in children and adolescents, alongside much weaker evidence at medium and long-term follow-up.
Read the BMJ review

Jacobs et al. (2023), “Experiences of children and adolescents with attention-deficit/hyperactivity disorder taking methylphenidate” — Developmental Medicine & Child Neurology
A narrative review of qualitative research focused on young people’s own experiences of methylphenidate. It examines self-esteem, control, academic performance, relationships and the pressure some young people felt to conform to adult expectations around behaviour and medication.
Read the review

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

After the paywall, subscribers get access to a toolkit I’ve used and adapted for myself, examples, and further reading and research. If you’d like access to those resources but can’t currently afford a subscription, please shoot me a message.

Journal. Go for a walk. Read. Cook something healthy. Reach out to a friend. Keep a routine.

All perfectly reasonable things to suggest to someone who’s struggling.

Now give them to someone with 4% battery.

That was the question I posted on Substack a few days ago:

Four percent is obviously a made-up number; I wasn’t walking around during the worst of my depression with a little battery icon floating above my head. Some days probably would have qualified, though.

I couldn’t journal. I couldn’t read through the brain fog. Eating sometimes depended on somebody putting food in front of me.

There was plenty of advice available for all three.

The problem was getting from the advice to the part where I actually did it.

Check the system requirements

Every piece of mental-health advice has minimum system requirements.

Take journaling.

From the outside, the instruction is tiny. Find some paper, write down what you’re thinking, and maybe answer a prompt.

Except you have to remember to do it. Start doing it. Concentrate long enough to work out what you’re thinking. Find words for it. Keep track of those words. Possibly sit with whatever you’ve just dragged onto the page.

There were periods when I couldn’t read a book because my brain fog was so bad, and a blank page wasn’t going to fix that.

Eating was similar.

“Make sure you’re eating properly” sounds like one task, when in practice, it can contain deciding what to eat, checking whether you have it, getting it, preparing it, eating it and dealing with whatever remains afterward.

Going for a walk has its own quiet collection of prerequisites. So does calling a friend. So does finding a therapist. So does building a morning routine.

The advice might take six words, but following it can take twenty decisions.

Then one reader, Anna Keenan, added another requirement I’d missed:

Fair - energy is only one currency.

The Australian Bureau of Statistics found that in 2024–25, 19.7% of people who needed a mental-health professional said cost was a reason they delayed or didn’t see one. For psychologists and psychiatrists, the figure was about one in four.

So now our hypothetical person has 4% battery, $5, brain fog, a pile of decisions to make and several people on Instagram explaining the importance of magnesium.

Good luck.

I couldn’t journal

Journaling is probably the clearest example from my own experience because it is such ordinary mental-health advice, and I’ve honestly got nothing against it.

I simply couldn’t do it when I was at my worst.

Reading disappeared too. I normally love reading, but there were periods when following a page through the brain fog was beyond me.

Making the target smaller wouldn’t have solved much.

One page still required reading.

One sentence in a journal still required working out what the sentence was.

This matters because “small” is usually defined by the person giving the instruction.

Write for five minutes.

Walk around the block.

Cook something simple.

Send one text.

On paper, each task has been successfully reduced, but the person actually doing it still has to pay the entry fee, and when everyone describes these things as easy, failing to do them can produce its own little verdict.

Other people can do this.

This is supposed to help.

I know what I’m meant to do.

I still can’t do it.

Eventually the advice becomes one more place to feel behind.

The things that survived

Some things were still available to me.

I could watch comfort shows and films I’d already seen thousands of times. They didn’t ask me to keep up. I knew what was going to happen. Losing concentration for a while didn’t matter because I’d probably seen the same scene ten times already.

I don’t know that watching familiar television was doing anything particularly therapeutic; it was simply something I could still do.

That distinction has become more interesting to me than whether an activity looks sufficiently like recovery from the outside.

Reading was good for me in all sorts of ways. At 4%, it wasn’t available, but a familiar film was.

Showering survived too, though less frequently. There was no impressive habit system behind it - Sometimes I could shower, and sometimes I couldn’t. The frequency dropped with everything else.

Looking back, that gives me one version of low-power mode.

Some things can be shrunk.

The full routine disappears. A smaller version remains within reach.

Other things need a substitute.

Reading demanded concentration I didn’t have, and Familiar films asked for much less of it, so i substituted. Neither required pretending I could access a version of myself who had temporarily gone offline.

And then there were things I couldn’t reduce far enough.

Sometimes the food has to appear

At my lowest, eating without somebody putting food in front of me could be difficult.

This is where “make it smaller” reaches its limit.

You can simplify a meal. You can remove choices. You can buy easier food.

There is still a point where somebody has to make something happen.

For me, sometimes that somebody had to be someone else.

The workable version of “remember to eat” was food appearing in front of me.

Most of the decisions had already been made.

Most of the work had already happened.

All I had to do was eat.

That seems embarrassingly obvious written down. It also changes the question.

A lot of self-help asks:

How can I make this easier for myself?

At 4%, another question becomes available:

Can somebody else carry part of this?

My partner and kids gave me space. They allowed me to rest. My workload was reduced.

Nobody handed me a better productivity method for completing the same amount of work while severely depressed.

There was less work.

That helped.

This is the part I think a lot of advice misses because advice is usually addressed to the individual.

Here’s what you can do.

Here’s the habit you can build.

Here’s what you can track, change, practise, challenge, cook, schedule or remember.

Eventually the person with the least capacity in the room has somehow been promoted to project manager of their own recovery.

Low-power mode

Phones already have an answer for scarce resources.

Low-power mode.

The basic idea is boring and sensible: when there isn’t much left, you stop behaving as though there is.

I think mental-health advice could use more of that logic.

Looking back at what was possible for me, low-power mode took several forms.

Showering could be shrunk. Less often was still more realistic than maintaining the normal routine.

Reading needed a substitute. Familiar films and shows survived because they demanded less concentration.

Eating sometimes had to be transferred. Somebody else did the work that happened before the food reached me.

And some demands were simply removed. My workload got smaller. My family gave me space. Rest didn’t arrive with another assignment attached.

I wouldn’t turn those four words into a new daily tracker.

The last thing somebody at 4% needs is an app asking them to calculate whether they are currently at 3.7%.

They’re questions, if they’re useful at all.

Can this be smaller?

Is there something else I can tolerate?

Can somebody else carry part of it?

Can this wait?

Sometimes the answer to all four will be no.

Journaling sat outside my available options for a while. Trying to rescue it with a two-minute version would have missed the problem.

Some useful things belong to days with more capacity.

A recovery toolkit can contain things that work at 40%, 60% or 80%. They don’t become bad ideas because you can’t access them on your worst Tuesday.

They just aren’t Tuesday’s job.

Spending the last 4%

One response to my original Note pointed out another problem.

Even when you can do something, you don’t know whether spending the energy will pay off.

Someone tells you exercise helped them.

Someone else swears by journaling.

Another person has a morning routine.

Another has breathwork, cold showers, supplements, meditation, gratitude, digital detoxes and a course available for twelve easy payments.

You have 4%.

Choose wisely.

That’s a strange amount of decision-making to hand somebody whose capacity for making decisions may already be shot.

And the benefits aren’t always immediate.

You can drag yourself through the thing someone promised would help and feel exactly as terrible afterward.

Maybe it helps eventually.

Maybe it doesn’t.

Either way, you’ve spent something.

I think that’s why the question of cost matters as much as the recommendation itself.

How much energy does this require before it gives anything back?

How much concentration?

How many decisions?

How much money?

How much social capacity?

And how much of that cost can disappear?

Because the things I remember being possible at my lowest were rarely impressive.

A film I’d already watched to death.

A shower when I could manage one.

Food somebody else had put in front of me.

Less work.

Space.

Rest.

Nobody looking at that list would mistake it for an optimised recovery routine.

That may be exactly why it worked at 4%.

When I couldn’t make myself eat, “eat properly” was useless as an instruction.

A plate of food in front of me asked much less.

Sometimes low-power mode is simply one less thing waiting for you to do it.

For paid subscribers: I’ve turned this idea into a small Low-Power Mode Kit for working out what can be shrunk, substituted, handed over or dropped when your usual coping tools ask too much, plus a companion page for the people trying to help without accidentally giving you another job.

Leave a comment

Read more

There’s a number printed on the side of the box. Repeats: 5. I have never once, in years of picking these up, thought about what happens when it reaches zero.

The pharmacist knows my name. The box and its contents live in the cupboard above the stove. I take them first thing (most days) with my cup of tea or a glass of water, which is to say I don’t take them at any particular time so much as I arrive at the moment and find it already happening. My hands know the routine better than I do.

Supply feels like physics. Water comes out of the tap. The tablet is in the packet. It’s been there every day for so long that it stopped being a belief and became a certainty.

Bradley Buswell

In 2021, a 42-year-old man in Western Australia started calling his friends and family to say goodbye. He sold his belongings. He believed he was dying of a serious physical illness.

His family found out he had schizophrenia. Years of sleepless nights and preoccupation with his health suddenly made sense.

He was admitted to the psychiatric ward at Bunbury Regional Hospital under WA’s Mental Health Act. His family later told the coroner what that ward was like: “often chaotic, overstimulating, and populated by patients in acute distress.” For Bradley, they said, who was gentle and quiet and already overwhelmed by his symptoms, it was traumatising rather than therapeutic.

He got worse on the ward. He was eventually released on a community treatment order, which legally required him to live with his grandmother, attend his appointments, and follow his treatment plan.

His psychiatrist, Dr Daniel de Klerk, started him on intramuscular olanzapine. An antipsychotic, injected once a month.

Dr de Klerk told the inquest why he chose the injection: “The choice to administer intramuscularly was a result of uncertainty about whether he was going to take oral medications.”

It worked. His schizophrenia was labelled treatment-resistant, but the monthly injection managed his symptoms. “His symptoms never completely resolved but he became much less distressed,” Dr de Klerk said.

His family have asked that he be remembered for his gentle nature, his love of family and friends, the outdoors, and the Fremantle Dockers. Not for the illness.

Bradley Buswell died in March 2025, in state care, eight months after he was taken off that injection. A coronial inquest heard evidence in July. The coroner has not yet determined what caused or contributed to his death, and will hand down findings by September.

Two words that put the failure in the patient

Compliance, in psychiatry, means whether the patient did what the treatment plan said.

Did you take it?

Did you turn up?

Did you follow the order?

Did you tell someone when things changed?

It gets recorded and it follows you between services. It shapes how the next clinician reads you before you’ve opened your mouth, and depending on where you are and what order you’re under, it can change your legal status.

Non-compliance isn’t filed as a logistics problem. It’s filed as a fact about your character. Insight lost and trust withdrawn.

Then there’s the other term; Bradley Buswell’s schizophrenia was labelled treatment-resistant.

The illness resists the treatment. The treatment doesn’t fail; it gets refused, down at the level of biology. Even the phrase for medicine not working puts the resistance inside the person.

Two words, both pointing the same direction. Whatever goes wrong, the sentence finds its subject in him.

I want to be careful here, because the alternative to this vocabulary isn’t no vocabulary. Adherence genuinely matters. Treatment-resistant is a real clinical category with real prescribing consequences behind it, and Dr de Klerk was making reasonable decisions with the drugs he could actually get. His evidence at the inquest reads as thoughtful. Bradley’s family said so themselves, and thanked his treating team.

The problem isn’t the clinicians. It’s that the language only conjugates one way.

What the shortage got called

In 2024, intramuscular olanzapine became abruptly unavailable in Australia. The medication had run out of contract. The overseas manufacturer wasn’t making it anymore.

It “became” unavailable.

It had run out of contract.

Nobody withdrew it.

It occurred, the way weather occurs.

WA Country Health Service’s acting executive director, Jeff Calver, expressed deep sorrow for the family’s loss and said he understood there were no shortcomings in the care provided. He added: “mental health challenges are complex, and we know a patient’s condition can change rapidly.”

Which is true. It’s also a sentence that locates the change in the patient.

And here’s the honest complication, which is worse than an accusation. The responsibility really is distributed. An overseas manufacturer, a supply contract, a regulator, a state health service, a prescriber, a pharmacy. Each holds a piece and none of them holds the outcome. The language slides into the passive voice because there’s no single subject available to put in the sentence.

Meanwhile the patient is always available. He’s the one subject the system can reliably name.

The patient fails morally. The system encounters circumstances.

The order and the supply

A community treatment order lets the state require someone to accept treatment without their consent, and act on it if they don’t. Bradley Buswell was on one. Live with your grandmother. Attend your appointments. Follow the plan.

Nothing in that order obliged anyone to be able to supply the treatment it compelled.

So when the injection vanished, the compulsion carried on and the thing being compelled changed underneath it. He was put on oral olanzapine instead. His mental health suffered. “It was not very long before we had reports from the family that he was saying goodbye again,” Dr de Klerk told the inquest.

Now look at what the shortage actually removed.

The injection existed because compliance was uncertain. That was the stated reason for choosing it. It was the system’s own answer to the exact problem the system worries about most, and it was working.

Take the injection away and you hand the problem back to the patient. You give a man a daily task that a psychiatrist had already judged him unlikely to manage, and you give it to him not because anyone decided it was better for him, but because a contract lapsed on the other side of the world.

He was then prescribed a different antipsychotic, also by injection, which Dr de Klerk said was less effective. The reason given was so his treating team would know he was taking something.

Not so it would work. So it could be seen.

I don’t say that as an attack. It’s a rational response to a bad situation. It’s also, precisely, what this whole essay is about. When supply failed, the priority that survived was visibility of compliance.

Bradley kept telling them. He admitted to his treating team, more than once, that he hadn’t taken his tablets for days. At his final review, days before he died, he said it again.

He was doing the disclosure the framework asks for.

Dr de Klerk’s answer, at the inquest, was this: “It doesn’t work from the bedside table, unfortunately, you have to swallow the tablets.”

He was found dead in his room days later, with 52 times the prescribed amount of olanzapine in his system. His family have said they were never told he’d been prescribed the oral olanzapine at all. They were told only about the monthly injection. Had they known, they told the inquest, they could have monitored his access.

They believe his death was accidental. They believe he was failed by a system rather than by any person in it. The coroner hasn’t ruled.

What ran out

I want to be clear that I’m not comparing my situation to his. I’ve never had schizophrenia, never been on a community treatment order, never had a long-acting injection. I take tablets, by choice, and I swallow them.

But I know what a supply running out feels like from the inside of an illness, because mine ran out once.

I’d come out of hospital after three months and a course of ECT. I was on lithium. At some point in the weeks after I got home I worked out that the tablets I had were going to run out before my next psychiatric appointment, which was a fortnight away.

I didn’t call to ask for an earlier one.

I’ve thought about that a lot. The straightforward reading is that I should have picked up the phone. And I should have. But I’d been out of hospital for a matter of weeks, I’d had electricity run through my brain, and I could not reliably do the dishes. The idea that I would advocate for myself inside a health system, that I would chase and negotiate and hold a line on the phone, assumes a version of me that hadn’t come back yet.

So I rationed them. Cut what I had to stretch it further, and made the supply last longer than it was meant to.

Here’s what I want you to notice about that. Rationing your psychiatric medication is, in the record, non-compliance. Not taking the prescribed dose. Deviating from the plan. In a file, that’s what it is.

In reality it was the single most conscientious thing I could do with a supply I couldn’t replace. It was stewardship. It was me trying, badly, to solve a problem that had been handed to me by a scheduling decision I had no part in.

I don’t actually know what was written down about it. That’s the part that stays with me. I’ve never seen the file. Compliance gets recorded about you and travels ahead of you, and mostly you find out what it says by how the next person in the room treats you.

I went back to hospital before the script was renewed.

And the question I’ve never been able to put down is this: somebody wrote that prescription, and somebody booked that appointment, and the gap between them was arithmetic. Count the tablets. Count the days. Anyone could have done it.

I didn’t. I was ill.

They didn’t. They were busy.

Only one of those has a word for it.

Reciprocity

If compliance is a fair thing to ask of a patient, it’s a fair thing to ask of the system doing the asking.

That could be concrete. Supply continuity recorded and auditable the way patient adherence is. A named person who holds responsibility when a compelled medication becomes unavailable, rather than a contract that lapses into nobody’s hands. Families of people on treatment orders told which drugs their person has actually been prescribed, which the Buswell family says they weren’t.

None of that is radical. It’s the same standard, applied in both directions.

There’s a number printed on the side of my box. Repeats: 5. I know now what happens when it reaches zero, and it isn’t that the tablets stop working.

It doesn’t work from the bedside table. You have to swallow the tablets.

That’s true of a patient. It’s also true of a medicine that nobody is manufacturing.

If you or anyone you know needs help:

Lifeline 13 11 14
Suicide Call Back Service 1300 659 467
Beyond Blue 1300 224 636
13YARN 13 92 76
MensLine Australia 1300 789 978

Source: Neve Brissenden, “Bradley died months after national antipsychotic medicine shortages, coronial inquest hears,” ABC South West WA, 22 July 2026.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

Content warning: depression, hospitalisation.

The ad came back, and this time the phone explained itself

I went to save it, and the menu offered me something I didn’t know was there.

Under the save button, a line about why I was seeing this ad. I’d tapped that before. Twice, in fact, and both times it told me the same two things and then stopped: Australia, and 18 or over. A country and an age. That’s the whole audience, apparently, and I’m in it, along with about 20 million other people.

This time there was more underneath. A heading about my activity, and a sentence I’ve now read enough times to have memorised.

“What you interact with helps us predict what ads you might like.”

Then a list of the things I apparently interact with. Broad categories, most of them unremarkable. One of them was health and wellbeing.

Sitting above all of this, still playing, was a young woman with a name badge. The badge says The Church of Jesus Christ of Latter-day Saints. Her first words are about depression.

I spent 7 months in a psychiatric hospital for the thing she names in her opening line. I write about it publicly, every week, on Substack, and every one of those posts goes out as a public post on my Facebook profile where people I went to school with argue about them in the comments. So the ad found a man who has done everything a person can do to make his depression a matter of public record.

That doesn’t tell me why it found me. It tells me why the question is worth 6,000 words.

Here’s what I can say with certainty, and it’s less than I want. Two screens told me who was allowed to see this advertisement. A third told me what Meta had been paying attention to.

The first one still matters

Two weeks ago I finished writing an essay about comparing yourself to other people’s social media presence. Then I opened Facebook and started scrolling reels, which is the kind of thing that would be funny if I’d done it on purpose.

I was flat. Not bad, not in any trouble. Flat, and wondering what I was going to write about next.

I scrolled past her. Then I went back, fast, because I’d caught the page name and the badge on the way past and my brain had already flagged it as something.

Sound was on. I heard her say it.

“If you’re struggling with depression and not feeling the comfort and peace that you need, I invite you to meet with missionaries from The Church of Jesus Christ of Latter-day Saints.”

Twenty-five seconds. I’ve watched it enough times now to have the shape of it memorised, and the shape is what I want to talk about, because the shape is doing more work than any single sentence in it.

It goes like this. A clinical condition, named in the first four words. Then an emotional shortage, which is where depression gets translated into a lack of comfort and peace. Then the people who can fix the shortage, who are missionaries. Then the thing they’ll give you, which is the Gospel of Jesus Christ, described as a message of love, hope and comfort. Then the button. Sign up.

Six steps from a diagnosis to a form.

Look at the verbs, because the verbs are careful. “We can help you learn the Gospel of Jesus Christ.” That’s the only promise in the ad, and it’s a promise about learning. Nobody says they can help with the depression. Nobody says anything will improve. The end card says “Feel hope and comfort in Jesus Christ,” which is an invitation rather than a claim, and you could put that sentence on a church noticeboard without anyone blinking.

So the ad names a psychiatric illness and then promises to teach you something. In between those two things there’s a gap, and the gap is where the word “help” sits, doing a job it hasn’t been asked to do out loud.

There’s a second thing in that opening sentence I didn’t notice for a week, and it’s the pronoun.

“If you’re struggling with depression.” Not people who are struggling. You. She’s talking to a specific person, and the specific person is whoever the phone has decided is watching. The conditional “if” is what makes it survivable, because “if” means she’s covered either way. If you’re not depressed, she’s addressed a hypothetical. If you are, she’s addressed you.

Meta’s own advertising rules prohibit ads that assert or imply a viewer’s personal attributes, and the policy names mental health conditions directly, with depression given as an example. I’m not in a position to say this creative breaches that rule. I don’t know what Meta’s reviewers saw, whether anyone complained, or how the policy gets applied to a conditional. Those decisions aren’t public and I’d be inventing a verdict I can’t support.

I’ll say the thing I can support instead. A rule exists against implying that the person watching has depression, and this advertisement opens by raising the possibility that the person watching has depression, and it ran to me in Australia for at least 2 weeks.

I want to be fair here, because the fair version of this is stronger than the angry one.

A church is allowed to advertise. A church is allowed to say it offers comfort, and to hope you’ll come and find out. Millions of people have found real support in religious community, and the research on that is genuinely mixed rather than dismissive. My objection isn’t to a church buying advertising, and it isn’t to a young woman on a mission believing every word she says. I have no reason to doubt she believes it. She probably does.

My objection is narrower than any of that. Of all the openings available, the campaign chose a named psychiatric condition as the door.

Depression is the doorway. Everything past the doorway is careful enough that there’s nothing to argue with, which is exactly what makes the doorway worth arguing about.

Two ads, one button

Today there was a second one.

Different missionary, different opening. “Listen, whatever you’re going through, we would love to sit down with you and teach you a message that can help you find strength to overcome it.” Then the same instruction: click the link below to have missionaries from the Church come and visit you. The end caption says “We’re here for you.”

Same verified account. Same distress-to-contact movement. Same Sign up button, and the same form design on the other side of it.

The difference between the two is a single design decision. The first ad names the illness. The second one names nothing at all, and lets you fill in the blank yourself. “Whatever you’re going through” covers a bereavement, a divorce, a redundancy, and a psychiatric relapse with equal comfort, and it does it without ever putting a clinical word on screen.

I have to be careful about what that similarity proves. I can see two creatives from one advertiser. I can see that both use a Sign up button and both open a form asking for the same fields. What I can’t see is a campaign ID, an ad set, or a form ID. Two ads with the same form design might be one campaign testing two hooks, or two campaigns that happen to share a template, and from where I’m standing those look identical.

So I’ll say the small true thing. One hook names the illness, the other names anything at all, and both of them end at the same button.

Australia, 18 or over

The advertiser’s own screen has never said anything about depression.

It’s headed “Why you saw this ad,” and the section that matters is called Advertiser choices. Meta introduces it by explaining that advertisers want to connect with people they think will enjoy their products, so they make choices about who to show their ads to.

Then the specifics, and there are two of them.

“Come Unto Christ wants to show ads to people who may have: Set their age to 18 and older. A primary location in Australia.”

That’s the entire disclosed audience. Nothing about mental health. Nothing about interests, behaviours, or any audience the Church might have built or bought. An age and a country.

For a while I read that as the end of the story. It isn’t, and understanding why took me longer than it should have.

Those two settings are boundary conditions. They describe who is eligible to see the ad, which in this case is essentially every adult in the country. They don’t describe who actually sees it. Between eligibility and delivery there’s a second decision, and that decision isn’t made by the advertiser at all.

Meta’s own product documentation is open about this. Its automated audience system is described as using AI to find the most relevant audience for a campaign, and the page setting it out is specific about the shape of the thing: minimum age, location and language are strict criteria the system won’t expand past. Everything inside those criteria is the model’s to decide.

The same documentation describes what happens if an advertiser does supply audience suggestions. The system prioritises people matching that profile first, then searches more widely anyway.

I can’t confirm which of these products this campaign used, or whether it used any of them, and I’m not going to pretend otherwise. What the documentation establishes is that the architecture exists, that broad settings are compatible with narrow delivery, and that on this platform a wide audience setting is a description of the fence rather than the search.

Something worth sitting with: on these systems, a wide audience is often the recommended setting rather than the lazy one. The more room the model has, the better it performs against whatever it’s been asked to optimise for. An advertiser who selects the whole adult population of a country hasn’t declined to target anyone. They’ve delegated the targeting.

I don’t know that’s what happened here. I know it’s what the product is built to do, and I know the visible settings are consistent with it.

Australia and 18 or over is the perimeter of the room. It tells me the door was open to 20 million people. It tells me nothing about who got walked to the front.

A broad boundary doesn’t mean random delivery.

Health and wellbeing

Which brings me back to the third screen, the one I found by accident while trying to save a video.

The screen is called Your activity. The sentence underneath is Meta’s, and I’m quoting it exactly: “What you interact with helps us predict what ads you might like.”

Then a subheading, Your activity on Meta technologies, and two rows.

The first: “You interacted with ads about sports, ads and marketing and puzzles and games.”

The second: “You interacted with posts, stories, reels and Pages about books & literature, health and wellbeing and business solutions.”

There it is. Third topic in the second row, sitting between books and business solutions like an item on a shopping list.

This is the strongest evidence I have, and I want to be precise about why, because there’s a version of this paragraph that overstates it and that version is worthless.

Everything I’d found before this was general. Meta’s Help Centre says it may use the Pages you like, the ads you click, the content you create or interact with, your check-ins, and activity from outside Meta’s own apps. That’s a description of what a platform is capable of. Every platform publishes something like it, and quoting it proves nothing about my ad.

This screen is different because of where it sits. It lives inside the explanation for one specific advertisement, the depression one, and there it names health and wellbeing among the activity categories behind the prediction. Meta is telling me, about this ad, that health-adjacent activity was part of how it decided.

That’s a real finding, and it’s the first one in this whole exercise that came from the platform rather than from me guessing.

Now the limits, and there are a lot of them.

The screen doesn’t identify a single post. It doesn’t say how heavily health and wellbeing weighed against books and literature, or against business solutions, or against the row about puzzles and games. Three topics in one sentence, no ranking between them.

And it carries Meta’s own caution, printed directly underneath: “Some of your activities may not be directly related to this ad. In these cases, we’ve made a prediction on which ads you may like based on others with similar activity.”

Read that carefully, because it’s doing a lot of work. Meta is saying that some of what it has shown me might have nothing to do with this ad, and that the prediction may have come from other people rather than from me.

The gap that bothers me most is between creating and consuming, because the screen collapses them into one word. Interact.

I write about depression. I also read about it, constantly, because I’m halfway through a Certificate IV in Mental Health and because it’s the subject I’ve organised my working life around. Those are different activities with different meanings, and a category that contains both can’t distinguish a person in distress from a person studying distress. The screen counts them the same.

Which means the honest reading is worse for my argument than I’d like. My public essays about my own hospitalisation might have contributed nothing. A fortnight of coursework reading might have done all the work.

That last one deserves more attention than it usually gets. Similarity-based prediction means the system can arrive at me without ever forming a view about me. It can notice that people who behave the way I behave respond to this ad, and serve it, and nothing in that process requires anything as specific as an opinion about my mental health.

So no, Meta has not written the word depressed next to my name. I have no evidence of that and I’m not claiming it.

What I have is a category wide enough for Meta to deny any interest in my diagnosis, and close enough to it that a depression advertisement arrived anyway. Health and wellbeing covers a man writing about a psychiatric hospitalisation. It also covers a man following a running app. The category can’t tell you which one I am, and the ad found me regardless.

I can’t decide whether that’s reassuring or worse.

The evidence has my fingerprints on it

There’s a problem with everything I’ve written above, and it’s mine.

The screenshot that proves health-adjacent personalisation was taken today. By today I had already seen the depression ad on Facebook a number of times, and I had already tapped Sign up on it, roughly a week after the first impression. I’d opened the form. I’d tapped through the advertiser explanation twice.

Meta’s own documentation lists ads clicked as one of the things that shapes which ads you see next. The screen I photographed today even includes a category for ads I’ve previously interacted with.

So the evidence for personalised delivery was collected after I spent 2 weeks personally demonstrating an interest in this exact advertisement.

I can’t get around that, and I’m not going to write as though I can. Everything the third screen establishes, it establishes about a later impression, in a feed I had already influenced. It says something true about how I’m being served ads now. It says much less about the first time, 2 weeks ago, before I’d clicked anything at all, when a depression ad arrived unprompted on a flat afternoon at the end of an essay about social media.

That first impression is the one I actually want explained, and it’s the one I’ve made permanently harder to explain.

The obvious response is to run it again properly. Wait for a clean impression, don’t touch it, document the conditions. I can’t. There’s no version of my feed left that hasn’t been taught something by the last fortnight, and there’s no reset button that clears what the system has learned without also clearing the account. Every future missionary ad that arrives is now evidence of my own behaviour as much as anything else.

For me this is the part that’s hardest to write around, and I’d rather say so than build the essay on a screenshot I know is contaminated.

The moment I clicked to find out why the ad had found me, I handed the system one more thing to learn from.

The form already knew my name

Roughly a week after the first impression, I tapped Sign up.

I’d like to report that this was a research decision. By then I’d been circling the thing for days and the circling had stopped being useful.

The form opened inside the app. I never left the feed. Across the top, a progress bar in three segments. The heading said “Share your info,” and underneath it my name was already sitting in the first field.

I hadn’t typed it. Meta had, because Meta has it, and the design of an instant form is built to remove the small window of effort in which a person might reconsider.

Continue.

“Hi Jacob, Need spiritual help? Meet some missionaries.”

Underneath, an explanation. Missionaries teach people God’s plan, help strengthen faith, and answer life’s biggest questions. Interested? Let’s get you started.

Sit with the phrase in the middle of that screen for a second. Need spiritual help?

The advertisement that brought me here says depression. Two screens later the same funnel calls it spiritual help, and nobody announces the swap.

Probably not a trick. These are the words the Church has for this, and the advertisement borrowed a clinical one to get the tap. The clinical word doesn’t survive past the tap.

Continue again, and the third screen asks for the rest. Full name. Street address. City. Country. Phone number. Some of it pre-filled again, some of it blank, all of it one submission away from a young woman I’d never met knowing where I live.

I want to be exact about what that form is asking for, because it’s easy to describe it as sinister and harder to describe it as it actually is. Nothing on the page is unreasonable for its stated purpose. If missionaries are going to visit you, they need to know where you are. A street address is the minimum viable answer to “come and visit you,” and there’s no dishonest version of that field.

What sits oddly is the pairing. The illness gets named in the advertisement. The address gets collected in the form. In between there’s no step where anyone asks what’s actually going on with you, whether you’re safe, or whether you have a doctor.

I stopped there. I didn’t submit it.

That matters more than it sounds, and it’s the single biggest limit on everything I can say in this essay. Meta’s own explainer states that your contact information is shared with an advertiser only if you submit the form. My recording ends before submission. So the Church has not received my name, my address or my phone number through this interaction, and I have no basis to suggest otherwise. Anyone reading this as an account of my data being handed over has read it wrong.

What I saw was a system with everything loaded and nothing fired. The form had removed almost all the friction between a depressed person watching a video and a stranger arriving at their door, and the only remaining piece of that machine was my thumb.

“Don’t worry, we will never share or sell your information”

That sentence is on the contact page, underneath the fields, in small text. I’ve quoted it exactly.

Start with the first 2 words, because they’re the most interesting part and nobody would notice them in the ordinary course of filling in a form. “Don’t worry” answers a worry you hadn’t got around to having yet. It arrives slightly before the concern it’s addressing, which is a strange place for reassurance to stand.

Then the promise. Never share. Never sell. Both categorical, no qualifiers, no link required to understand it. An ordinary person reads that sentence and concludes their details are going to one place and staying there.

Now the Church’s published privacy notice, updated on the site in April 2026. It says the Church processes contact details for missionary and teaching purposes. It says it uses and analyses that data. It says it shares personal data with third-party processors. It says data may be transferred to Church entities for Church purposes, and that it may be stored in the United States.

Those 2 documents use the same ordinary word in opposite directions.

There’s a technical answer available, and I think it deserves stating properly rather than being waved past. In data protection language, a processor is a contractor handling data on your behalf and under your instruction, which is a different thing from selling a list to a stranger. Transfers between entities of the same organisation are likewise not what most people mean by “sharing.” A lawyer could construct a defence in which every word of the privacy notice and every word of the form are simultaneously true, and it might well hold up.

The defence has a problem, and the problem is the reader. Nobody filling in that form is applying the technical definition. They’re applying the English one, having watched a video about depression, with their address already typed in.

There’s a detail here I didn’t know when I started, and it cuts both ways.

Meta requires a privacy policy link on every lead form. Not as a recommendation. An advertiser can’t run a lead ad without supplying one, and Meta’s guidance says the linked policy has to explain how the lead data is collected and used.

Which means the missing document isn’t missing. Somewhere attached to that form is a privacy policy the Church has nominated as governing what happens to my details, and it exists because Meta made it a condition of running the ad at all.

I haven’t yet confirmed which document that is. The Church publishes a global privacy notice and a separate Australian one, and working out which is linked to this exact form is a piece of work I need to finish before I’d stand fully behind this section. So treat what I’ve written as a comparison between the form’s promise and the Church’s most prominent public notice, rather than a proven contradiction between the form and its governing terms.

The part that survives either way is smaller and harder to shake. Somewhere behind that form sits a document explaining the real data practice, and directly above it sits a sentence beginning “Don’t worry.” Only one of those is written to be read.

With that caveat carried openly: a categorical promise never to share appears difficult to reconcile in plain English with a notice that describes sharing and transfers in some detail. If both are accurate, the form is relying on a distinction it never explains to the person being asked to trust it.

What a submitted form becomes

I didn’t submit, so everything in this section is reconstructed from public documents rather than from anything that happened to me. It’s the Church’s own material and Meta’s own material, and it’s the clearest picture available of where that button goes.

The Church’s published training for missionaries has a name for people who make contact after seeing an advertisement on Facebook or Google. They’re called Media Referrals. The training says to make contact within 24 hours where possible, suggests text as the best method, and covers updating something called the Area Book Planner, which is the record-keeping system missionaries use to track the people they’re working with.

That 24 hour window is worth pausing on. It’s ordinary sales practice, and it’s also the interval between a person in distress tapping a button and a stranger contacting them, before anything about the impulse has had time to settle.

So on the institutional side, a person who fills in that form becomes an entry. That’s true of every organisation that has ever run a contact form, and there’s nothing sinister in the existence of a database.

The training material also contains something I think should be quoted rather than skated past, because it complicates the case I’m building. Missionaries are warned not to mislead people by turning up and substituting a different offer from the one that brought them there. Somebody at some point wrote down that you shouldn’t bait and switch a person who has asked for help. That’s a real safeguard, published openly, and any honest account of this funnel has to include it.

On the platform side there’s a second thing happening at the same moment. Meta’s developer documentation describes how advertisers retrieve lead data, and it’s specific about what comes back. A single request returns the time the lead was created, the contact fields the person filled in, and identifiers for the form and the advertisement that produced them. The documentation names ad_id and campaign_id directly.

I want to be careful here, because this is where I could overreach and the overreach would be tempting. The documentation establishes that a submitted lead can technically be linked back to the creative that generated it. It does not establish that this campaign retrieves those fields, that the Church stores them, or that any missionary ever sees which video brought you in. That integration is invisible to me and I’m not going to guess at it.

What the public record supports is narrower and still worth sitting with. A form submitted from a depression advertisement can exist in 2 places at once. On one side, a name and an address for somebody to visit. On the other, a measurable outcome attributed to a particular piece of creative, feeding back into a system that learns what works.

The person is the referral. The person is also the result.

What health and wellbeing establishes, and what it doesn’t

Here’s the whole case, stated as plainly as I can manage.

A religious organisation ran a paid advertisement that opens by naming a psychiatric illness and ends at a form requesting a home address. It ran a second advertisement with a broader hook into the same form design. Its visible audience controls were a country and an adult age limit, and nothing else. Meta’s explanation for that first advertisement named my interaction with health and wellbeing content as one input into predicting I might like it. The form promised categorically that my information would never be shared, next to a public notice describing sharing in some detail. I stopped before submitting.

That’s it. That’s everything I can prove, and I’ve spent 2 weeks establishing it.

Now the things I can’t prove, because the essay is worthless without them.

I can’t prove Meta inferred that I have depression. Health and wellbeing is a category, not a diagnosis, and the same category contains people training for marathons. I can’t prove any particular essay of mine was read, analysed or weighted by anything. I can’t prove the Church selected a health-related audience, and the visible evidence points away from it. I can’t prove the 2 creatives shared a campaign. I can’t prove anyone broke any law, and I’ve deliberately not written a sentence that implies I can.

I also can’t prove the thing I most wanted to prove, which is why the first ad arrived at all, before I’d clicked anything, on an ordinary flat afternoon.

Then there’s the counterargument, and it’s stronger than the shape of this essay would lead you to expect.

The Church publishes its own material on depression, and it’s better than most of what circulates online. An article in its own magazine separates ordinary sadness from major depressive disorder and says the second is a mental illness requiring professional treatment and medication. Another, by one of its senior leaders, tells members that aberrant thoughts and feelings come from disease rather than transgression, and sends them to skilled health care providers. A third walks readers through the difference between a psychiatrist, a psychologist and a counsellor, and how to pick one.

Read that second one properly. A church elder, writing in an official Church magazine, telling members their mental illness is not a moral failure. That’s a better sentence than a lot of what gets said about depression in secular settings, and it was published in 2005.

So this is an institution that has formally, repeatedly and in public told its members to see a doctor.

None of that is buried, and I’m not producing it as a disclaimer. It’s the strongest thing anyone could say in the Church’s defence, and it means the campaign I’ve been describing sits at odds with the organisation’s own published position on the illness it’s advertising against.

I’d add something else, because the alternative is condescension. People find real support in religious community. The research on religion and mental health is genuinely mixed rather than damning, and negative religious coping and positive religious coping produce measurably different outcomes. A person who is depressed and who responds to that advertisement isn’t a dupe, and hasn’t lost the capacity to make a decision about their own life. Depression takes a great deal. It doesn’t take that.

So what’s left, once I’ve given away everything I can’t prove?

A gap, mostly, though the ground moved under it last month and I nearly missed it.

Meta’s advertising rules prohibit implying that a viewer has a mental health condition, with depression given as an example, and this creative appears difficult to reconcile with that rule, though whether Meta’s reviewers agree is not public.

Then there’s the Australian privacy position, which changed on 24 June. The Privacy Commissioner published two determinations against Medmate Australia and Monash IVF, both health providers, finding they’d interfered with people’s privacy by collecting health information through third-party tracking pixels on their websites and using it to target those people with advertising on social media. No consent, no notice. The Commissioner found contraventions of three separate Australian Privacy Principles, covering collection, notification and direct marketing. A companion inspection of 50 health provider websites found 96% running tracking technologies of some kind.

The core finding is the part that matters here. Watching what somebody does in a health context, and then using it to decide what advertising reaches them, can amount to collecting sensitive information that requires their consent.

I want to be careful, because my situation and theirs are not the same and I’d be overreaching to pretend otherwise. Those were health service providers, placing pixels on their own clinical websites, tracking their own patients. Nobody put a pixel on me. The Church isn’t a health provider. And the personalisation I’ve documented happened inside a platform, from activity categories, rather than through a clinic handing data to an advertiser. A determination about a fertility clinic doesn’t decide anything about a missionary advertisement.

What it does is remove an argument I expected to have to fight. Six weeks ago, anyone writing this essay would have been told that health-adjacent digital activity used for ad targeting is an interesting theoretical concern with no Australian authority behind it. That’s no longer where the regulator stands.

Whether the Privacy Act reaches this particular ad still depends on which entity is collecting, whether a category as broad as health and wellbeing amounts to health information, and several other things the screenshot on my phone cannot answer.

The rest of the map is thinner. Consumer law is a stretch, because a promise of hope and comfort isn’t the kind of claim it was built to test, though the categorical promise about my information is a good deal more testable than the spiritual one. The advertising code has a clause about content that encourages behaviour contrary to community standards on health and safety, and applying it to an advertisement that never once discourages treatment would be novel. The therapeutic goods rules don’t apply, because nothing therapeutic is being sold. The health practitioner advertising rules don’t apply, because nobody here is claiming to be a clinician.

That last pair is the whole problem in miniature. The campaign is clinical enough to name the illness in its opening line, and spiritual enough that not one regulator built to police clinical claims has jurisdiction over it.

I’ve been thinking about which of the 3 screens actually told me something.

The first said Australia. The second said 18 or over. The third said health and wellbeing, hedged immediately with a note that some of it might not be related to the ad at all.

Three explanations, each one more specific than the last, and the most specific of them describes about half the country.

Meanwhile a young woman I’ve never met opened by naming my illness in 4 words and got it right.

I still haven’t saved the video.

What’s coming next

This is the first of three.

I’ve spent this essay on what I can see, which is a screen, a form and a set of documents. The next one goes further back, because none of this is new. Long before anyone could buy a distressed audience, there was a fight over who gets to say what suffering is and what the sufferer must do about it. Priests, physicians, asylum keepers, Quakers who thought madness deserved gentleness, psychiatrists who thought it deserved restraint. That history isn’t a story about superstition losing to science, and I’d rather not write the version where it is.

The third one puts the two together and tries to draw a line. Not between religion and medicine, and not between belief and unbelief, but between offering someone support and using their condition to acquire them. There’s a difference. I think it can be tested. Seven questions, applied to this campaign and to anything else that comes looking for you on a bad day.

Part 2 lands in a week or so. It starts with epilepsy, in Greece, about 2,400 years ago, with a doctor calling out a racket.

Where all of this came from

I’ve written an essay accusing a large institution of using a psychiatric illness as an advertising hook, and a larger one of helping it find me. Anyone reading that is entitled to check it.

So here’s everything I used, what each document actually is, and what it can and can’t carry. A few of them cut against my argument, and those are in the list too. Where a page is live and editable, I’ve archived a dated copy, because a source that quietly changes underneath a published essay is worse than no source at all.

Meta’s own documentation

• Privacy Violations and Personal Attributes, Meta Transparency Center. Meta’s rule against advertising that asserts or implies a viewer’s personal attributes, with mental health conditions including depression given as examples. This is the source of the sharpest criticism in the essay, and also of its limits: there’s no public enforcement decision here, so the most anyone can say is that the creative appears difficult to reconcile with the rule.

• About Facebook Ads, Meta Help Centre. The general explanation of what feeds ad delivery and where the “Why am I seeing this ad?” tool lives, covering Pages liked, ads clicked, content created or interacted with, and activity from outside Meta’s apps. Useful mostly as background now, since the ad-specific screen on my phone turned out to say more than the general page does.

• About Advantage+ Audience, Meta Business Help Centre. The product documentation for Meta’s automated audience system, described as using AI to find the most relevant audience for a campaign. It’s also the page that explains how minimum age and location function as strict criteria the system won’t expand past, which is what turned “Australia, 18 or over” from an ending into a beginning. It establishes that delegated targeting exists as a product. It doesn’t establish that this campaign used it, and I’ve been careful not to imply otherwise.

• About lead ads with instant form, Meta Business Help Centre. How the in-app form works, including pre-filled contact fields. Read this if you want to understand why the form already had my name in it.

• About privacy policies for lead ads, Meta Business Help Centre. Covers the privacy policy an advertiser must attach to a lead form, and Meta’s own position that the policy has to explain how lead data is collected and used. Relevant to the section on the assurance sentence.

• Meta Privacy Policy. The broad description of what activity and content feed personalised advertising. Supporting context rather than evidence about any particular impression.

• Retrieving Leads, Meta for Developers. Developer documentation showing that lead data can be retrieved alongside identifiers for the ad and form that produced it, with ad_id and campaign_id named directly. This is what supports the point that a submitted form can be a person and a measured result at once. It says nothing about what the Church’s actual integration does.

The Church’s own material

• Preparing for Referrals, Come Unto Christ. Training material describing people who respond after a Facebook or Google advertisement as Media Referrals, encouraging follow-up within 24 hours where possible, and covering the Area Book Planner. It also tells missionaries not to turn up and substitute a different offer from the one that brought the person in, which is a real safeguard and belongs in any fair account.

• Request a Missionary Visit, Come Unto Christ. The Church’s public web form for arranging missionary contact. Worth comparing with the in-app form, though they aren’t the same thing and I haven’t treated them as interchangeable.

• Privacy Notice, The Church of Jesus Christ of Latter-day Saints. Describes processing contact details for missionary and teaching purposes, analysis of that data, sharing with third-party processors, transfers between Church entities, and storage in the United States. Read it next to the sentence on the form and draw your own conclusion. I’ve flagged in the essay that I haven’t yet confirmed this is the notice governing that specific form.

• Privacy Notice, Australia. A country-specific version of the notice above. Since the ad was served in Australia and the form asks for an Australian address, this is arguably the more relevant of the two, which is part of why the question of which notice governs the form is still open.

• Data Privacy Office. How the Church describes its own privacy governance and the routes available to people who want to ask about their data. Included for completeness; I haven’t made a request.

• “Depression”, Ensign, February 2016. Separates ordinary sadness from major depressive disorder and states plainly that the latter is a mental illness requiring professional treatment and medication. This is the strongest thing in the Church’s favour and it’s why the campaign is worth writing about rather than dismissing.

• “Myths about Mental Illness”, Alexander B. Morrison, Ensign, October 2005. Rejects the idea that mental illness is a matter of sin or something a person should snap out of, and points people toward skilled health care providers.

• “Finding a Mental Health Professional Who’s Right for You”, Liahona, January 2019. A guide to the difference between psychiatrists, psychologists and counsellors, and how to choose one. Together these three make it hard to argue the institution is hostile to treatment, which is the strongest objection to my own essay.

Australian law and standards

The short version: several regimes come close to this campaign and none of them obviously lands on it. Everything below is a map of what might apply, not a finding that anything was breached.

• Privacy Act 1988, current consolidated text. The federal statute containing the Australian Privacy Principles, definitions, entity coverage and exemptions. Whether any of it applies here depends on which entity is collecting, which the public record doesn’t settle.

• APP 3, collection of solicited personal information, OAIC. Explains the extra consent required for sensitive information, which includes health information, along with data minimisation and the requirement that collection be by lawful and fair means.

• APP 5, notification of collection, OAIC. What an organisation should tell you at or before the moment it collects your data, including its identity, its purposes, who it usually discloses to, and any overseas recipients. This is the most relevant one to the assurance sentence on the form.

• APP 6, use or disclosure, OAIC. Covers using data for the purpose you collected it for, and when secondary use is permitted.

• APP 7, direct marketing, OAIC. Sensitive information can only be used for direct marketing with consent. Only becomes relevant if follow-up contact occurs and the entity is covered.

• Tracking pixels and privacy obligations, OAIC. About a different technology, but the principles on how ordinary digital activity can reveal or imply health information are the closest available guidance to what happened on my phone. I’ve used it as an analogy and labelled it as one.

• Privacy Commissioner finds privacy breaches in third-party tracking pixel investigation, OAIC, 24 June 2026. Two determinations finding that Medmate Australia and Monash IVF interfered with privacy by collecting health information through third-party pixels and using it for targeted advertising without consent, in contravention of APPs 3.3, 5.1 and 7.1. A companion inspection of 50 health provider websites found 96% used tracking technologies and 52% used at least one third-party pixel. This is the closest thing to relevant Australian authority I’ve found, and I’m not offering it as a precedent that settles anything about a missionary advertisement. What it shows is a regulator now looking hard at health context and targeted advertising.

• AANA Code of Ethics. The self-regulatory code for Australian advertising. Section 2.6 concerns advertising that encourages or condones unhealthy or unsafe behaviour having regard to prevailing community standards. Applying it to a missionary advertisement that never discourages treatment would be novel, and I don’t know of a decided case like it.

• False or misleading claims, ACCC. Consumer law turns on the overall impression an advertisement creates, and a business has to be able to substantiate its claims. Its application here is doubtful, because free religious outreach and a promise of hope and comfort aren’t what the regime was built for. Included to explain why consumer law isn’t the route.

If you find an error in any of this, or you know something about how these campaigns are configured that I don’t, tell me. I’ll correct the essay and say what changed and when. An explanation that can’t be revised isn’t worth much, which is more or less the argument of the whole series.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open for subscribers (free and paid); I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

I’d been on the couch a while by then. A full day at TAFE behind me where I hadn’t concentrated on a word or contributed anything worth hearing, and I’d come home and sat down and started working through whether I was cut out for the course at all.

Then someone from my class messaged to ask if I was okay. They’d noticed I seemed off that day, and they were right, and I still don’t know why I answered honestly.

I typed the whole thing out. Five sentences, maybe six. I ran a coffee shop. It got to be more than I could handle. I walked out mid-shift, went to the emergency department up the road, and stayed in hospital for seven months.

I read it back before I sent it. It looked clean. Cause and effect, in order, a complete account with nothing left over.

I sent it. Then I felt better than I had in weeks.

Not long after that I started writing here.

What I actually sent

What I sent was an argument, and I’d been making it for years without once looking at how it was built.

Reduce the account to its working parts and this is what’s left.

I could not keep up with what the work needed from me.

I fell apart.

Therefore something was wrong with me.

The first two lines are true. There are records. There are people who watched it happen and would confirm every detail of it. And the third line does not follow from them. Nothing in those two sentences produces that verdict. Someone could accept the whole account and refuse the conclusion, and I’d have had no idea what to say to them, because I had never noticed there was a gap there to defend.

I’d told that story hundreds of times by then. To family, to admitting nurses, to people who asked at parties and regretted it. Each telling made it smoother. None of them made it truer. Repetition works on the delivery and leaves the structure completely alone, which is why a story can get more convincing every year while the hole in the middle of it stays exactly the same size.

The premise nobody says

D.Q. McInerny spends most of Being Logical on arguments laid out in full. Premises visible, every part available for inspection. Real arguments almost never look like that.

An argument with a piece left out is called an enthymeme, and it’s how nearly everybody talks nearly all the time. Someone says he’s a surgeon, so he can handle pressure. The sentence doing the work there, that surgeons can handle pressure, never gets spoken, because speaking it would be strange. The listener supplies it without noticing. So does the speaker.

Leaving that piece out is necessary. Conversation would be unbearable otherwise, and every exchange would turn into a deposition.

The trouble arrives later. An unstated premise never gets argued with. It isn’t in the room. Nobody can disagree with a sentence that was never said, and that includes the person leaning on it hardest.

My missing line

Between “I fell apart” and “something was wrong with me” there is a sentence I never wrote down.

A person who can’t meet reasonable demands has something wrong with their character.

That’s the hinge. Put it in and the argument runs. Take it out and the account is a sequence of events with no verdict attached to it.

I never said that sentence to anyone. If someone had put it to me about a stranger I’d have argued with them, and I’d have argued well, because as a general claim about people it’s obviously rubbish and I’ve believed it was rubbish my whole adult life. Applied to me it was the floor. Floors don’t get examined. They get stood on.

Nobody was hiding it from me either. That’s what I keep circling back to. An enthymeme isn’t a trick, and there was no version of this where somebody had quietly slipped a false line into my reasoning while I wasn’t looking. I did it myself, in daylight, and the reason it worked is that I never wrote it down.

Here’s the part that still bothers me.

I already had the counterexample.

I was diagnosed with ADHD about five years before it all came down. I spent two of those years medicated, and in those two years I opened the businesses. Built them from nothing and got them running.

Then I came off the medication. That happened because of a misdiagnosis, which is its own essay and I’ll write it when I can stand to. It matters here only because it isn’t finished. I still haven’t got back on stimulants. I’m closer than I was.

Some time after that I was standing in an emergency department telling a triage nurse I couldn’t take it anymore.

The businesses I closed were the businesses I’d built while treated. Same hands, same trade. One variable moved. That’s about as close to a controlled experiment as a life ever gets, and the result was sitting in my own history with a date on it.

And it never entered the argument. Not once, in five years of explaining to myself why the businesses ended. I was carrying evidence that pointed somewhere else entirely, dated and lived through and impossible to dispute, and my account of myself went on without it.

An unstated premise never faces scrutiny at all. That’s the whole mechanism. Evidence can only contradict a claim, and there was no claim on the table to contradict. So the two years I spent medicated and building sat in my memory as a good run that happened to me, filed under biography, doing no work whatsoever in the case I was assembling against myself.

Why it took so long

Nobody handed me the sentence. There was no session where a clinician said it out loud and I sat there with my mouth open. It came apart slowly, over years, in pieces I couldn’t put back in order now if I tried.

For a long time I read the delay as a failure of attention. Diagnosed, treated, unmedicated, hospitalised, discharged, and still running the same two-premise argument with the same buried third line underneath it.

The reason is structural. The premise sat outside the part of the argument I could see, underneath, keeping everything else upright.

An unstated premise arrives as the world itself. It doesn’t announce itself as one option among several, so there’s nothing to compare it against and no moment where checking it would occur to anyone. That’s the problem with auditing your own story and I don’t have a clever solution to it. The correction has to come from outside the argument. Something has to put the premise into words before it can be found false.

For me that was partly the coursework, which gave me the vocabulary to hear my own reasoning as reasoning rather than as description. Partly the writing, which forces sentences into the open where they can be looked at properly. Mostly it was time. That was my route and I’d be cautious about recommending it to anyone, because five years is a long time to be wrong about yourself and I don’t believe the length of it was necessary.

What I’d send now

The account hasn’t changed. Coffee shop, mid-shift, emergency department, seven months. Every word of it still stands, and I’d type it the same way.

What changed is the sentence underneath, and it changed the verdict without touching a single fact.

A brain with impaired attention regulation, unmedicated, under sustained executive load, will eventually stop.

Same evidence. Different conclusion. Nothing that happened happened any differently.

I’ve been answering that message ever since. Everything I’ve written here is another line of the reply, which is either an unusually long response to a simple question or exactly the right length for one.

That night on the couch I’d decided I wasn’t cut out for the course, on no evidence beyond one bad day, and I hadn’t noticed I was doing the same thing again in miniature.

What I don’t know is what’s under the floor now. Whatever premise I’m running on today is, by definition, the one I can’t see, and I’ll be standing on it right up until somebody says it out loud where I can hear it. That’s the thing about arguments I didn’t understand for five years. The ones I make about myself are the only ones I can’t hear.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

The final part of Nothing Bold. Begin with Part One: How Was the Trip, or if you missed it read Part Three: The Second Mug.

He clocked off at seven and walked home. The morning was grey and windless and the woman in scrubs was at the bus stop and did not look up. This time he found nothing fine about it. It was just two people not looking at each other. You could go your entire life inside the not-looking and it would hold you like water holds a stone.

In the flat he sat on the edge of the bed in his vest with the name badge still pinned to it. First name only. The bed made behind him, the bench wiped, the fridge full against its list, the two chairs for no guest: everything squared and maintained and accounted for. A month of proof and no examiner. He had been keeping the books for a life nobody had asked him to justify.

The deadline was Monday. Two days. He had invented it on a ward bed with a paper bag between his feet, the arithmetic of a frightened man. Monday would arrive and inspect nothing.

He took the badge off and turned it over and sat there holding it. Outside, the wattle went on with its slow work. He did not sleep.

The light came into the flat the way it always came, first along the ceiling and then down the wall above the table. He had not slept. Somewhere around eight the block woke around him, pipes ticking in the wall, a door on the ground floor, a car reversing out of the carport with Triple J already going. He got up because sitting had stopped being anything.

He made the bed. He had not slept in it but he made it, then washed the mug from yesterday and wiped a bench that had nothing on it. The cloth went over the laminate in three long passes. At the end he waited for the small click and it did not come. The form had stopped accepting entries. He rinsed the cloth and looked at the kitchen, which was perfect and proved nothing.

He took the vest off and hung it on the chair by the door with the badge face down. The kettle boiled and clicked off. He stood at the window while the steam thinned over the bench. Sunday under a high white sky. The wattle had shed yellow onto the roof of someone’s ute. Across the units, a towel and two work shirts hung dead still on a folding rack.

He kept returning to the second mug. The coworker had loaded a telescope on the chance of average sky and packed another mug on the chance of company. He had poured for a man he knew three facts about. In return, he had received one true sentence and watched it turn into a joke.

Telling the coworker would have been a beginning. A first true sentence could be followed by a second; a person could begin to know him. The email was different. She already knew enough to notice the absence. She had asked carefully, without demanding an account, then stood back from the question. He had spent a week drafting answers to something she had not asked. Her question was only: are you there.

Monday would certify nothing. The flat was real. The job was real. The twenty-two-minute walk was real. The deadline was not. The raft had floated; that did not make it land.

He got the laptop from beside the bed and set it on the table before anything in him could convene a meeting.

He read the email again, straight through, and stopped at the last line. I just wanted you to know someone noticed. Then he pressed reply. The empty white field opened under her name and the cursor went on and off like a slow instrument reading, and he typed, Hi. Thanks so much for reaching out, it really

He deleted it. Really was doing something in the sentence he did not trust.

He typed, Hi. Sorry for the slow reply. I’ve been meaning to write back all week. The truth is I wasn’t on any kind of trip, I had a health issue that needed dealing with and it took a long time, and coming back has been

He looked at that for a while. It was true and it was also a report, it had the sound of a man reading from a page at a lectern, and she had not written him a report. She had written him the projector meeting and the thermometer joke. He deleted back to Hi and sat with his hands off the keys.

Outside a car went past with its window down and a fragment of someone’s music arrived and left.

He typed, Hi. Thank you for writing. I’m sorry it’s taken me a week to answer, I’ve opened this email about nine times. I wasn’t on a trip. I was dealing with something medical, and it took about six months, and I’m okay now. Or getting there, which I’m told is the same thing said honestly.

He stopped. He read it. The last clause had a shine on it he did not like, it was performing, it wanted her to think something of him. He deleted, which I’m told is the same thing said honestly, and left, Or getting there.

Then he typed, I remember the projector meeting too. I can’t believe you kept that joke. It wasn’t even a good joke.

He looked at that and left it.

Then he typed the last part slowly, one word at a time, the way you carry something full across a room. You said you wanted me to know that someone noticed. It means something that you did. More than I would have guessed, honestly. Thank you for asking when nobody else did.

He read the whole thing twice. Six sentences and change. It said nothing about the ward and everything he was prepared to have exist outside his own head. The gap between those things was enormous, and the message did not pretend otherwise. It stood on its own side and waved.

He moved the cursor to send and there was a moment, one full second, where the whole apparatus stood up in him at once, every argument, all of it well made, and he pressed send while it was still getting to its feet.

The email went. The screen showed the sent folder for a moment, one item, and then the inbox, empty and grey, nothing bold in it anywhere.

He closed the laptop. He sat with his hands flat on the lid, and the flat went on around him, the fridge running, the wattle at the window, a door somewhere below. There was no sensation of arrival. He noted this. He had half expected one and there was only the room, and himself in it, and the small fact of the sent thing, already out there somewhere in the machinery, already hers as much as his.

He got up and filled a glass from the tap and drank it standing at the sink, looking at nothing, the whole glass, and washed it and set it on the rack to dry.

Thank you for reading Nothing Bold. You can find all four parts here: Part One · Part Two · Part Three · Part Four.

This story is fictional, though some of its emotional territory is drawn from my own experience of returning home after a long psychiatric hospital stay.

I’ve had this final part edited and ready to publish for a few days now. I set myself a deadline to be finished this weekend, and here I am on Sunday at 10.35pm, almost not making it, mirroring the story. Funny that.

I have further ideas for some fiction pieces, but I’m not sure I have the confidence to post any more. I hope you read all four parts. I hope you felt something from it. I hope you enjoyed it.

- Jacob

An index has no argument. It only tells you where things are.

Bold page numbers mark days spent in hospital. Italic page numbers mark days spent telling people it was fine. The numbers are the only invention here.

Absence, requesting, see Leave, denied
Anhedonia, see Taste, loss of
Antidepressants, 91, 96, 103
Apron, still tied on, 12 \

Bed, not leaving, 31, 47, 88-112
Being Logical (McInerny, D.Q.), 60-74, passim
Bins, going unemptied, passim
Business, the, see Coffee shop; Walking out \

Camera, unpaid, 20
Care, self-, see Rabbit, the
Certificate IV in Mental Health, 126
Christmas, on the ward, 101
Chungking Express, 4, 4, 4 [sic]
Coffee shop, owning, 3-11; leaving, 12 \

Degree, computer science, 1-90 (incomplete)
Diagnosis, arriving late, 118
Discharge, 112
Doctors, not trusting; later trusting, 70, 89 \

ECT, 88-112
rounds of, 25
Emergency department, going in through, 87
Employees, the people I employed, 6
England, leaving, age 17, 1; see also Leaving, patterns of \

Family, in the corridor, 88-112, passim
Fine, “I’m”, passim \

Good day, a, 38
GP, the ten minutes with, 82 \

“Help me. I can’t take it anymore.”, 86
Hope, see Waiting
Hospital, see Bed, not leaving; ECT \

Insomnia, passim \

Kettle, standing at the, passim
Keys, left in the door, 12 \

Leave, denied, throughout
Leaving, patterns of
as a child, 1
mid-shift, 12
six months from the end, 90
Lithium, see Mood stabilisers \

Machine, keeping it running on will alone, 5-11
Mood stabilisers, 96
Monopoly Deal, 95-104
worked better than the electricity, 104 \

Nothing, notes of, see Taste, loss of \

One to ten, on a scale of, 90 \

Phone, face down, passim
Publishing, weekly, 130 \

Rabbit, the, 128
Recovery, see Recovery \

Shower, the, counted as an outing, 52
Sitting up, the negotiation before, 44
Suicidality, rating of, see One to ten, on a scale of \

Taste, loss of, 50-58; return of, 129
Television, communal, 99 \

Waiting, passim
Walking out, 12; see also Leaving, patterns of
Weight of it, the, 91-112
Well, getting, 214 \

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

It’s 9pm and you’re horizontal on the couch, thumb moving, watching a girl you knew at 16 stand on a beach in Portugal.

Then an engagement.

Then a promotion for a job title you didn’t know existed. By the tenth post you’ve run a rough tally in your head, and you’re losing, and you can’t remember deciding to count.

This many milestones for them, this many for you, subtract, wince.

That sum was rigged before the first post loaded.

For me it was graduation photos. A whole season of gowns and proud parents while I sat on the couch, having dropped out of uni three times.

Then holiday photos, while I was between work again and couldn’t have afforded the taxi to the airport.

Then the one with teeth: old colleagues posing outside their coffee shops. Busy counters, queues out the door. I’d closed my own businesses down. Out of work and depressed, doing the maths on people I used to manage.

The gowns and the beaches were other people’s lives. The coffee shops were mine, running on without me. Same industry, same skills, and the feed was showing me the version of me that didn’t fold. That tally cut deeper than envy. It felt like evidence.

Each verdict arrived as plain fact. I’m behind. It read like a measurement. That’s the trick of depression’s arithmetic: the readout looks like an observation.

But look at the inputs. On your side of the ledger you have everything. Every 3am, every unanswered text, every hour on this couch, every version of you that didn’t survive the draft folder. On their side you have a beach and a caption.

You’re grading your director’s cut against their trailer. The reshoots and the scenes that got you the R rating, weighed against 90 seconds someone chose to show. The fight before the beach photo never gets posted. The loan behind the shopfront stays off the books. Their bad year doesn’t make the feed. Yours is the only bad year you have full footage of, so it wins every comparison by default.

The sum comes out the same every time because the inputs are fixed. That’s what rigged means.

So what do you do with a calculator you can’t put down?

I learned the first move from people whose job it was to teach me, when things were at their worst. When you catch the counting, say what it is. “This is depression doing maths.” Out loud if you’re alone, in your head if you’re not.

It sounds too small to work.

Here’s the mechanism. Arguing with the total keeps you inside the calculation: was it really that bad, am I actually that far behind. Every argument accepts the premise that the sum deserves an answer, so you end up checking rigged maths on the same rigged calculator, and it confirms itself every time.

The label steps outside instead. “This is depression doing maths” puts the problem upstream of the numbers. The verdict gets demoted, from fact about your life to symptom with a name. Weather.

I’d like to tell you it stuck completely. It stuck-ish. Even now, I still find myself mid-tally before I’ve noticed the counting start. The naming works and I still need it, both at once. There’s no finish line here, and anyone selling you one is selling.

The second move is cutting the fuel. The tally runs on the scroll and the quiet of the 9pm couch, so my circuit-breakers are embarrassingly small.

Phone in the other room. Kindle instead of the feed, because a book is a screen that can’t rank me against anyone. Walking with music loud enough to crowd out the counting.

It’s cutting the power to a machine that only runs while I feed it. Your wiring is different, so your breakers will be too. The principle is the part I’d bet on: the sum stalls without fuel.

You’ll still catch yourself counting. I do. The counting starts on its own, and that part may never file its resignation.

The difference is the signature. The tally can run all it likes, but a verdict needs someone to sign off on it, and you can decline. Watch the machine do its arithmetic, note the readout, feel the old urge to sign, and put it down unsigned.

It’s 9pm again, sooner or later. The couch, the girl on the beach in Portugal. Somewhere around the third post you feel the count start up, and this time you recognise the hum of it, and the phone goes in the other room mid-sum.

The total never gets announced. It can’t finish without you.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

Part Three of four. Begin with Part One: How Was the Trip or read Part Two: No Version on File.

Saturday morning he sat on the edge of the bed with the laptop open on his knees. The email had been there almost a week. He clicked on it.

It was longer than the preview had promised. She apologised twice before arriving anywhere. She said she knew they had never really known each other and that this might be strange to receive. She said she remembered the planning meeting where the projector had died and he had run the whole slide deck off printed pages without missing a beat. She still thought about a joke he had made in the tearoom, about the fundraising thermometer poster, and quoted it back to him word for word. Then she wrote that she hoped whatever had happened had worked out, followed immediately by, Sorry, that’s a stupid thing to say when I don’t know what happened.

I noticed when you stopped coming in, she wrote. I asked around a bit and nobody seemed to know anything, or nobody wanted to say, and I didn’t want to push. But it’s been sitting with me. You don’t have to answer this. I just wanted you to know someone noticed.

He read it twice. The wattle moved outside the window and the traffic ran on the highway and the flat held its ordinary morning quiet around him.

Then he closed the laptop and set it on the floor beside the bed, squared to the skirting board, and lay back on the made bed in his clothes and looked at the ceiling for a long time.

He arrived at the store Saturday night with the email in him like a stone swallowed at breakfast.

He clocked on and got his cage and started on aisle four, and within the hour he had shelved the long-life milk one bay too far left and put the powdered custard in with the jelly crystals and fronted half a run of sauces before noticing he had done the same run twenty minutes earlier. He stood at the pallet at one point with the box cutter in his hand, blade out, not cutting anything, and he did not know how long he had been standing there. Two minutes possibly. The scrubber went past the end of the aisle and the driver did not look at him.

And here, the narrator said, the Western Australian Male demonstrates the freeze response characteristic of his species. Observe that no predator is present. Researchers remain unable to account for this.

He retracted the blade and cut the pallet wrap and went on working, and the narrator, which had been company for a month, kept offering observations of that kind, and he noticed the change in its material. It had turned on him at some point in the last two days. The commentary used to be about soup. Now it was about him, and it knew where he kept things.

She had quoted the joke word for word. That was the part he kept handling. Not the kindness, though the kindness was its own problem. The quote. He had said something forgettable in a tearoom to a room of people who had been paid to be there, and one of them had carried it out of the building and kept it for a year and produced it now as evidence, and the evidence proved something he had not been prepared to have proven, which was that he had been visible the entire time. He had gone through his old life believing himself to be a kind of pleasant furniture and someone had been taking notes. There was no way to file this. The filing system was built on the premise that nobody was looking.

The coworker was on again. They crossed at the dock at one and worked opposite sides of grocery until the break, and at half three the coworker stuck his head down aisle seven and said, I loaded the scope. Sky’s average but I’m having my break outside anyway. Come out if you want.

He looked at the pallet, which could wait, and at the tins, which could wait, and everything in the store could wait, that was the entire nature of the store.

Yeah, he said. All right.

The dock at night gave onto the back carpark, empty except for their two cars and the coworker’s ute and the row of skip bins along the fence. The coworker had set two milk crates upside down near the tailgate and the telescope stood on its tripod beside them, aimed at a sky that was mostly a grey lid, cloud from one fence line to the other, one dim star showing in a gap over the loading bay and then not showing.

Waste of a setup, the coworker said. Saturn’s behind all that somewhere. He sat on a crate and produced a thermos and two enamel mugs from a canvas bag, as if two mugs were standard equipment, as if the second had always been coming along on the chance of company. He poured without asking and handed one over. The tea was milky and far too sweet, and better than anything the urn had produced in the store’s history.

They sat. The cloud moved east in one piece. Somewhere over the fence a reversing truck beeped its way through a yard and stopped.

I nearly bought a proper mount last month, the coworker said. Motorised, tracks the sky for you. Eight hundred bucks. Then I thought, mate, you point a tube at the sky and look through it. The moment you’re spending eight hundred to have a machine do the pointing, something’s gone wrong with the hobby.

Seems right, he said.

The tea steamed off in the cold. The coworker talked for a while about nothing in particular, the roster, a possum that had got into the dock two winters back and lived above the freezer aisle for a week, and then let a silence run, and the silence was not the kind that asked to be filled. He drank his tea inside it. On the far side of the carpark the skip bins stood in their row and the security light above them clicked off, satisfied, and the dark came down a grade.

How’d you end up on nights, he asked. He heard himself ask it. It was the first question he had asked anyone in a month that was not about stock.

The coworker turned the mug in his hands. Few years back I was in a bad way with money, he said. Not drama, just a hole. Owed people. Days I was doing removals, cash, and I needed a second thing that didn’t care who I was, and nights don’t care who you are. That’s the whole appeal. You turn up, nobody asks you anything, the work’s just the work. He shrugged. Then the hole filled in and I found out I liked it. The quiet. Days I’ve got mum, she’s got her thing going on, she needs someone around till about lunch. Nights suit that. So now it’s on purpose.

He said all of it in the same voice he used for the beetroot. It was not a confession. It was inventory, offered across a tailgate, and when it was done the coworker topped up his own mug and looked at the cloud as if the cloud might have developed an opinion.

He sat with his tea and understood only this much: the problem had never been vocabulary. A true sentence, once it entered another person, would become a fact in the world, no longer soft and unfinished and entirely his. The flat, the list on the fridge, the twenty-two minutes: he had arranged the month to keep it that way.

The cloud shifted. The one star came back over the loading bay.

I wasn’t working, before this, he said.

The coworker looked over. Not sharply. The way you look at a person who has started speaking.

I was away for a while, he said. Not travelling. He turned the mug a quarter turn on his knee. Away away. Not by choice, exactly. It was

The sentence stopped there. Six words, seven more would finish it. The carpark was empty and the man beside him had just laid his own bad years on the tailgate like tools, and he still could not make the rest come out.

Long service leave, he said. From life. Highly recommend it.

The joke landed with a small dry click. The coworker looked at the sky for a moment and said, Well. Tea’s better out here anyway. Then he returned to the mount he had not bought and whether Saturn would be up tomorrow, leaving the unfinished sentence where it was.

They went back in at four. He worked the rest of the shift in silence and the work was clean, the tins fronted, the cardboard folded, the cage emptied and returned, and the narrator said nothing at all, the whole rest of the night, not one observation, and the silence where it had been was a new kind of silence and he did not like it.

The final part arrives this weekend.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

The notification lands at 9pm. How was your day?

Five faces wait under the question, grim on the left, beaming on the right. Your thumb finds the third one before you’ve finished reading. Tap. Done. The app throws a little confetti: 47 days in a row.

Somewhere in there is a month view too. A grid of coloured dots, one per day, most of them the same agreeable yellow-green. You scroll past it sometimes on the way to something else and feel a small, vague pride, the way you’d feel about a plant you’ve kept alive.

The whole ritual takes about 11 seconds. The phone asks, the thumb answers, the streak grows. You picked this app because a therapist suggested tracking, or a friend swore by it. A small act of maintenance, like flossing. The app agreed with you, and promised a year-in-review that would reveal your patterns.

Eleven seconds a night to know yourself better. That was the deal.

Then comes the bad night.

A day that sat on your chest from the morning, where every task took 3 times its usual weight and you cancelled the one thing you’d been looking forward to. The notification lands at 9pm, same as always. How was your day?

The honest answer is the second face. Maybe the first.

And your thumb hesitates.

Because a dip looks bad on the graph. Because 46 yellow-green dots and 1 grey one reads like a problem, and you don’t feel like being a problem tonight. Because if you log the second face, the app might ask a follow-up question, and you don’t have 11 seconds of honesty in you, let alone a paragraph.

So you tap the third face. The week stays clean.

You lied. And the strange part is who you lied to. There’s no clinician reading this graph, and no algorithm deciding anything that matters. Nobody sees that chart but you. You protected the graph from its only reader.

In 1975, the British economist Charles Goodhart noticed that statistical relationships tend to break down once they are used for control. The idea was later popularised in a cleaner sentence: when a measure becomes a target, it ceases to be a good measure. Banks do it with regulatory targets. Schools do it with test scores. And you, at 9pm, did it with a cartoon face.

You weren’t trying to improve your mood. You were trying to improve the record of your mood. The measure had become the thing being managed.

The deliberate lie has a quieter cousin. Think back over the last three weeks of third faces. How many times did you stop, run the day back, weigh it, and answer? The thumb learned the route long ago. The tap happens somewhere below thought, the way you sign your name.

On those nights the log records attendance: proof you showed up to the question, with no trace of an answer. A month of that and the grid is fiction by inertia, no bad night required.

The app’s pitch was the opposite. A mood tracker has one job: catch the slide early, while it’s still a slope and you can still walk back up. But the slide is exactly when the curating starts. The worse things get, the more tempting the third face becomes, because the honest dots would pile up into a picture you don’t want to see. So the graph stays yellow-green while the floor drops, and the tool goes quiet right when it’s needed.

Come December, the year-in-review shows a calm, even sea of yellow-green, and tells you less about your year than your bank statement does.

A fire alarm that turns itself off when it smells smoke.

Take the target away from the app instead. Four ways to do that.

log facts, drop verdicts

There’s an old distinction in logic between statements of fact and statements of value. “I slept 5 hours” is a statement of fact. It’s true or it’s false, and you know which. A face emoji is a statement of value: a verdict on the whole day, delivered by the same tired judge who lived it.

Verdicts get performed. Facts mostly don’t. It’s easy to round a “meh” day up to a smiling face; it’s much harder to claim you slept 8 hours when you watched the ceiling until 3am. Behaviour leaves receipts.

So change what you log. Pick 2 behaviours in advance, ones that move when you do: hours slept and whether you left the house, or meals eaten and plans cancelled. Log those instead of the face, or at least beside it. You’re swapping a report card for a logbook, and a logbook is much harder to flatter.

log daily, read weekly

Psychology has a name for what the nightly chart-check does to you: reactivity. The act of measuring changes the thing being measured. Korotitsch and Nelson-Gray’s review in Psychological Assessment (1999) treats it as a standard, expected effect of tracking your own behaviour. Clinicians sometimes use that reactivity deliberately. Smoking is one of the classic examples: asking people to record their cigarettes can itself alter how much they smoke.

The chart is never merely watching you. You are watching yourself through it.

Step on the scale every morning and the number starts running your breakfast. Check the mood graph every night and the graph starts running the mood, or at least the report of it.

The fix is distance. Log every day, but read the chart once a week. Sunday, coffee, 2 minutes. If the app lets you hide the graph view, hide it. If it pushes streak notifications, kill them.

A week of dots read together tells you something a single dot never can: direction. One grey day is weather. Five grey days is a slope. You can only see slope from a distance, and the nightly check keeps your nose pressed against the glass.

treat the caught edit as data

You will still nudge a score. Some night the thumb will drift up a face and you’ll catch it mid-move, the way you catch yourself saying “I’m fine” in a voice that convinces nobody.

That moment is worth more than the rest of the month combined.

The nudge marks the exact night the performing instinct beat the noticing instinct. When you catch the edit, log the honest number, then write one plain sentence in the notes field about why you wanted to lie. “Didn’t want to break the streak.” “Didn’t want it to be real.” Whatever’s true.

The lie you almost told becomes the data point.

make the app reapply for the job

Some trackers can’t be saved, and here’s the test: if the number changes how you feel more than what you do, it loses its vote.

A useful measure changes behaviour. You see 5 hours of sleep stacking up across the week, so you go to bed earlier. That’s a measure doing its job. A useless measure changes feelings. You see a 61 where yesterday there was a 74, and you carry that dip around all morning while doing nothing differently. That’s a number collecting rent.

Run the test properly: take a full week off the app. Log nothing. At the end, ask what you actually missed. If the answer is “I noticed a slump on Tuesday and adjusted,” reinstall it. If the answer is “I felt lighter,” you have your answer too. An app that only ever generated guilt was manufacturing a mood of its own.

Everybody’s line sits somewhere different here. Some people run trackers for years and get real early warnings out of them. Some people are better off with a paper diary and no graphs at all. The test is whether the method still tells you the truth.

The notification will land tonight at 9pm, same as always. How was your day?

Maybe now you answer with a fact. Slept 6 hours, kept the dinner plan. Maybe you write one honest sentence on a night you’d rather round up. Maybe the streak breaks at 47 while you’re doing something better with the 11 seconds, and the confetti never comes, and you notice you don’t miss it.

The app exists to help you notice yourself. Noticing is the part that matters, and it survives without the graph. People noticed their own slides for a few thousand years before anyone drew a dot grid over the month. They did it the unautomated way: a kitchen that tells on you, the third cancelled plan in a row. The dot grid added one thing to all that: a record that doesn’t flinch when you’d rather it did.

A graph nobody else sees has exactly one job. Tell the truth to the person holding it.

Let it do that, or let it go.

references and further reading

Goodhart, C. A. E. (1975). “Problems of Monetary Management: The U.K. Experience.” In Papers in Monetary Economics, Volume 1, pp. 1–20.

The original source of what became known as Goodhart’s law. Goodhart’s formulation concerned the instability of statistical relationships once they were used as instruments of economic control.

Strathern, M. (1997). “‘Improving Ratings’: Audit in the British University System.” European Review, 5(3), 305–321. DOI: 10.1002/(SICI)1234-981X(199707)5:3<305::AID-EURO184>3.0.CO;2-4.

Strathern applies the principle to systems of institutional assessment and gives the familiar formulation: when a measure becomes a target, it ceases to be a good measure.

Korotitsch, W. J., & Nelson-Gray, R. O. (1999). “An Overview of Self-Monitoring Research in Assessment and Treatment.” Psychological Assessment, 11(4), 415–425.

A review of the uses, accuracy and reactivity of self-monitoring. It discusses how recording a behaviour can alter that behaviour, as well as the factors that can make self-recorded information more or less accurate.

further reading

Caldeira, C., Chen, Y., Chan, L., Pham, V., Chen, Y., & Zheng, K. (2017). “Mobile Apps for Mood Tracking: An Analysis of Features and User Reviews.” AMIA Annual Symposium Proceedings, 2017, 495–504.

An analysis of 32 mood-tracking apps and their user reviews. The authors found that apps offered substantial support for collecting and displaying information, but considerably less help with deciding what to do with it.

Schueller, S. M., Neary, M., Lai, J., & Epstein, D. A. (2021). “Understanding People’s Use of and Perspectives on Mood-Tracking Apps: Interview Study.” JMIR Mental Health, 8(8), e29368. DOI: 10.2196/29368.

Interviews with people who had used mood-tracking apps in everyday life, covering why they began, what they found useful, why they stopped and the limitations of reducing emotional experience to predetermined categories.

Epstein, D. A., Ping, A., Fogarty, J., & Munson, S. A. (2015). “A Lived Informatics Model of Personal Informatics.” In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 731–742. DOI: 10.1145/2750858.2804250.

A useful account of self-tracking as something people do unevenly rather than continuously. The model includes lapsing, resuming and abandoning tracking, instead of treating uninterrupted adherence as the only successful outcome.

Epstein, D. A., Kang, J. H., Pina, L. R., Fogarty, J., & Munson, S. A. (2016). “Reconsidering the Device in the Drawer: Lapses as a Design Opportunity in Personal Informatics.” In Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 829–840.

Research on people who had stopped using activity trackers. The paper argues that returning to tracking is not always the right or only positive outcome; sometimes the useful result is retaining an insight and leaving the device behind.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and try to answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

It’s past 3am and I’m running the audit. The same audit, most weeks, in the same order, and item one never changes: I am awake, and awake is on the list of things to watch for.

That’s the trap built into it. Sleep is an early-warning sign. So lying here is either a bad night or a symptom, and the difference matters, because everybody has bad nights, and I have also had bad nights that turned out to be the first domino in a sequence that ended with seven months in hospital. So I check. Appetite: fine. The writing: happened this week. Showered: yes, if Tuesday counts as recent. Four items in, a second voice points out that healthy people don’t audit themselves at 3am, which means the audit itself belongs on the list, and now the audit is auditing the audit, and everything I find is either fine or

Tonight is supported by Restfold. If you’re lying awake at 3am, your mattress could be the reason. Restfold’s triple-layer cooling gel adapts to your body in real time, so you get the deep, restorative sleep you deserve. Try Restfold free for one hundred nights. Go to restfold.com slash breakdown and use code AWAKE for fifteen percent off your first order. Restfold. Sleep like nothing’s wrong.

Ads are placed automatically. Tonight is supported by listeners like you.

evidence. Fine or evidence. There’s no third category at this hour.

The audit moves to the accounts. I did this arithmetic years ago and I’ve never liked the answer: whatever it is in a person that gets them through treatment, I don’t think mine is infinite. I spent a lot of it across two admissions. Nobody can tell me the balance. Daylight me finds that uncertainty liveable. 3am me reads the statement over and over, looking for the line where the balance is printed, certain it’s low, certain that the not-knowing is

Running the numbers again? Sanctum is the meditation app for minds that won’t close the books. Guided sleep journeys, breathwork, and soundscapes recorded in real rainforests, designed to bring you back to balance. Try Sanctum free for fourteen days, $34.99 a month thereafter. Cancel anytime. Sanctum. Quiet, delivered.

Ads are placed automatically. Tonight is supported by listeners like you.

itself a symptom. See item four.

Somewhere past the arithmetic, the audience files in. Same seats every time. The people who knew me when I owned things. My family, who watched all of this happen once, and who in the spiral’s staging of it are watching for it to happen again, politely, the way you watch a glass near the edge of a table. And this year, a new row at the back: readers. I write a publication about getting well. A relapse now would have subscribers. The spiral is delighted with this material. It’s the freshest thing I’ve given it in years.

I know none of these people are awake. I know the theatre is empty and I built it. Knowing doesn’t close it. The show runs anyway, and underneath the commentary a smaller voice asks what I actually need right now, and the answer it receives is

Verdure Daily Greens. Seventy-four ingredients in one scoop, supporting immunity, digestion, energy, mood, and cognitive clarity. No more shelves of half-finished supplements. One scoop, once a day, everything handled. Go to verdure.com slash breakdown and use code SEVENMONTHS for twenty percent off your first tub.

Ads are placed automatically. Tonight is supported by listeners like you.

not that.

Nearly 5 now. The gaps have gone. Item after item, no air between them. I’ve stopped auditing and started bracing. If it comes back. If this is night one of it coming back. Whether I’d see it in time. Whether seeing it in time has ever helped. I don’t know wheth

Confide. Talk to a licensed therapist from your own bed, matched to you in under five minutes by our clinically informed algorithm. Video, voice, or text, whatever feels safe. Therapists available in all states and territories. Because you deserve support, and support should be as easy as everything else you order at night. First month half price with code BREAKDOWN.

Ads are placed automatically. Tonight is supported by listeners like you.

The following portion of tonight is unsponsored.

er I can do it again. There it is. The thought the whole night was walking toward, arriving on schedule, interrupted by nothing.

Seven months, across two stays. The wards, the paperwork, the visiting hours, the year. If it comes back, all of that is the price again, paid out of an account nobody can show me. I don’t know if I can pay it twice.

That’s the entire thought. It doesn’t escalate. At this hour, nothing cuts away from it.

I lie in it for a while. Minutes, probably. Long ones.

The alarm goes off. That’s the ending. These nights adjourn; the audit closes without filing a report. The day arrives with no interest in the verdict, and I get up, because getting up is the only appeal available.

But the format wants a sponsor list, so, for the record, recovery had one. It was brought to you by Medicare, which paid for the wards and the doctors and asked for nothing at checkout. By a card game in a common room, dealt at exactly the hours the audit likes best. By the people around that table. None of them offered a trial period or a code. Nothing that worked was for sale, which might be why nobody advertises any of it.

And the boilerplate was nearly right. It’s the one true line those ads ran all night. I was supported by listeners. Mine sat across a common-room table at terrible hours, and they listened.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

The trip, he said.

Yes, she said. Six months. Honestly. Some of us were very jealous, I’ll just say that. Where did you even go.

The dispensary assistant had drifted to the far end of the counter and was doing something with a clipboard. Behind her, a man in hi-vis was reading the back of a box of antihistamines. Nobody in the shop had any idea that anything was happening, and he understood that this was because nothing, by any visible measure, was.

Away, he said. Down south for a bit. Around.

Down south, she said, and put a hand on her hip as if he had confirmed a bet. See, I said Margaret River. Karen said Bali, and I said no, he’s not a Bali person, no offence, and she said everyone’s a Bali person, and I said not everyone. Was it lovely. It would have been lovely.

It was quiet, he said.

Quiet, she said, the way you would say the name of a dessert. God. Six months of quiet. Do you know what I’d give.

He watched her hold the basket against her hip and he did the arithmetic he could not stop himself doing, the ledger he kept without wanting to keep it. She knew a version of him that had ended seven months ago. The version had a lanyard and a desk near the kitchenette and made a certain kind of joke in the lift. She was speaking to that version now, over his shoulder almost, and the version was answering her, and he was standing slightly behind it holding a stapled paper bag, working the bag around behind his hip in increments.

The office is exactly the same, she said. Exactly. You haven’t missed a single thing, I promise you. Well, Deb retired. And they redid the carpet on two, which took forever, we were all in the annex for a month breathing glue. But apart from that. Same circus.

That’s good, he said.

Karen’s still Karen, she said, and laughed at something interior to herself. And what’s-his-name, Tom, from accounts, he’s got a baby now. A little girl. There’s photos everywhere.

He nodded at the news of Tom from accounts. He could not have picked Tom from accounts out of a lineup of two men. She kept going, names moving past him like traffic, and he found the rhythm of it, the small nod, the sound in the throat, and the rhythm carried the conversation the way the travelator by the trolley bay carried trolleys. It required nothing from him. It had never required anything from him, and he had once thought of that as a mercy, and standing there in the chemist he could not locate the mercy in it anywhere.

And what are you doing now, she asked. Are you back around here. Are you working.

I’m working, he said. Just something low-key for now. Nights, mostly.

Nights, she said. God, I couldn’t do it. I need my eight hours or I’m a monster. She shifted the basket to her other hip and looked at him with an expression of real warmth, and this was the part he found hardest afterwards, that the warmth was real, that none of it was pretend on her end, that she was a kind person being kind. Well listen, she said. It is so good to see you. Honestly. You look great, you’ve lost weight, it suits you. We should grab a coffee sometime.

We should, he said.

She gave a small wave with the hand not holding the basket, three fingers, and turned down the aisle toward skincare, and he heard her say something bright to the man in hi-vis about reaching the top shelf, and the man laughed, and then the shop was ordinary again.

Credit or savings, the assistant asked.

He paid. He walked home along the highway with the bag in his jacket pocket, and the traffic went past in its long dull tide, and the sky over the roofs had gone the flat white it went before rain that never arrived. Six months of his life had just been handed back to him with a resort attached. She would go home and tell someone, ran into him at the chemist, he’d been down south, he looks great. The story would enter the office and circulate and become the truth, because it was easy and there was nothing pushing back on it, and the person who could have pushed back on it had stood in the aisle and said, it was quiet.

Relief arrived first. The cover had held under a surprise inspection and held without effort. Under it was the feeling of a lift travelling the wrong way. He had been erased in the pain relief aisle by a woman smiling at him, and he had done most of the erasing himself.

At the flat he made the bed although the bed was made. He noticed himself doing it and did it anyway. He wiped the bench and started a shopping list he did not need and stood at the window while the easterly worked the wattle over, branch by branch. He opened the laptop. The email sat where it had sat, bold, and the receptionist’s voice ran under it, how was the trip, how was the trip. He closed the lid and got three hours of sleep in the striped light with his forearm over his eyes.

Friday night the crew was thin. Two call-outs and the flu going around, so it was him and one other down the whole grocery side, and the other was the coworker with the telescope.

They had overlapped before. He knew three facts about the coworker. The coworker chose nights and had chosen them for years. The coworker owned a telescope that lived in a ute tray in a padded case. The coworker could talk continuously without asking anything of the person being talked to, which made him, by the standards of the month, the easiest company available anywhere in the world.

They crossed at the dock around midnight, both of them wrestling pallets off the truck, and the coworker nodded at the load and said, someone in ordering has gone mad on tinned beetroot. Look at this. That’s a nation’s supply of beetroot.

Big weekend for beetroot, he said.

Must be. The coworker jacked the pallet up and leaned on the handle. You see the whiteboard. Someone’s complained the trolleys are too cold. Written a whole paragraph. The trolleys. Too cold.

They’re not wrong, he said. The trolleys are freezing.

That’s what I said. First customer feedback I’ve ever agreed with. The coworker walked the pallet off toward dairy, talking as he went, about a film he had watched that afternoon that was so bad it had circled around into being good, and the plot summary lasted the length of the dock and required from him only the occasional syllable, and he supplied the syllables and found that an hour had gone somewhere painless.

They took their break at three in the tearoom with its urn and its noticeboard and its two-year-old safety poster. The coworker looked out the window at the sky over the carpark.

Clear tonight, the coworker said. I was going to bring the scope out but I couldn’t be bothered loading it. Saturn’s up at the moment, early, before dawn. You can see the rings even with a cheap one. First time people look at Saturn through a scope they always think it’s a sticker. They think you’ve stuck something on the lens. It looks fake. Nothing real is allowed to look that neat.

Maybe next time, he said, and found that he partly meant it.

Yeah, no pressure, the coworker said. It’ll still be there.

The coworker tipped back in the plastic chair and drank the urn coffee without complaint. He seemed entirely inside the moment he was occupying, without a second self standing off to the side taking notes. He studied this the way you might study a man speaking a language you had once known.

So what’d you do before this, the coworker asked. The question came in on the same flat trajectory as the beetroot and the trolleys, no weight on it at all.

Office stuff, he said. Boring. This is better.

Yeah, the coworker said. It usually is. And that was the whole of it, the coworker moved on to the urn and whether the coffee in it had ever once been fresh, whether there was a founding coffee from which all subsequent coffees descended, and the moment closed over without a ripple.

For half a second, between the question and his answer, a different answer had stood up. Something true had been available: a person across the table with no version of him on file. He had reached past it and taken down the script instead. The script worked perfectly. Nobody at the table lost anything except him.

He finished the shift fronting up the tins. The narrator stayed quiet on the subject.

Read Part 3 here.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee

It’s 7:40 on a Tuesday and you’re brushing your teeth. The tile is cold. The kettle’s rumbling downstairs and your phone holds two texts you’ll answer at some point today. Nothing about this morning is wrong. Nothing about it feels like victory either.

And somewhere under the foam, a quiet audit is running. You’ve been getting better for a while now. Months, maybe. The medication is doing something, or the therapy is, or time is. People have started saying you seem brighter, in the careful voice people use when they’ve been worried.

So why does it still feel like nothing?

Blame the montage.

Every story about getting better ends the same way. The morning the light comes through the curtains differently. The scene where someone laughs for the first time in two years and the music swells so you know: they’re back. Films do it. Memoirs do it. So does every “I got my life back” testimonial, and every before-and-after that’s really about the inside of someone’s head.

So that’s the shape you’re waiting for. A return. A specific morning you could point to later and say, that’s when I became myself again.

And while you wait, you measure your actual days against it, and your actual days lose. You did the dishes. You answered one of the two texts. You were in bed before midnight. Weighed against a rebirth, that’s nothing. Weighed against a rebirth, everything is nothing.

This is where it gets expensive. The stretch of recovery that feels like nothing is usually the stretch where everything is working, and people stop right here. The meds get quietly dropped. The appointment gets moved, then moved again, then cancelled. The reasoning sounds sane while you’re in it: nothing’s happening, so why keep paying for it in side effects and waiting-room mornings.

The plateau was the climb. It never announces itself.

Change the unit of measurement

Try measuring recovery in availability. As in: how much of ordinary life is available to you now that wasn’t three months ago?

Availability is a boring metric. That’s its whole strength. Feelings will tell you nothing is happening for months and then ambush you twice in one week. Availability moves slowly, in one direction more often than not, and it leaves evidence.

Learn to recognise it. The signs are boring by design: food has taste again. You notice the weather. A text gets answered the day it arrives. The shower happens without an hour of internal negotiation. You read four pages of something before sleep.

None of these will ever feel like recovery, which is why you need the list. Hold your last week against it and count what’s back.

Count repetitions. Day nine of brushing your teeth is evidence. Day one can be luck, anyone can have one good day; the second time is a pattern forming.

Keep the tracking stupid: a note in your phone, a tally mark when the thing happens again. Skip anything with streaks or badges. An app that celebrates you will also, by design, have an opinion about the days you miss.

And you will miss days. Repetition tolerates gaps; that’s the difference between a practice and a test. A missed day is information. A missed fortnight is a conversation worth having with your doctor. Neither one is proof the progress was fake. And if the counting itself ever starts to feel like a stick you’re hitting yourself with, put it down for a week. It’ll keep.

One caveat that stays true no matter what: this is a way to measure, and it might not be yours. Everybody’s ordinary looks different. Someone’s tally says “left the house”; someone else’s says “stayed off my phone till noon.” Anyone who tells you exactly what recovery must look like is selling something.

Availability can return a yes this week. A definition of recovery that needs a dramatic return keeps reporting failure right up until the miracle morning arrives, and the miracle morning was always cinema.

What you’re actually waiting for

So watch the horizon less. The person you keep scanning for, the old you making the grand return, may have already walked back in and started quietly doing the dishes.

Returns like that skip the music. A Tuesday you can show up for, then, if things keep going, another one.

Tomorrow at 7:40 you’ll be back on the cold tile, brush in hand, and it will feel like nothing again. Count it anyway. Somewhere between day nine and day ninety, the feeling catches up with the evidence. Or it hasn’t yet, and the evidence keeps you going anyway.

Both count as recovering. Only one of them gets filmed.

Hello, I’m Jacob.

If this essay gave you language for something, share it with someone who needs the same words. And the comments are open; I read and answer everything.

Share

Leave a comment

I study and write about mental health: living with depression and ADHD, sometimes through film, sometimes through logic. New essays most weeks.

Everything is free to read right now. Paid subscribers get first claim on the extras I’m building (workbooks first), and some future pieces may end up behind the paywall.

I write these while I can. A paid subscription buys the time to keep going, so if you’re able, it’s the best way to back the work. A one-off coffee on Ko-fi helps too.

Subscribe now

Buy me a coffee