By the time this reaches you there will be ten days of August left. Not enough to start anything. Too many to ignore. The holiday, if you had one, has been had. Term has not begun. The out-of-office is still on, but nobody quite believes it any more.
Most people describe this stretch of the year in the language of waiting. They are waiting for September. And they tend to say it with a slight apology, because waiting is not supposed to count as living – it is what you do in corridors, at bus stops, outside consulting rooms. Ten days of it feels like something being taken rather than something being had.
This letter is about that feeling, which is far more physiologically expensive than it looks, and about what can be done with it.
The last fortnight of August is qualitatively different from the first, and everyone knows it without being able to say why. Early August has a reality of its own. It is simply summer. Late August has quietly become an anteroom to something else – you are still technically in one season while functioning as though you were in the next.
The French psychiatrist Eugène Minkowski, writing in the 1930s, made a distinction that has never been improved upon: we do not measure time so much as inhabit it. Lived time has texture, direction and weight. The future is not a place we travel towards at a fixed rate. It is something already present in how we hold ourselves now – leaning towards, bracing against, reaching for.
In late August the future has arrived in the room without the present having left. Both are here, and neither has your full attention. This is not a failure of mindfulness. It is what a threshold is.
Robert Sapolsky’s well-known observation is that a zebra outruns the lion and then stops. Its stress response is exquisitely designed and mercifully brief. Ours is identical in machinery and catastrophically different in duration, because we can switch it on by thought alone and leave it running for weeks. We are the only animal that can be exhausted by something that has not happened.
Current neuroscience frames the brain less as a device that reacts to the world and more as one that continuously predicts it, then corrects itself against what arrives. Prediction is metabolically cheap when it resolves quickly. It becomes expensive when it is held open. An unresolved prediction is not filed away and forgotten; it stays live, consuming resource in the background, in much the way Bluma Zeigarnik observed a century ago when she found that waiters recalled unpaid orders vividly and forgot them the moment the bill was settled.
Late August is a month of unpaid orders. Everything is on the tab and nothing has been settled.
Bruce McEwen’s work on allostatic load makes the same point in the language of physiology. Acute stress is not what damages us. What damages us is the failure to switch the response off – the low, sustained hum of a system that has been told something is coming and has not yet been told what. Two weeks of that leaves people genuinely tired, and then puzzled by their tiredness, because on paper they have been resting.
There is a striking piece of evidence for this. In a study published in Nature Communications in 2016, participants turned over virtual rocks, some of which concealed a snake that delivered a mild electric shock. Their stress was measured through pupil dilation and skin conductance as the odds shifted.
Stress did not peak when a shock was certain. It peaked when the probability sat at around fifty per cent. The people who knew for certain they would be shocked were calmer than the people who did not know. Uncertainty, not adversity, produced the strongest response.
Late August is a fifty per cent month. You know roughly what September contains. You do not know exactly when it will make its first real demand, how hard the first fortnight will be, whether the thing you have been half-dreading will turn out to matter. The mind cannot close the file, so it keeps it open, and keeps paying for it.
In 2010, Matthew Killingsworth and Daniel Gilbert sampled the moment-to-moment experience of thousands of people and found something uncomfortable: minds wander roughly half the time, and wandering minds are reliably less happy than present ones. What mattered was not what people were doing. It was whether they were there while they did it.
This is why late August so often fails to restore. There is more free time available than at any other point in the year, and less of it is being lived. People find themselves physically on a beach and mentally in a meeting scheduled for the third week of October. The rest is available; it is simply being spent in a room they are not in.
Which is the actual difficulty. Not that there are ten days left, but that they are being held as a corridor rather than a room.
For autistic and ADHD adults, and for neurodivergent children and their families, this fortnight is rarely a matter of mood. It is a matter of load, and it is worth naming precisely, because it is so routinely misread as reluctance.
The concept of monotropism describes attention that runs deep rather than wide – pulled into a small number of interests with real intensity. It is an efficient way to think and an expensive way to change direction. Six weeks of summer is long enough to settle fully into an attention tunnel. September requires exiting it, on a fixed date, on somebody else’s schedule. The cost of that switch is real, and it is not proportional to how much anyone likes or dislikes school or work.
For ADHD, the difficulty is usually structural. Self-regulation leans heavily on externalised scaffolding – timetables, deadlines, other people’s expectations, the shape a week is given from outside. Summer removes almost all of it. September reinstates the whole apparatus at once. And because time in ADHD is frequently experienced as now or not-now rather than as a measured quantity, ten days does not feel like a countdown. It feels like nothing at all, right up until it feels like an emergency.
What families often read as not wanting to go back is, more often, the anticipation of switching – which is a different problem, and a solvable one.
Anticipatory load also shows up in the body before it shows up in language. Interoception — the sense of one’s own internal state – tends to be less finely calibrated in autistic people, which means the first available evidence of dread may be a stomach ache, poor sleep or an unexplained shortening of temper, several days before anyone can say what it is about. Taking those signals literally, rather than looking for the emotion behind them, usually gets further.
None of this argues for enjoying the last fortnight more. It argues for stopping treating it as a waiting room.
1. Give the not-yet a boundary.
Decide when September starts for you and write it down — a date, not a vague sense. Uncertainty is the expensive part, and it can be reduced by decision even when it cannot be reduced by information. A named threshold turns an open file into a closed one.
2. Rehearse exactly one thing.
Prediction error falls with data. Open the diary and look. Do the school run route once. Answer three emails. One concrete contact with September does more to settle the system than a fortnight of thinking about it, because thinking generates no new evidence.
3. Move the wake time before the deadline, not on it.
Body clocks shift by roughly an hour a day, and they respond to morning light and wake time far more than to bedtime. Starting a week out costs very little. Starting on the first morning of term costs a fortnight.
4. Separate resting from waiting.
They look identical from outside and are entirely different inside. Resting is being here with nothing required. Waiting is being somewhere else, on standby. If an afternoon leaves you more tired than it found you, it was probably the second.
5. Lower the stakes on the first week.
The first week of anything runs on novelty and adrenaline, and tells you almost nothing about how the season will go. Judge it in week three. Decisions made in the first fortnight of September are made on borrowed energy and unreliable information.
6. Notice if it is disproportionate.
A fortnight of low-grade apprehension before a return is ordinary. Sleep that has not been right for weeks, dread that does not lift when the thing actually starts, or an autumn that arrives the same way every single year, are patterns rather than seasons — and patterns respond well to being looked at properly.
There is nothing wrong with a threshold. Every culture that has lasted has built rituals for them, and the rituals are almost never about speeding the crossing up. They are about marking it — making the in-between into a place rather than a gap, so that it can be occupied instead of endured.
We have largely stopped doing this, and late August is what we have left of it: an unmarked ten days that everyone experiences and nobody names. Naming it is most of the work. The rest is deciding to be in the room you are actually standing in.
September will begin whether or not you spend August braced for it. It may as well find you rested.
If anything here felt closer to the bone than a seasonal dip — sleep that has not settled for some weeks, dread that does not lift once the term or the work actually starts, or the same difficulty arriving at the same point every year — our team is available across our London practices and online. Individual psychotherapy, psychological assessment, and support for autistic and ADHD adults and young people.
