Outside Signs

You look at your wrist tracker in the morning. It says you got 54 minutes of deep sleep. The question is not whether that number is accurate. The question is what you are actually trying to measure, and why — because the answer changes everything about how useful the device can be.

WHOOP measures proxies. Heart rate, heart rate variability, respiratory rate, movement. From these it infers sleep stages through a proprietary algorithm. Deep sleep — slow-wave sleep, N3 — is fundamentally an EEG phenomenon, a pattern of electrical activity across the cortical surface. You cannot directly measure it from the wrist. What you get is a shadow of something real, cast at a distance, through several layers of inference.

The 54 minutes might be 38. It might be 71. The device cannot tell you which. This is not a reason to throw it away. It is a reason to use it correctly — as a trend indicator across weeks, not a precise measurement on any given night. The direction of change over time, with variables isolated, is meaningful. The number itself is not.

Caffeine, cannabis, chronic stress — all suppress or distort sleep architecture. The problem is they rarely arrive alone. Most nights carry several inputs simultaneously, which makes the signal unreadable. You cannot understand what any one variable is doing until you have a clean baseline to measure against.

This is what a caffeine taper is actually doing when approached seriously. Not wellness optimisation. Basic experimental hygiene. Strip the inputs back, let the system stabilise, then reintroduce variables one at a time and watch what the trend does. The tracker becomes useful precisely when you treat it this way — as an instrument in a personal experiment rather than a health score to be maximised.

The standard picture of sleep and recovery treats them as a block appended to the end of a day. You work, then you recover. The recovery happens at night, in a defined window, producing a score. This framing is wrong in a way that matters practically.

The body does not experience a day as a block followed by a recovery block. It runs a continuous curve of stress accumulation and discharge across the full 24 hours. What arrives at the sleep window is not a fixed quantity waiting to be processed — it is whatever has accumulated since the last genuine discharge point. Two days with identical sleep can produce very different recovery if the daytime stress curves were different.

A 20-minute yoga nidra session is not 20 minutes of benefit. It is a reset point. The intervention is the divergence in trajectory after it — the difference between two accumulation curves, one that resets mid-afternoon and one that runs continuously from morning to midnight. That difference, integrated across the full day, is what arrives at the sleep window. It is the area under the curve that matters, not the duration of the practice.

This reframes what rest practices are doing entirely. The question is not how much recovery a nap produces in itself. It is how much lower your accumulated load is by the time you lie down — and therefore how cleanly your sleep architecture can do its actual work.

Every culture that lived close enough to its own biology arrived at the same answer. Distributed rest across the day. Not as a health practice, not as a philosophy — as a direct response to biological signal that was still audible.

These are not related traditions. They did not borrow from each other. They are independent convergences on the same biological solution, arising wherever people had not yet been required to override the signal.

There is a temptation to place high-stress environments on one end of a health spectrum and low-stress Mediterranean ones on the other — hardship versus ease, north versus south, the brutal and the graceful. This framing misses what actually matters.

A Norse winter — cold, hunger, genuine physical danger, hard labour — is a high-stress environment. A Mediterranean summer — warmth, full stomach, afternoon rest — is a low-stress environment. But both are T1 in the framework's sense. Both have coherent signals and genuine resolution. The Viking stress profile is intense and episodic: the raid, the storm, the hunger — followed by the feast, the hall, the long dark winter rest. Hard signals the body evolved to receive, followed by genuine discharge. The Mediterranean profile is moderate and distributed: small stressors absorbed by embedded recovery throughout the day.

Different oscillation frequencies. Both within biological range. What they share is the thing that actually defines T1: the signal is legible, and the recovery follows it.

T4 is not harder than a Viking winter. It is illegible. The body keeps waiting for a discharge signal that never arrives.

Chronic low-grade psychological stress does not resolve the way physical stress resolves. There is no feast after the open-plan office. No winter rest after the quarterly review cycle. Artificial light removes the circadian anchor. Caffeine suppresses the adenosine signal that would otherwise enforce rest. The body is not overwhelmed — it is confused. It keeps generating a readiness to respond that has nowhere to go, and the accumulated unresolved load drifts upward across days, weeks, years, below the threshold of conscious awareness.

Pre-industrial European life was not a golden age. Cold, disease, food insecurity, genuine violence — the total stress load was enormous. But the stress profile was different in kind, not only in degree. The canonical hours structured the civilian day with multiple pauses. Work was seasonal. Longer summer days meant earlier starts and genuine midday rest during peak heat. The body and the light were still in rough correspondence.

The shift happened gradually through the 16th to 18th centuries. Artificial lighting extended the evening. This pushed sleep later, which compressed the night, which eventually collapsed whatever pre-industrial rest patterns existed into a single efficiency-optimised block. The Protestant Reformation added the ideological layer — idleness as moral failure, continuous productive wakefulness as virtue — but the lighting change was probably the more fundamental driver. The ideology arrived to justify and accelerate what technology had already started.

What replaced these patterns was not a better system. It was a Northern European Protestant productivity model exported globally — fixed hours, short lunch, no midday rest — removing climate adaptation, circadian accommodation, and distributed recovery in a single move. The populations who lost most were the Jhapakee workers: manual labourers in hot climates who now work through peak heat without the shade break, accumulating load the evening cannot fully clear.

Polysomnography — the gold standard against which every wrist tracker is measured — records surface EEG, eye movement, muscle tone, respiratory and cardiac signals. It classifies sleep into stages by their external electrical signatures. This classification system is clinically useful. It is not the same as understanding what is happening inside the cell during those stages, or what drives the transitions between them.

The staging categories are a shadow-map. N1, N2, N3, REM — correlates of states, not the mechanisms producing them. We are very good at recognising the shadow. We have considerably less certainty about what casts it.

These are not processes we can measure from outside the skull with an electrode on the scalp, let alone from a wrist with a photoplethysmography sensor.

Melatonin adds a further layer of complexity. It is not only a pineal hormone signalling darkness to the brain. It is produced locally in many tissues, and then:

Whether a deep parasympathetic state — yoga nidra, genuine rest, a mid-afternoon siesta — can produce some component of that mitochondrial response independently of PSG-defined sleep stages is not known. The mechanisms overlap directionally: reduced sympathetic tone lowers oxidative load, shifts cellular redox environment, reduces the conditions that favour aberrant condensate formation. Whether the overlap is functionally significant or only resembles the real thing is genuinely open. The research at this intersection does not yet exist.

You started with a device on your wrist producing a number. The number is a weak signal of something real — a proxy of a proxy, an outside sign of inside processes we do not yet fully understand. Used correctly, as a relative trend indicator across weeks with variables isolated, it tells you something. Used as a score, it tells you almost nothing.

What the number is trying to point at is whether your nervous system had enough genuine discharge built into the day that it could do what it evolved to do when you finally lay down. Whether the cellular environment your mitochondria entered at sleep onset was one they could actually work with — low enough oxidative load, low enough unresolved stress signal, low enough accumulated condensate burden — that the night's maintenance could run cleanly.

The Hadza labourer lying down after genuine physical work under an open sky, the Mediterranean worker finishing a meal in the midday heat, the woman in Delhi finding shade for twenty minutes between shifts — none of them needed a device to tell them whether they had recovered. The signal was still audible. The body knew what it needed and the day had space to answer.

The wrist tracker exists because we have moved far enough from that coherence that the signal is no longer reliably audible without instrumentation. It is a compensation device for a broken feedback loop — useful, necessary even, but downstream of the actual problem. The actual problem is a day architecture that accumulates load continuously, removes discharge points systematically, and then asks a single overnight window to clear what 16 hours of T4 exposure produced.

No algorithm running on wrist-worn photoplethysmography can fix that. But it can show you, over time, whether you are moving toward it or away from it. Which is, perhaps, enough to start.