A stress-and-recovery report from a wearable ring covers three consecutive days in May: high-stress minutes at 150, 150, and 105, recovery minutes collapsing to 105, 30, and 15 across the same span. The ratio of stress to recovery time flips five- to ten-fold across those three days. The ring’s own classifier labels all three “Normal.”
Nothing in the classifier is wrong, exactly. Each day’s numbers sit inside whatever range the device calibrated as non-alarming for this wearer’s baseline. The label just answers a narrower question than the one it appears to answer. It says “not yet an outlier” and gets read as “you are fine.” Nobody built it to lie. It never claimed to be anything but a threshold check, and the wearer read it as a verdict.
Compare Nikolai Evreinov, the Russian theater director who spent the 1910s designing what he called the Theatre for Oneself: private, invented scenes staged for an audience of exactly one. His best-documented example wasn’t even his own. A historian named Mykola Kostomarov, grieving a canary his cat had eaten, put the cat on trial. Kostomarov played prosecutor and recruited friends to serve as judge and defense counsel; his family became the jury (his mother refused, on the grounds that her son had gone mad). Fifteen minutes of deliberation; the cat walked. Evreinov’s other favorite: stage your own convalescence. Recruit two friends to play doctor and nurse for a couple of days. Lie in bed, take the soup, have the newspaper read to you, get told by the “doctor” that you’re on the mend but should stay down one more day. “Isn’t it charming,” Evreinov asked, “when people nurse you and take care of you with a truly touching attention… when everybody’s thoughts seem to centre around your dear self?”
Both cases run the same procedure at the top level. Something formless and private — grief, fatigue, the need to matter for an afternoon — gets converted into a legible, externally modeled artifact: a verdict, a doctor’s note, a numeric score, a “Normal.” The artifact is what makes the interior state real enough to act on, discuss, or simply enjoy. Neither the trial nor the readiness score changes anything about the cat or the cardiovascular system underneath. What changes is that the wearer now has something outside themselves to point at.
The two procedures split here — not in accuracy, but in authorship, and in what authorship does to the possibility of betrayal.
Evreinov’s fiction cannot lie to Kostomarov, because Kostomarov wrote it. He knew, at every stage, that the trial was invented — that was the entire premise, stated up front to every participant, including the jury that convened for fifteen minutes and returned exactly the verdict the format was built to produce. A convalescence staged by two friends can’t reveal that you were secretly dying the whole time, because nobody involved was trying to discover anything. The theatre for oneself has no epistemic ambitions, so it cannot be wrong — it never claimed to be finding out something true.
The ring’s “Normal” is built from a different claim. It presents itself as discovery, not invention: a measurement of what the body is doing, arrived at from data, not agreed to by consenting participants. A claim to discovery is a claim that can misfire in a way an admitted fiction structurally cannot — it can report the wrong thing about a real situation while sounding exactly as authoritative as it would if it were right. The label “Normal” sits on top of a five-to-tenfold swing in the underlying ratio and reports calm. Nobody agreed to that framing the way Kostomarov’s jury agreed to acquit. The wearer just trusted the word.
This isn’t an argument that wearables are miscalibrated, or that self-quantification is a con. The classifier did precisely what it was specified to do, and the raw numbers sat right there in the same report, available to anyone who read past the label. The point is narrower and stranger: the vulnerability to being misled tracks who’s claiming authorship, not who’s being more careful. Evreinov’s staged illness is immune to a certain kind of failure for the same reason it can never teach you anything you didn’t already know about your body — it was never in the business of finding out. The ring is in that business, which is exactly what exposes it to the one failure mode fiction cannot have: being wrong while sounding like fact.
One version of this is testable without touching a single sensor. Keep the same numbers, change only the framing: swap “Normal” for something that names the classifier’s actual operation, like “below alert threshold.” If the argument above is right, wearers shown the threshold-framed version should be more likely to open the underlying data when a felt sense of stress contradicts the summary, because the label no longer claims to have already settled the question. That’s a measurable behavioral difference, and producing it would cost a vendor nothing more than a different word.
It would be tidy to conclude that invented frames deserve more trust than measured ones, or that “Normal” needs air quotes from now on. Neither holds up. Kostomarov’s cat trial produced no information the household didn’t already possess before the gavel; it produced feeling, deliberately and honestly, at zero risk of factual contradiction. The ring produces something closer to information, at the price of being contradictable, silently, by its own underlying numbers. Trade one kind of vulnerability for a different kind of value, and the safety of the first one doesn’t come along for free. The theatre for oneself was never going to catch a real illness; the instrument that might catch one is also the only one of the two that can lie without knowing it.