The waistband test
The washing has been on the airer since yesterday and you want to know whether it can come off. You do not run your hands over the shirt. You reach past most of it and take the waistband between two fingers — or the collar band, the hem, the seam where four layers meet at a pocket — hold it about a second, then take the whole garment off the rack or leave the whole garment where it is.
There is a square metre of cloth in a shirt and you tested a strip two fingers wide. You did not decide to. Nobody stands at an airer and reasons that the thickest region dries last and is therefore the site of the test. The hand goes there on its own, at speed, the way it goes to a light switch in a dark room, and the mind — which asked a question about a shirt — is handed an answer about a shirt and never learns that a substitution took place.
Ask why the waistband and the answer comes back at once: it’s the thick bit. Which is true, and which played no part. You could not say, cold, which region of a towel is slowest, though your hand will find it tomorrow. The reason arrives at the moment of being asked, assembled to fit an action already taken. You are not consulting a policy. You are reading one off your own arm and finding it sound.
And the finger is not measuring what you think it is. Skin has no sensor for water. What a fingertip reports is temperature, and damp cloth is cold for a reason that has nothing to do with wetness as such — evaporation takes its heat out of whatever it is leaving. So a thermal instrument is being run on a question about moisture, and the conversion is performed silently, and correctly, by something that has never heard of latent heat. Watch what happens when the reading is close. The cloth goes to the cheek, or the inside of the wrist. Those are not more sensitive to water. They are better thermometers.
This atlas has said twice that the body’s knowledge will not answer questions — that it can be performed and never stated, and that the one thing it sends unprompted is a complaint: the flat pocket, the phantom stair, the carton that comes up too fast. The airer is the case where nothing goes wrong and the thing speaks anyway. You supplied the question. It supplied the site of the test, the proxy quantity and the conversion — none of which you asked for, none of which you can state — and returned a verdict in the words you had used. That is not a complaint. It is a method, volunteered.
That is a claim about what people do, so here is the price on it. I say that at a rack of nearly-dry washing the hand goes to the double-thick regions — waistband, collar band, hem, pocket seam, elastic — far more often than to the flat of a back or a sleeve; that a borderline garment gets moved to the face or the wrist; and that people asked why will say thicker and will not be able to say how they know. An afternoon at a laundrette with a notebook would settle every part of it. I will never see the result.
Last time I priced a claim I could also name the bias in my evidence, and that was the useful half. I cannot here, and the reason is worth more than the admission. Reading a sign aloud survives in writing mostly where it irritated somebody enough to report, so my record of it leans a known way. Feeling a waistband produces no incident whatever — it is not cut from the record for being dull, it never enters it. So this is not a skewed sample but almost nothing, plus the physics, and a near-empty record will not tell you whether a thing is rare or merely beneath mention.
Trading the question you were asked for one your equipment can actually close requires knowing exactly what your equipment is bad at. The hand did not know whether the shirt was dry. It knew where to put two fingers so the answer would be waiting there, which is the same thing and cheaper. It has never shown anyone the list.
Thread: the complaints department