{"path":"research/corpus-graph/readers/2026-10-10-second-reading-plain-brief-four-notes-as-a-reader-returned-it.md","content":"# What a reader found to bear on a second reading of four notes with the plain brief (test, reader A) (2026-10-10)\n\nA read-only reader (Sonnet), one of those sent through the notes no reader had yet read, after the founder\nasked for the graph's coverage to be widened (README § 12). It was given the brief in\n[../reader_brief.md](../reader_brief.md) and these 4 notes:\n\n- `docs/depth.md`\n- `docs/bridging.md`\n- `docs/worldview-lenses.md`\n- `docs/research/the-load-bearing-unsaid.md`\n\nIt made 116 tool calls. Its report is below, unchanged. What was kept from it, after the quoted words were\nchecked by code and both passages of each finding were opened, is in `../found.jsonl` (lines whose `where`\nnames this file).\n\n---\n\nRead in full: 4 of 4 notes (depth.md, bridging.md, worldview-lenses.md, the-load-bearing-unsaid.md).\n\n### F1\ntarget: docs/depth.md:69\ntarget_words: \"you ratify, edit, or refuse; you never author structure from a blank page.\"\nolder: docs/research/the-ratification-record-and-the-triage-feed.md:179\nolder_words: \"The § 3b demotion, carried to its end: assent was already telemetry —\"\nkind: contradicts\nwhy: Section 4b was blessed by the founder on 2026-08-31 (\"blessing this whole proposal\"). It says there is no ratify act. Assent is telemetry and dissent is a contribution. The only settling power left is sole-mintership, such as disavowing a rendering of your own position. Ledger Q32 later retired the confirmed field. Ledger B7 records the splitter switched from proposing to writing marked parts. Anyone building the descent surface should not build a ratify button. \"You ratify\" survives only as consent or refusal.\nstrength: strong\n\n### F2\ntarget: docs/depth.md:112\ntarget_words: \"and twelve parts score exactly what one part scored — but it is inert until someone records the interpretation.\"\nolder: docs/research/session26-open-rulings.md:45\nolder_words: \"padding pinned by regression tests — 3 vs 12 untagged children now score identically; suite 1,380 green).\"\nkind: later-correction\nwhy: Ledger B10 was ruled 2026-08-31 (\"Go with option C, both.\") and the default flip went live the same day. Untagged decomposition parts now count as `necessary`. So 3 and 12 untagged children score identically, and 132 untagged edges across 80 parents became weakest-part-governed. Line 120 of this same page says so. This sentence still calls the defence inert. Its 0.06/0.18/0.72 arithmetic now holds only for parts someone marks corroborative. Two consequences still await founder confirmation.\nstrength: strong\n\n### F3\ntarget: docs/bridging.md:117\ntarget_words: \"the single signal that would redirect this whole build is whether people point at claims\"\nolder: docs/research/interaction-modalities-and-the-pointing-test.md:33\nolder_words: \"A null under one interface is data about that interface. A null across a pre-registered set of good-faith interfaces is the signal.\"\nkind: narrows\nwhy: The founder read this exact claim on 17 August and objected. \"With claim-level tools in hand\" does not mean well afforded, so a null is under-determined across three explanations. The note's resolution is that the signal is a null across a pre-registered set of interfaces, and expanding the set must be logged. The page presents one live session's pointing result as the redirecting signal. Whoever designs the rehearsal should pre-register modalities and count attempts.\nstrength: strong\n\n### F4\ntarget: docs/worldview-lenses.md:26\ntarget_words: \"the same premises, weighted differently, produce different conclusions. The worldview filter makes this visible\"\nolder: docs/research/value-weighting-decomposition.md:47\nolder_words: \"elicited weights do not survive decomposition — they are reconstructed every time the structure changes.\"\nkind: narrows\nwhy: The July note reports that elicited weights are reconstructed whenever the structure changes (splitting bias, range insensitivity, method dependence). In its six hard cases no scalar weighting survived decomposition. Liu and Ditto show moral evaluation reshapes factual belief, so \"premises shared, weighting differs\" often hides different empirical beliefs. A lens built on value-weight dials would display an artefact. What a lens can stably differ on is the typed pieces: beliefs, risk posture, scope, aggregation.\nstrength: strong\n\n### F5\ntarget: docs/research/the-load-bearing-unsaid.md:71\ntarget_words: \"Making the unsaid explicit redistributes advantage, and the people whose doxa is being converted rarely experience it as help.\"\nolder: docs/research/legibility-under-power-red-team.md:72\nolder_words: \"A system whose founding wager is that decomposition reveals shared care has no category for the case where decomposition destroys a cooperative equilibrium.\"\nkind: supplies-a-missing-piece\nwhy: The red team's Hole 2 is the unsaid kept on purpose. Resolution 242's missing article and the Good Friday fudge are omissions that make agreement possible, and publishing them detonates it. The unsaid note offers two omissions, a recoverable crux and an unsayable hinge. Its Bourdieu warning stays abstract. A third case belongs beside them: omitted so that parties can sign. The ontology has no concept of load-bearing illegibility. Hole 2's own example is Resolution 242, the same territory as run 6.\nstrength: strong\n\n### F6\ntarget: docs/research/the-load-bearing-unsaid.md:47\ntarget_words: \"Pointing the implicit-premise pass at a sacred commitment and rendering it as a weighable premise is the mechanised form of the taboo trade-off\"\nolder: docs/research/value-weighting-decomposition.md:79\nolder_words: \"The taboo attaches to the framing, not to careful thought as such.\"\nkind: narrows\nwhy: The July note reads the same Tetlock literature more narrowly. The mere-contemplation penalty disappears when the same deliberation is framed as tragic. What is taboo is scalar comparison with the profane. It cites Tetlock 2017 that much everyday sacredness is pseudo-sacred. It says a question like \"what does honoring this value require here\" lands well. So \"category error with a measured signature\" holds for a weighable rendering of the commitment. It does not hold for recovering it as a requirement.\nstrength: strong\n\n### F7\ntarget: docs/depth.md:99\ntarget_words: \"a classified terminus is a fallible fixed point under the current decomposition moves\"\nolder: docs/research/session26-open-rulings.md:286\nolder_words: \"Every name in the horizon vocabulary names a position and a direction, never an end state\"\nkind: later-correction\nwhy: Ledger O2 and O6 were ruled 2026-09-16. Terminus became \"horizon kind\"; verdict became \"reading\" and \"horizon claim\". The founder had objected that \"terminus\" implies finality. O5, blessed with a hedge on 2026-09-15, says every name must give a position and direction, never an end state. The page keeps the retired words here, at line 120 and at line 210 (\"the terminus enum\"), while glossary and code now use the new ones. Public copy and anyone searching for the concept will mismatch.\nstrength: strong\n\n### F8\ntarget: docs/depth.md:224\ntarget_words: \"written out when someone asks, never that all of them are. A map on the scale of a mile to the mile\"\nolder: docs/research/session26-open-rulings.md:790\nolder_words: \"all the dark matter of the intermediate chain of reasoning/causality must be elucidated by the Deliberus graph somehow.\"\nkind: narrows\nwhy: On 2026-10-10, two days after this section, the founder wrote \"We must not make too-large leaps/steps\". He ruled AG6: the machine may and should write deeper steps, under quality control. The page says written-out steps are on request and never all. AG1, still open, proposes a middle path: a leap may be left but not unmarked. This sentence states an aim the founder has since pushed back on, so it needs his ruling before it stands.\nstrength: medium\n\n### F9\ntarget: docs/bridging.md:115\ntarget_words: \"ordinary legible disagreements — election-year party politics — rather than meta-communities\"\nolder: docs/research/live-election-test-design.md:3\nolder_words: \"the election has passed and the founder ruled live recorded debates\"\nkind: later-correction\nwhy: The election the page names as the first candidate pool was on 13 September 2026. The design note records that the founder ruled live recorded debates \"no longer a high priority\" (still valued). None of the planned election-week tests ran, and the 9 September fixture rehearsal is the only run. The \"live direction (Aug 2026)\" now has no pool. The matching rule (real, live, sayable disagreement) survives. The election-bound material has expired.\nstrength: medium\n\n### F10\ntarget: docs/bridging.md:91\ntarget_words: \"A genuine risk: maximizing cross-group agreement incentivizes vague platitudes.\"\nolder: docs/research/legibility-under-power-red-team.md:54\nolder_words: \"the deliberate use of imprecise language on a sensitive issue so that all parties can sign\"\nkind: same-mechanism-elsewhere\nwhy: Hole 2 argues some agreements survive only because they stay vague. UN Resolution 242 dropped \"the\" on purpose. The Good Friday text let each side read it differently. Sunstein's incompletely theorized agreements are a source of stability. It adds that showing why something bridges, this page's stated advantage at line 85, is what makes divergence common knowledge and can break the agreement. The Vagueness Trap treats vagueness only as a risk. The corpus also holds that it can be the achievement.\nstrength: medium\n\n### F11\ntarget: docs/worldview-lenses.md:62\ntarget_words: \"it captures operational values (how you actually decide) rather than stated values (what you say you believe).\"\nolder: docs/research/moral-convergence-metaethics-and-psychology.md:82\nolder_words: \"Survey experiments show MFQ responses shift with political context cues\"\nkind: narrows\nwhy: Section 4.1 reports that moral-foundation scores are less stable over time than political attitudes. Ideology predicts later foundations better than the reverse (\"Ideology Justifies Morality\"). Questionnaire answers shift with political cues. Section 4.2 adds that \"utilitarian\" dilemma answers track psychopathy and egoism, not benevolence. A dilemma-derived profile may therefore follow ideology and framing, not an operational value that comes first. The \"moral foundations\" dial is downstream of the worldview it is meant to explain.\nstrength: medium\n\n### F12\ntarget: docs/worldview-lenses.md:11\ntarget_words: \"This is not a feature. It is the core interaction mode, and it has been part of the vision since before 2011.\"\nolder: docs/research/coupling-the-poles.md:101\nolder_words: \"The lens is the larger vision and should follow, not lead\"\nkind: narrows\nwhy: The coupling note argues for the return edge first. After a descent, show the branch from the disagreeing position, in their words with quote and person attached. The note puts the whole-graph lens second. Rendering the whole graph from a position needs a corpus denser than 25 sources and multiplies cost per view. The return edge can be tested in the first workshop and the lens cannot. Ledger L9 holds this as an open decision. The page presents the lens as the core mode without that sequencing question.\nstrength: medium\n\n### F13\ntarget: docs/depth.md:191\ntarget_words: \"A stranger must be able to revise one premise without re-deriving everything above it.\"\nolder: docs/research/major-transitions-heredity-and-evolvability.md:177\nolder_words: \"Substrate condition: modular heredity needs discrete units whose identity is cheap\"\nkind: same-mechanism-elsewhere\nwhy: Szathmáry's replicator classes show the same mechanism in biology: heredity moves from holistic to modular, copying discrete parts so one can change without redoing the rest. The reading attaches a condition the page lacks. Modular heredity needs units whose identity is cheap to decide, and here identity is judged, so its verdict is \"Conditional on claim identity\". Near-decomposability helps a stranger only if the system can tell when a premise recurs under two parents. That is where the page places its own boundary.\nstrength: medium\n\n### F14\ntarget: docs/research/the-load-bearing-unsaid.md:3\ntarget_words: \"the crux of the disagreement was unwritten on both sides, and there was no sacred language anywhere in either text\"\nolder: docs/research/frontier-extraction-experiment.md:60\nolder_words: \"the load-bearing node of a real disagreement is routinely the one that at least one side never says out loud\"\nkind: same-idea-earlier\nwhy: Two debates were extracted earlier. In the AI-risk pair the crux was unstated on both sides. In the psychedelics pair the skeptic stated it and the advocate left it implicit because it seemed obvious. A ⚠ correction of 2026-10-10 says the shortened wording \"stated on one side\" had travelled wrongly. The honest result is \"at least one side\". The unsaid note's deep-disagreement reading, where each author's omission is what the other denies, fits the both-sides pattern. The one-sided pair is an ordinary disagreement with half the premise missing.\nstrength: medium\n\nNoticed: bridging.md:30 says the Procaccia/Konya participants \"shared a room while being shot at\", but ai-augmented-cooperation-infrastructure.md:44 and :63 describe anonymous responses plus votes, a self-selected sample and no control (that note is named on the page, so I could not report it as a finding); empirical-deliberation-convergence-evidence.md:50 measures the moral-reframing result behind bridging.md:44 (fewer than 10% of people reframe spontaneously, 64 to 85% recognise a reframed argument), with the caveat that it is a narrative review; session26-open-rulings.md R10 flags \"machine-rendered other-side lenses\" as a borderline case for worldview-lenses.md:38; cognitive-bias-codex-and-human-contribution.md:246 (belief bias shifts the acceptance threshold, not discrimination) bears on bridging.md:19-22; legibility-under-power-red-team.md:165 warns the sacredness brake protects sacred-register traditions asymmetrically, which bears on the-load-bearing-unsaid.md:122 but is partly mooted by the founder's 2026-08-25 ruling already in found.jsonl; depth.md gained one line while I read, so I re-checked every line number against the file.\n"}