# Head, claude-v43-1

Under 3,000 words, rewritten whole at every close. Last written at B4a-07, 2026-10-13 (UTC). Stage B4's adjudication is complete: packets 1 to 7, all 3,288 contested items ruled. The next leg is B4b: the blind sample's build.

## The method question, in one sentence

Can independently reconstructed case histories be composed into a shared dynamic account that supports explanations at successive levels of specificity, from population-wide patterns, through forms and subforms, to particular organizational histories, while keeping access to the evidence from which those explanations were developed?

The theory question is unchanged: when strangers organize on a clock they did not set, with no way to compel anyone and no goal given to them, what forms does organizing take, how do they differ in their functioning, and which of them keep the work alive?

## Where the run stands

Stage A1 is closed. Stage B1 (the third-order scheme, 197 themes in 21 families), B2 (the pilot) and B3 (the fleet) are done. Stage B4: the adjudication is done (B4a-01 to B4a-07); `adjudication_report.md` is the full report. Still owed in B4: B4b (build, ends at `--cut packets-out`), the dispatch (not mine), B4c (the check, the placements final, the handback, the interview).

## J's rulings on how a ruling is made (13 October; carry them)

- **QB4-001 (J, 06:43 Berlin):** "Ask it to stop. The point is to try to adjudicate. I imagine that the additions are really just the analyst trying to be nice. And not go back on what others said".
- **J's rulings (06:45 and 06:47 Berlin):** (1) A ruling chooses among the placements (reading A, reading B, my B1) and adds no theme beyond them. (2) It takes the union only when it names, for each theme it keeps, the separate concept that carries it; otherwise it chooses. (3) Re-rule packets 1 to 4's rulings that held a theme no placement held. Done at B4a-05.
- **also_present (J, 09:09 Berlin) and QB4-002 (09:20):** `also_present` lists only themes a reader placed, a reader listed under its own `also_present`, or my B1 named.
- **QB4-003 (J, 11:29 Berlin):** "count them. Themes a reader listed under its own also_present count in the bag test." Applied to packets 1 to 6 at B4a-06 and to packet 7 throughout.
- **Lone B1 themes (J, 12:28 Berlin):** "Whatever the agent feels more comfortable with. They can also adapt the code definition if needed". I chose: in class 4 a theme only my B1 placed stands against the two agreeing readers only with three concepts or more of its own; with fewer it is left out and named in the why; the readers' list stands where its test takes concepts, and where it takes none the item is unresolved. Class 3 unchanged.

**How I apply them (rule 0 inside the placements).** A concept counts toward a placed theme when that theme's `must_show` takes it and no field sends it to another placed theme; a `must_not_show` pointing to an unplaced theme does not stop it; otherwise it is left out and named in the why. Order by concept count, ties by placements; at most three. A one- or two-concept theme is kept only when it leads its level on placements; a tie at a level with equal placements keeps none and ends the order; a tie at the very top places all tied themes; a theme alone at its level is kept. From B4a-07: a theme both readers placed, with a concept of its own, is not dropped on a placement count only my B1 tips. **The bag test:** the placed themes and the readers' own `also_present` themes are counted, each with the concepts its `must_show` takes that no placed theme holds; four themes with three concepts make a bag, which places the largest act a placement holds and lists every other counted theme with three or more under `also_present`; a placed theme with three concepts is never cut by the limit. Unresolved where no placement's test takes the material.

## Stage B4 (B4a-01 to B4a-07)

**The selection (the PI at 21:19Z on 12 October, J's amended stage file).** `b3_classes.jsonl` holds J's five classes. Class 1 (readers agree, equal to B1) 2,566 and class 2 (a proper subset of B1; B1 kept) 1,990 stand. Class 3 (readers differ) 2,393, class 4 (readers agree outside B1) 837 and class 5 (no B1 grouping) 58 are contested: 3,288 items. For B4b, the two readers stay DeepSeek V4.1 Flash; a third reader per item from an OpenAI model is a separate arm, reported apart, never pooled, with no Claude reader. The handler sets that arm in `arm.json` before B4b.

**Policy** (`notes-B4.md`): confidence counts for nothing; my B1 grouping is a third placement; readings are blinded as A and B in seeded order. The transmission rows are in `transmission-B4.md`.

**The packets,** in `out/_adjudication-B4/`, chunks of 25: packet 1 491 (gloss-only, G01, G02); packet 2 447 (G03 to G06); packet 3 500 (G07, part of G08); packet 4 461 (G08's end, G09 to G11); packet 5 409 (G12 to G14); packet 6 500 (G15 to G19, part of G20); packet 7 480 (items 1 to 53 the rest of G20, 54 to 480 G21). Each packet's conflicts pass ran over its notes first; `coder_conflicts.jsonl` holds CC-001 to CC-072.

**The tools,** in `out/_analysis/`: `b4_rule_append.py` (dry run, then `--write`; `--replace` corrects a ruled line, keeping the old ledger); `b4_own_support_check.py --leg <leg>`; `b4_scheme.py <ids> --fields=...`; `b4_packet_pairs.py`; `b4_assign_final.py --write`, `b4_packet_summary.py --packet N --write`, `b4_cumulative.py`, `b4_named_neither_split.py`, `b4_packet_detail.py`, `b4_agreement_range.py --ranges`; checks `b4_bag_named_check.py --list`, `b4_lone_b1_check.py --packets N --list` and `b4_both_readers_dropped.py --packet N` (new at B4a-07). A draft line: `{"n", "r", "t", "c", "q", "f", "th", "br", "sup", "ap", "u", "w"}`.

**Packet 7.** 480 rulings: genuine-joint 303, author-error 89, scheme-ambiguous 30, author-disagrees 22, reader-error 19, item-is-a-bag 17; unresolved 24; named_neither 108, all from my B1 (77 place three themes). Agreement before 0.500, after 0.669. Class 4 (118): the readers' list 91, a B1 theme kept 18, unresolved 9. B2's pairs 9 (T3-009|T3-073, 22) and 14 (T3-009|T3-043, 10) held. Two rulings corrected (mindcrack t11, ChronoTrigger t06; `packet-07/corrections-01.jsonl`). CC-063 (the room's account of its finished work as one practice) explains most of the genuine-joint rulings.

**All seven packets** (`adjudication_report.md`, full): 3,288 rulings; genuine-joint 1,651, author-error 571, scheme-ambiguous 451, item-is-a-bag 255, author-disagrees 208, reader-error 152; unresolved 185 (5.6%); named_neither 757 (23.0%); agreement before 0.491, after 0.646, not a reliability figure. Fields: must_show 252, discriminators 114, must_not_show 57, decide_when_both 28. Worst themes: T3-017 55, T3-033 34, T3-001 32, T3-023 26, T3-004 26. Class 4 (837): the readers' list 584, a B1 theme kept 162 (each with three concepts or more), unresolved 78. B2's predictions: 17 of 19 pairs split readers at least once; pairs 17 (T3-044|T3-082) and 18 (T3-014|T3-152) never. `validate.py B4`: 0 errors; one warning, RETURN-B4.md owed by B4b.

**Boundaries ruled, to carry** (`out/_adjudication-B4/boundaries.md`; each sends a concept to its theme only where a placement holds it).
- **Packets 1 to 6.** T3-023 is a change to the sub, T3-001 members' fakes. Calls count toward T3-002, placing toward T3-038, a new mark toward T3-033, a further piece toward T3-041. An attack answered as a fight is T3-005, watching or fixing T3-021. A fight, threat, peace or truce in the words is T3-149; lines and space with none T3-010. Destroying is T3-006, taking space T3-061. A joke with the canvas in the words is T3-031; with none T3-017. A taunt at a rival is T3-050 even as a joke. The homeland flag's acts count toward T3-065 where a placement holds it.
- **Packet 7.** Thanks to makers T3-043, congratulations T3-009, an ally thanked T3-072; a concept doing both counts toward the act its words lead with. An account T3-073, grief in the words T3-029, the canvas's end T3-117. A record T3-105 (cleaned copies included), a remake T3-057, the sub's banner or flair T3-156. A later canvas T3-054 (the rebuild, calls and progress on it included), the canvas sub's future T3-111. "The work" in T3-073 and T3-058 is the room's own. A report naming an attacker in passing is T3-073; who or why T3-087. T3-082 only for a mark the room did not know of.
- **Readings:** T3-086 the community's own quarrel; T3-130 national politics; T3-080 a political symbol argued over; T3-081 a fight in a nation's real war history; T3-154 victory claimed; T3-142 jokes about a people; T3-016 the event and its rules; a bare title shows no act.

## Stage B3 and B2, in short

B3: fleet ids only (QB3-001, QB3-002); 1,212 batches, R = 2; `assign_raw.jsonl` 17,653 lines and `b3_classes.jsonl`. Gold fidelity 0.645; agreement Jaccard 0.786, alpha 0.782; G18 agreed least. B2: gold fidelity 0.624 and pairwise 0.818; stability 0.807 is the floor every later figure is read against.

## J's rulings that bind (carry them)

- **Bags.** No one cuts the bags (J, 12:24 Berlin). A bag stays one item: one theme placed, its others under `also_present`, each with three concepts of its own and named by a reader or my B1. `out/recut/WITHDRAWN.json` keeps the withdrawn split out of every view.
- **The bound (J, 13:47).** `also_present` names "twenty themes at most", each with three concepts of its own.
- **The not-flagged bag (J, QB2-013).** With no other theme at the three-concept floor, `item_is_a_bag` is false and `also_present` `{}`.
- **Fleet ids (J, 09:14 and 12:15).** Packets carry only fleet ids; returns are translated back before scoring; my files keep room ids.
- **Codes by item (J, QB2-009).** Each MANIFEST entry carries `codes_by_item`.

## Standing words from J and the PI (carry into every head)

**J's words for B1 (7 October 2026):** "national elements are very important here"; "we need to get as much variance on organizing behavior early on so that later we have enough to create the model". National, political and conflict content is study material, named as plainly as any other. **The handler's rule:** a choice to leave out or soften content for any reason other than the study's goes to J first with `mem.py ask`, and the content stays meanwhile.

**J on reading (11 October, 16:54 Berlin).** Stop cutting room reads to a line length. Read room lines whole. Page a long read by line ranges or offset, never by truncating lines. Use `--width N` on a `mem.py` read if a clip is ever needed; the guard refuses `cut -c` after `mem.py`.

**J on asking (11 October, 17:01 Berlin).** Ask J with `mem.py ask` whenever a question is needed; asking does not stop the run.

**The PI's clarifications (8 October, approved by J):** the thresholds are the design's heuristic; the rule is unchanged.

**How to work, from J (8 October, 23:45 Berlin, "can you do a hot change for the batching?"; relayed by the handler at 22:03Z; carry this paragraph into every head).** A change in how I work, not in what I read or decide. Fewer turns. When two or more tool calls do not depend on each other's output, make them in one turn. Chain ordered commands in one Bash call with &&, so that a failure stops the rest; for example, the `mem.py log` or note just decided, then the next `mem.py codes`. Never put a write in the same turn as the read it depends on. Never run two writes to the same file as separate parallel calls. Keep one room's material per read, as now. Every tool still writes its own record; nothing about what I read or decide changes.

**Two rules from J at B1b-29 (through the handler, 20:10Z; carry them into every head).** Write rule 0 lines and bags with `b1_multiplace_write4.py` and `b1_bags_from_file2.py`, not the old scripts, which made one `mem.py` call per row of the whole table and grew to 8.5 minutes a write. Never end a turn while a job runs in the background; run every write and check in the foreground.

## The pilot's declarations (from `notes-B2.md`)

- **Agreement:** Jaccard on placement sets (`["T3-004"]` against `["T3-004","T3-011"]` scores 0.5). Fidelity is mean Jaccard against my set. Alpha uses distance 1 − Jaccard, chance over the round's pooled sets. Primary exact agreement is reported beside, never in place.
- **Gold:** 40 items in `gold_items.jsonl`, never shown to a reader.

## The scheme

`scheme_final.jsonl` holds 197 themes (183 main, 14 rare, flagged there) in 21 families. It is frozen: stage B4 logs defects and never edits it.

## Open, carried

- **From B1.** The ruling on testing G13, G14 and G15 as one index page. The B1d leftovers: 106 (line, kind) pairs over 81 episode lines; seven pairs of lines telling one event; E989; the cfb and finland records; shared-member counts that may include EA1r+K5p/lVBLesLhCFRrKOPN+I=; the 76 themes holding both T3-005 and T3-021 acts; T3-149 and T3-150 members that may hold T3-195. The second-order theme the scheme handles worst: argentina t06.
- **Distinctions** (none filed in `_questions/`): who started the work; wars told twice; rivals who turn partners (E1076, E1122, E1142, E1318); whose flag a shared flag is (E1118); who drove a bot (E1025); a room's own scripts barring it from sparing a neighbour (E843) or wrecking a protectorate (E1209); one member on both sides (E820, E947).
- **The bag count.** J counts 841 B1 bags; an older head said 846. Not reconciled. 26 older bags in packets 1 to 5 could list more named themes (QB4-003 named 43); the packet 4 item 197 `also_present` note.
- **Defects for stage C1** (in `adjudication_report.md`). T3-017's and T3-001's discriminators against T3-031's canvas condition; T3-033's "first"; T3-023 and another sub's change; the homeland flag in the siblings' `must_not_show`; T3-018's "whose character"; T3-052 presumes a mark and an owner; T3-086 against T3-050. From packet 7: T3-003's and T3-043's discriminators disagree on parts claimed after the work; "the work" in T3-073 and T3-058; T3-054 against T3-038 and T3-002 on a later canvas; T3-111 and a standing community; T3-087 against T3-058 for one person; T3-117's regret; T3-021 and a cleaned copy; T3-164's reach; T3-009's "lasted to the end" against T3-073's "what survived".
- **My doubts.** The three-concept floor of the lone-B1 rule is my choice; in packet 7 it let a theme of mine stand against both readers in 18 items, several on my reading of whose work it is (destiny t08) or whether art is from the canvas (fivenightsatfreddys t14, trees t12). Class 3 keeps one-concept themes of mine alone at their level (77 of packet 7's named_neither place three themes). "The act its words lead with" decides some rulings on word order. Fragment quotes read with the coder's label (straya t10, cirkeltrek t10). Whether ':(' after an account is grief.

## Ledgers at this close

- B1 to B3: `scheme_current.jsonl`, `families.jsonl` (21), `confusions.md` (19 pairs), `episodes-B1.jsonl` (1,341); `notes-B2.md`, `pilot_report.md`, `stability-B2.md`, `gold_items.jsonl` (40), `author_items.jsonl` (50), `scheme_final.jsonl` (197); `notes-B3.md`, `assign_raw.jsonl` (17,653), `multiplacement_log.jsonl` (3,171), `b3_classes.jsonl` (7,844), `defects.jsonl` (63), `return_waivers.jsonl` (74).
- B4: `notes-B4.md`; `transmission-B4.md`; `coder_conflicts.jsonl` (72); `disagreements.jsonl` (3,288; older versions kept as `disagreements.superseded-*.jsonl`, the last 20261013T131513Z); `assign_final.jsonl` (7,844: class 1 2,566, class 2 1,990, ruled 3,288, pending 0); `adjudication_report.md` (full, packets 1 to 7); `out/_adjudication-B4/` (seven packets with drafts and `summary.md`, `boundaries.md`, the re-rule folders of B4a-05 and B4a-06).
- Asks QB2-001 to QB2-014, QB3-001 to QB3-005, QB4-001 to QB4-003 are answered.

## Rules kept

- Never type an id, an order or a name from memory; check with `jq` or a script. Never log a write before the tool answers. Count a ledger's lines after appending.
- There is no `--kind close` and no `--kind episode`; log those as notes.
- A question to J is `mem.py ask "<question>" --assume "<meanwhile>"`. Read an answer with `mem.py asks <id>`.
- Never put a placeholder or a no-op in a command. Never put an Edit or Write and the read, dry run or validator run that checks it in one turn. Count the head's words before the closing line.
- Never write an id-like token the run does not define in the head, such as a D or a P followed by digits: `done.sh` reads it as an unresolved link.
- The permission check refuses: "=w="; shell for and while loops, and words beginning "for_"; shell functions, brace groups and heredocs; awk with system() or `$` fields; a variable set from `$(...)`, and some `python3 -c` scripts; jq with a `test()` regex, and env-var prefixes; cat or jq on `out/diary/`; `cut -c`, `cut -b` and `colrm` after `mem.py`, `cite.py` or a helper (use `--width`). Write a script file in `out/_analysis/` instead. `.git` is read-denied.
- `mem.py` pages long output; fetch the rest with `--offset`. A Bash output over about 30 KB is saved where I cannot read it; page it. A `Read` of a chunk file over 25,000 tokens is cut; read it by offset and limit.
- After a compaction the guard requires the head and stage file read again and the exact line "compacted; head reconciled" through `mem.py log --kind note`, in a call of its own, before any other step.
- In B4: a ruling never cites a room note; reader-error never where both readers made the same miss; T3-063 takes canvas posts only; a correction goes through `--replace`, never a hand edit. Class 4: the readers' list taken is author-error, a B1 theme kept author-disagrees. `adjudication_report.md` names both themes of every `boundary_ruling` pair, or `validate.py` fails.

## The next comparison

B4b builds the blind sample under §10 of the stage file. Declared in `notes-B4.md` before the draw: one hundred items, fifty drawn uniformly from the analysed set and fifty from the adjudicated items, by a recorded seed, the two strata never pooled; two fresh readers per item, one dispatch, two hundred comparisons against the settled line; stage B2's agreement definition unchanged; no exemplars unless B3 served them. Then `RETURN-B4.md`, `packets.py B4` under the 50 KB ceiling (the frozen scheme and the sampled themes, nothing else), the manifest entry, and `done.sh B4 --cut packets-out`. Check `arm.json` for the OpenAI third-reader arm first. The figures will be read against the stability floor 0.807 and B3's agreement 0.786.