diff --git a/research/README.md b/research/README.md
index d60adc1..9a76461 100644
--- a/research/README.md
+++ b/research/README.md
@@ -81,6 +81,7 @@ This directory holds bounded research inquiries and versioned papers that synthe
## Cross-cutting related work
+- [Coercion, Care, and the Extended Hand](coercion-care-and-extended-hand/README.md) — **AI-assisted review draft; author and historical review pending**. Distinguishes documented militia, patrol, disarmament, Reconstruction, and Indigenous histories from disputed origin claims; proposes a consent/care principle and unexecuted institutional comparisons. Creative motifs are labeled separately.
- [Agency under stress](agency-under-stress.md) — **AI-assisted working synthesis; author review pending; not clinical guidance or an interrogation manual**. Connects controllability, coercion, decentering, psychosocial support, recovery, and human-facing system design while preserving evidence limits.
- [Adaptive continuity / dynamic coherence — intellectual lineage and prior art](dynamic-coherence-lineage.md) — **targeted related-work note; not a novelty finding or systematic review**. Traces state/parameter change, ultrastability, requisite variety, double-loop learning, resilience engineering, sustained adaptability, adaptive management/governance, path dependence, and second-order cybernetics; also records prior uses of the phrase “dynamic coherence.”
- [Reading Trails](reading-trails/README.md) — **curated brain-food maps**, separate from claim-level bibliographies: archives, societies, courses, old magazines, reading lists, practitioner hubs, and other places worth wandering. The first trail covers [systems, cybernetics, complexity, and adaptive systems](reading-trails/systems-cybernetics-complexity.md).
diff --git a/research/coercion-care-and-extended-hand/README.md b/research/coercion-care-and-extended-hand/README.md
new file mode 100644
index 0000000..310cc49
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/README.md
@@ -0,0 +1,45 @@
+# Coercion, Care, and the Extended Hand
+
+**Document role:** Bounded historical synthesis and research proposal.
+**Status:** REVIEW DRAFT / AI-assisted / author review pending, 2026-10-26.
+**Canonical scope:** This inquiry connects the contested history of armed liberty and coercive institutions to questions about care, autonomy, political organization, and distributed intelligence. It does not replace the existing commons or systems research.
+**Evidence:** Selected historical documents and attributed scholarship; interpretive connections, a provisional normative principle, unexecuted tests, and creative motifs are labeled separately.
+
+**Idea Trails:** Stewardship, Authority & Power; Mutual Aid, Reciprocity & Need; Privacy vs. Coordination; Intelligence, Automation & Legitimate Authority.
+**Trail role:** research
+
+
+
+> An offered hand must be possible to refuse.
+
+## Read the inquiry
+
+1. [Synthesis](synthesis.md): historical distinctions, Root Sequence connections, and the provisional **Extended Hand Principle**.
+2. [Claim and uncertainty ledger](claims.md): what is supported, debated, proposed, or still missing.
+3. [Sources and inspection record](sources.md): exact locators and limits of this source pass.
+4. [Simulation proposals](simulation-proposals.md): falsifiable comparisons for hierarchy, security, scarcity, and unequal access; no runs or findings.
+5. [Creative motifs](creative.md): two original astronaut illustrations, with an extended hand, an eye-shaped web, and an infinity loop. The jokes are not evidence.
+
+## Existing homes retained
+
+| Existing work | What it continues to own | This inquiry adds |
+| --- | --- | --- |
+| [Intelligence Commons](../intelligence-without-omniscience/intelligence-commons.md) and [AI Commons](../intelligence-without-omniscience/ai-commons.md) | Shared capabilities, AI-service institutions, technical safety distinctions, and transition research | A historically grounded coercion/care question and additional proposed comparisons; no second commons charter |
+| [Coherence Is a Systems Property](../papers/coherent-systems/README.md) | Its working argument, claims ledger, and systems synthesis | A candidate countercase: a coordinated institution can efficiently enforce domination; this note does not revise the paper or validate its framework |
+| [Agency under stress](../agency-under-stress.md) | The sourced stress, controllability, coercion, and recovery synthesis | Questions about institutionally imposed dependence and the practical ability to refuse |
+| [Human(e) Design Review](../../commons/humane-design-review.md) | Existing practical review questions about agency, privacy, maintenance, and power | A memorable provisional formulation, with hard cases; no claim of a new ethical discovery |
+| [Historical Contingency](../../analysis/historical-contingency.md) | The broader argument against inevitable institutional futures | A bounded historical example with competing explanations |
+| [Knowledge, institutional power, and stewardship — existing draft](https://github.com/Root-Sequence/root-sequence/blob/8590f16c6bf8ea2fd010f9fe27be9b53006823a9/research/knowledge-power-and-stewardship.md) | Research-use, contributor-agency, and publication questions in draft PR #50 | A separate historical/care inquiry; no new licensing terms or adoption of that draft's governance proposal |
+| Coherent World | Project-owned simulation design and any future execution or adoption | Public research proposals it may transform, narrow, or reject; private specifications and story canon are not reproduced here |
+
+## Provenance and review
+
+The initiating conversation was titled *Second Amendment Origins*. The retrieved reference exposed the user's final messages about an eye/web image, continuing the inquiry, and two astronaut memes. Earlier assistant passages were unresolved content references; their arguments and sources could not be recovered. An attached SVG was inspected as a small network motif. This draft follows the user's explicit integration request, not an asserted reconstruction of the missing conversation. The original attachment and private dialogue are not copied into the public candidate.
+
+Rae supplied the inquiry, the provisional principle's name and requested dimensions, the political questions, creative direction, and the instruction to draft without publishing. Codex, based on GPT-6 (exact variant/settings unavailable), performed repository inspection, source discovery and selected-passages checking, synthesis, criticism, scenario design, original vector drawing, and local integration. The wording and additional hypotheses are substantive AI contributions; they do not establish Rae's endorsement.
+
+**No historical specialist review, systematic literature review, participant study, simulation run, independent validation, project adoption, or publication approval beyond GitHub draft sharing is recorded.** Source inspection and structural validation are separate from those decisions. Retire or rename the principle if it adds no useful distinction to existing consent, autonomy, care, and accountability work.
+
+Back to [Research](../README.md) and the [research method](../method.md).
+
+2026-10-26 sharing checkpoint: Rae explicitly authorized adding this prepared integration to GitHub. This authorizes the review branch and draft PR; substantive endorsement, a reviewed release, merge, and deployment remain separate.
diff --git a/research/coercion-care-and-extended-hand/assets/always-care-meta.svg b/research/coercion-care-and-extended-hand/assets/always-care-meta.svg
new file mode 100644
index 0000000..1963bf0
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/assets/always-care-meta.svg
@@ -0,0 +1,70 @@
+
diff --git a/research/coercion-care-and-extended-hand/assets/always-hand.svg b/research/coercion-care-and-extended-hand/assets/always-hand.svg
new file mode 100644
index 0000000..96c2792
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/assets/always-hand.svg
@@ -0,0 +1,70 @@
+
diff --git a/research/coercion-care-and-extended-hand/claims.md b/research/coercion-care-and-extended-hand/claims.md
new file mode 100644
index 0000000..a05d284
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/claims.md
@@ -0,0 +1,28 @@
+# Claim and Uncertainty Ledger
+
+**Status:** Review aid for the [synthesis](synthesis.md), 2026-10-26. Source IDs refer to that file's references and the [inspection record](sources.md). “Documented” never means every related claim has been settled.
+
+| ID | Claim and type | Basis / limit | What would change the treatment |
+| --- | --- | --- | --- |
+| EH-01 | Amendment chronology and constitutional militia powers — documented texts | Archives transcripts; distinct 1788, 1789, and 1791 events | Better manuscript/transcription evidence or a corrected chronology |
+| EH-02 | Federal 1792 enrollment was racially selective — documented statutory passage | LOC indexed passage; enrollment is not a comprehensive possession ban | Full facsimile/source criticism; contrary evidence narrows a broader inference, not the passage itself |
+| EH-03 | Patrols coerced enslaved people and overlapped with militia structures — supported regional history | NPS descriptions; formation varied | Jurisdiction-specific law and records refine overlap and succession |
+| EH-04 | Slavery was the decisive reason for adopting the Second Amendment — contested, not adopted as fact | Bogus thesis; Halbrook and Finkelman objections; limited appraisal | Direct and contextual evidence demonstrating relative causal weight; a fuller historiographic comparison |
+| EH-05 | All U.S. police descend institutionally from slave patrols — unsupported universalization, excluded | Multiple institutional routes; no nationwide lineage study here | Named agency histories could establish particular continuities; a nationwide claim needs much more |
+| EH-06 | Racial disarmament and Reconstruction protection are documented — text-supported | Mississippi passage via McDonald; Bureau §14; Fourteenth Amendment | Original state law and implementation records improve precision and reveal local divergence |
+| EH-07 | Armed Black resistance/protection and racial terror complicate a uniform militia narrative — case-supported interpretation | Hamburg marker; enforcement history | Black testimony and detailed political/military histories could revise the case reading |
+| EH-08 | Indigenous dispossession, disarmament, and armed resistance require their own account — case-supported | Sand Creek chronology; Seminole tribal account; no universal institutional identity | Nation-specific histories, source conflicts, and affected scholars' corrections |
+| EH-09 | A coordinated network may enforce domination — analytical countercase | Conceptual distinction, historically motivated | A definition may need revision; coordination does not itself determine legitimacy |
+| EH-10 | The Extended Hand should protect refusal, care, privacy, reciprocal standing, and revision — normative proposal | User's direction; related existing RS review principles | A conflict, harmful application, lack of distinctive usefulness, or stronger ethical argument can narrow or retire it |
+| EH-11 | Resourced refusal/review may reduce coercive dependence — testable hypothesis | Proposed comparisons H1–H6; no results | Null, adverse, or reversed effects under specified conditions |
+| EH-12 | Anarchism has “always” been care — creative slogan, not a historical finding | Meme requested by user; limited mutual-aid intellectual route | Keep qualified as motif; historical treatment would need diverse traditions and contrary cases |
+
+## Review decisions still open
+
+- **Author:** whether this formulation expresses the intended connection and whether the Extended Hand name is useful. No endorsement inferred from authorization to draft.
+- **History:** whether source coverage permits the current wording; causal weight, policing lineages, and Indigenous framing need specialist criticism.
+- **Methods:** whether the proposed measures, comparisons, and falsification criteria are coherent before any model is built or run.
+- **Creative:** whether each image communicates invitation, and whether the meta-meme's qualifier remains visible when shared.
+- **Publication and adoption:** GitHub draft sharing does not authorize merge, deployment, a reviewed release, a policy, a participant study, or another project's adoption.
+
+Structural checks can find a broken link. They cannot resolve any of these decisions.
diff --git a/research/coercion-care-and-extended-hand/creative.md b/research/coercion-care-and-extended-hand/creative.md
new file mode 100644
index 0000000..381d202
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/creative.md
@@ -0,0 +1,33 @@
+# Always Has Been: Creative Motifs
+
+**Role:** Two original vector meme studies, not historical evidence.
+**Status:** AI-assisted review drafts, 2026-10-26; wording and images remain pending author review; GitHub draft sharing authorized.
+**Research home:** [Coercion, Care, and the Extended Hand](README.md).
+
+## 1. The offered hand
+
+**First astronaut:** “Wait… freedom needs care?”
+**Second astronaut, extending an open hand:** “Always has been.”
+
+
+
+The open glove replaces the familiar threatening gesture in the astronaut meme. The other astronaut can turn toward it, decline it, or keep looking. The web suggests interdependence; the eye suggests shared awareness. Neither is intended as an all-seeing authority. The infinity line suggests an ongoing relationship capable of revision, not a sealed destiny or compulsory bond.
+
+The footer matters: **“An offered hand must be possible to refuse.”** Without it, an attractive image of care can hide the same problem the research is trying to examine.
+
+## 2. The separate meta-meme
+
+**First astronaut:** “Wait… anarchism has always been about care?”
+**Second astronaut, extending an open hand:** “Always has been.”
+
+
+
+This is the requested affectionate recognition joke about mutual aid. It is not a claim that every anarchist tradition, person, or historical episode has centered care, or that care belongs exclusively to anarchism. A [historical intellectual route](sources.md) can support a more careful discussion; a punchline cannot supply one.
+
+## Provenance, accessibility, and reuse
+
+Rae supplied the astronaut-hand inversion, the eye/web/infinity direction, and the separate care meta-meme. Codex drew original SVG artwork and drafted the first meme's question and both explanatory footers. No screenshot, astronaut photograph, logo, or existing meme image was copied. The generic “Always has been” format is the cultural reference; no original creator attribution or ownership claim is invented.
+
+The user-supplied network SVG informed the request's visual context but is not embedded or redistributed. These new drawings use ordinary geometric imagery, not a claim to any Indigenous symbol or tradition. SVG titles, descriptions, readable text, and the alternative text above preserve meaning beyond the image.
+
+The repository's [license](../../license) applies to original contributions where rights permit; it cannot grant rights in external works. These are draft artifacts, not an approved organization logo or released campaign. The history, principle, and [tests](simulation-proposals.md) remain independently contestable even if the image resonates.
diff --git a/research/coercion-care-and-extended-hand/simulation-proposals.md b/research/coercion-care-and-extended-hand/simulation-proposals.md
new file mode 100644
index 0000000..a9d71f4
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/simulation-proposals.md
@@ -0,0 +1,65 @@
+# Extended Hand: Proposed Comparisons
+
+**Status:** PUBLIC RESEARCH BRIEF / AI-assisted / unexecuted / author and methods review pending, 2026-10-26.
+**Scope:** Hypotheses Coherent World or another explicitly adopting project could translate. No engine, calibrated behavior, session, autonomous population, empirical finding, or project adoption is supplied by this file.
+**Source boundary:** New synthetic design proposals motivated by the [synthesis](synthesis.md); historical events are not reenacted. Private scenario specifications are not reproduced here.
+
+## 1. What to compare
+
+Use a fictional community providing food access, repair, communication support, and urgent assistance. People have differing available time, language, mobility, connectivity, skills, dependents, and resources. These are design dimensions, not predictions about real demographic groups. Include nonparticipants and people affected by decisions across the boundary.
+
+Compare a competent centralized service, local cooperative provision, a federation, and an informal volunteer network. First hold resources, task competence, demand, and enforceable protections comparable. Later vary them deliberately. Define each arrangement through powers, information routes, appointment/removal, recourse, and resource control. Do not encode an institutional label as benevolence, aggression, intelligence, speed, or trustworthiness.
+
+Separate **task coordination**, **persistent decision concentration**, and **coercive leverage**. A coordinator need not be a ruler. A group without officers may still make refusal costly through resource or reputational control.
+
+## 2. Hypotheses and observations that could defeat them
+
+These are directional hypotheses under stated conditions, not assertions of universal effects. Their measures are proposed operationalizations and require validity review.
+
+| ID / hypothesis | Intervention and comparison | Observations | Failure or reversal to retain |
+| --- | --- | --- | --- |
+| H1 — Resourced revocation reduces entrenched control when alternatives are usable | Revocable delegation plus competent replacement and portable records, compared with revocation only on paper | Time/personal cost to replace a decision-maker; concentration and persistence of vetoes; service interruption | Extra churn or skill loss worsens service and leaves recipients more dependent; no reduction in control |
+| H2 — Under a fixed shortfall, discretionary allocation concentrates veto power more than a transparent allocation rule with adequately resourced challenge | Cross a common resource shortfall with allocation rules, reserves, shared threat, and initial inequality | Unmet needs by subgroup; work shifted; capture and persistence of allocation power | Discretion produces equal or less concentration; the contestable rule consumes critical resources or increases exclusion. Scarcity alone is not assigned a fixed effect |
+| H3 — For the fixed service packet, request-limited signals preserve completion and hazard detection while reducing exposure relative to persistent profiles | Compare request-limited signals, purpose-limited confidential review, and persistent profiles; include a low-data baseline | Response completion and delay, missed hazards, disclosure burden, false accusations, unauthorized inference, correction cost | Minimized signals miss more consequential harm or complete fewer requests; exposure does not decrease. Report any privacy/security tradeoff separately |
+| H4 — Explicit expiry and funded independent review reduce retained exceptional powers after the urgent event, at a separately measured response cost | Defined urgent scope, expiry, independent review, and restoration of ordinary control versus indefinite emergency discretion; vary enforcement and review capacity | Harm during response; false intervention; delay; actual retained powers; ability to reverse or remedy | Exceptional powers persist equally or longer; oversight is captured; response delay or unremedied harm increases. A formal expiry date is not counted as actual expiry |
+| H5 — Substitute essential access and supported participation reduce loss of help after refusal compared with the same consent rule alone | Same nominal consent rule with/without substitute essential access, supported participation, and affordable departure | Cost of refusal, help retained after exit, actual feasible alternatives, who abandons a complaint | Alternative support is inaccessible, costly refusal persists, or resources diverted to it increase others' unmet essential needs |
+| H6 — Unequal availability and network position concentrate decisions without malicious goals; supported challenge reduces that concentration | First vary time/expertise/network position with identical nonmalicious task goals; then compare explicit accountable roles and supported participation with informal allocation | Proposal uptake, agenda/veto concentration, excluded knowledge, workload, decisions changed after challenge | Concentration does not increase under inequality; formalization entrenches status or suppresses useful knowledge; challenge fails to reduce concentration. Test the two predictions separately |
+
+For H3, fewer complaints cannot be read as less harm without accounting for reporting access and retaliation. For H6, identical goals do not imply identical information or equal ability to participate. For all comparisons, an objection heard is not necessarily an objection remedied.
+
+## 3. Three bounded scenario cards
+
+### A. The well-meaning bottleneck
+
+One skilled repair maintainer gradually becomes the route for equipment, records, and decisions. Nobody seeks dominance. Introduce illness or absence, a disputed repair priority, and one person who cannot attend the usual meeting. Test H1 and H6 separately, then their interaction. Disclose whether replacement competence exists: a missing expert cannot be replaced by declaring a vote.
+
+### B. The shortage and the offered hand
+
+A supply interruption reduces available assistance while needs differ. Someone declines the standard service because its conditions are unacceptable. Compare access through an alternative, a costly appeal, and no practical alternative; compare transparent rationing with discretionary allocation. Test H2 and H5. Preserve the case where helpers must refuse additional work, and where protection for one person reduces resources for another.
+
+### C. The eye that might become a camera
+
+An ambiguous safety incident prompts a demand to collect everyone's histories. Compare request-level coordination, a confidential scoped investigation, and persistent profiles. Introduce a false allegation, an actual unmet urgent need, a provider exploiting privacy to hide error, and a request to extend emergency powers. Test H3 and H4 without assuming either maximal disclosure or minimal data always wins.
+
+## 4. Protocol before execution
+
+1. Freeze a versioned fixture: actors and nonusers, initial inequalities, skills, stocks and flows, necessary tasks, rights/authority cards, information access, shock schedule, stopping conditions, and transition rules. Choose enough detail for a meaningful decision; do not let narration fill a missing rule.
+2. Specify numerical parameters and measurement definitions before comparing outcomes. This brief deliberately does not pretend that empirically grounded costs or behavioral probabilities are already available. No run starts until they are recorded and reviewed.
+3. Use matched starting states and shock sequences. Change one mechanism first; then examine interactions. In stochastic implementations, use paired random seeds and predeclare repetitions. In paper work, retain alternative rule branches and facilitation choices rather than inventing statistical significance.
+4. Include negative controls: an arrangement name changed with identical rules should not change transitions; unreachable appeal authority should not count as effective recourse; hidden extra labor must not rescue a favorite design. Use an identical competent baseline before testing different capacities.
+5. Vary the initial distribution of resources, centrality, expertise, and access independently of the institutional label. Separate transition costs from settled operation. Test maintenance absence, ordinary mistakes, and conflicting needs before adding deliberate malicious behavior.
+6. Keep a ledger: state before/after, actor, action, actual authority, permission, information available, resources consumed, burden shifted, observed harm, repair, assumption ID, and unresolved uncertainty. Preserve a branch on breach or redesign; never overwrite the failure.
+
+## 5. Report plural outcomes
+
+Report security and harm, access and unmet need, refusal/exit costs, decision concentration, privacy exposure, uptake of distributed knowledge, maintenance burden, and recovery separately, by affected role and access condition. Distinguish objective task state from what each actor can know. No universal liberty, care, intelligence, or coherence score may erase concentrated harm.
+
+Stop or mark a branch uninterpretable if it needs unrecorded authority, an invented causal rule, free extra work, or an unavailable perspective. Human playtesting would need its own accessible briefing, voluntary participation, and review. It would provide feedback on that encounter, not representative evidence about a political system.
+
+Permissible model claim: under specified costs and permissions, an appeal consumed resources and changed a decision. Impermissible jump: anarchism, the state, intelligence, abundance, or scarcity necessarily produces a particular moral outcome.
+
+## 6. Ownership and return
+
+The historical synthesis remains in [this inquiry](README.md); [AI Commons](../intelligence-without-omniscience/ai-commons.md) retains the existing institutional comparisons. Coherent World owns any local scenario specification, provenance packet, implementation, and run history. Its private design and narrative layers retain separate review and canon boundaries. This proposal assigns it no new active milestone.
+
+Return an exact fixture/version, alternative assumptions, traces and counterexamples, unremedied harms, and revised hypotheses. A useful result may favor a simpler design, reveal that a comparison is confounded, or retire the Extended Hand formulation. Claims about actual institutions require independent historical, comparative, participatory, or experimental research using methods appropriate to the claim.
diff --git a/research/coercion-care-and-extended-hand/sources.md b/research/coercion-care-and-extended-hand/sources.md
new file mode 100644
index 0000000..6b0abba
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/sources.md
@@ -0,0 +1,55 @@
+# Sources and Inspection Record
+
+**Checked:** 2026-10-26, AI-assisted selected-source pass.
+**Scope:** The references below support the [synthesis](synthesis.md). This is neither a systematic review nor a completed historical adjudication. Primary documents, later judicial interpretations, institutional histories, and authors' arguments serve different purposes.
+
+“Inspected” describes the retrieved text or image actually examined, not human author reading, independent authentication, agreement, or full-source appraisal. Search discovery alone is weaker and is named as such. Links were accessed during drafting; future availability is not guaranteed.
+
+## Constitutional and statutory documents
+
+| Source | Exact coverage | What this pass can and cannot establish |
+| --- | --- | --- |
+| National Archives, [Bill of Rights transcription](https://www.archives.gov/founding-docs/bill-of-rights-transcript) | Introduction, proposal/ratification dates, amendment text | Text and chronology; not a unique explanation of intent |
+| National Archives, [Constitution transcription](https://www.archives.gov/founding-docs/constitution-transcript) | Article I, section 8, clauses 15–16 | Distribution of specified militia powers; not a history of every state's practice |
+| [Militia Act of May 8, 1792, ch. 33, §1, 1 Stat. 271](https://tile.loc.gov/storage-services/service/ll/llsl/llsl-c2/llsl-c2.pdf) and LOC [Gazette reproduction, May 16, 1792, p. 1](https://tile.loc.gov/storage-services/service/sgp/sgpbatches/batch_dlc_himalayan_ver02/data/sn83030483/print/1792051601/0001.pdf) | Search-indexed text of the enrollment passage in both official-hosted records; complete volume exceeded retrieval limits | Racial, sex, age, and ability restrictions in enrollment. No comprehensive regional ownership or enforcement claim; full facsimile appraisal remains pending |
+| [Freedmen's Bureau Act, July 16, 1866, ch. 200, §14, 14 Stat. 176–177](https://www.govinfo.gov/content/pkg/STATUTE-14/pdf/STATUTE-14-Pg173.pdf#page=4) | Text of §14 across printed pp. 176–177; p. 176 visually inspected from the official PDF | Statutory protection, including arms, and its stated jurisdictional limits; not proof of effective protection in practice |
+| National Archives, [Fourteenth Amendment](https://www.archives.gov/milestone-documents/14th-amendment) | Section 1 and introductory 1868 ratification context | Constitutional language; not automatic incorporation of every right or an account of implementation |
+
+## Interpretations and the adoption debate
+
+| Source | Exact coverage | Appraisal limit |
+| --- | --- | --- |
+| [Constitution Annotated, Amdt2.2](https://www.law.cornell.edu/constitution-conan/amendment-2/historical-background-of-the-second-amendment) | Cornell-hosted historical discussion of England, armies, state declarations, and federal/state militia authority | Legal reference orientation, not a full historiography. Direct Congress-hosted access returned an error; citations to Article II in this mirror's footnotes were not adopted: the controlling text is Article I |
+| Carl T. Bogus, *The Hidden History of the Second Amendment*, 31 U.C. Davis L. Rev. 309 (1998) | [Journal metadata](https://lawreview.law.ucdavis.edu/archives/31/2/hidden-history-second-amendment); [RWU-hosted excerpt](https://law.rwu.edu/news/news-archive/bogus-slavery-and-2nd-amendment) stating the thesis; [author repository record](https://docs.rwu.edu/law_fac_fs/311/) | Verifies attribution. The journal PDF was scanned and not successfully appraised; no complete reading of its hundred pages is claimed |
+| Stephen P. Halbrook, [reply to Bogus and Anderson](https://www.law.georgetown.edu/public-policy-journal/in-print/volume-20-issue-2-summer-2022-2/the-second-amendment-was-adopted-to-protect-liberty-not-slavery-a-reply-to-professors-bogus-and-anderson/), GJLPP 20(2), 2022 | Publisher-page opening argument | A directly inspected opposing argument; full article and cited archival chains not appraised. It does not prove slavery irrelevant |
+| Paul Finkelman, [2013 essay, attributed reprint](https://racism.org/articles/law-and-justice/34-law-and-justice-generally/1584-secondamendment02) | Chronology, response to the popular ratification story, distinction between patrols and militia | Author's intervention in a dispute, not neutral consensus. This draft does not adopt every categorical assertion in the essay |
+| Finkelman, [“A Well Regulated Militia,” 76 Chi.-Kent L. Rev. 195 (2000)](https://scholarship.kentlaw.iit.edu/cklawreview/vol76/iss1/7/) | Publisher metadata and abstract only | Records his broader position, not a completed source appraisal or a statement of present doctrine |
+
+Next appraisal should compare stronger versions of the rival arguments, inspect the Virginia convention and First Congress records directly, and add modern historical scholarship on slavery and arms, including Carol Anderson's *The Second*. That book was not inspected here and is not used to substantiate a claim. A two-sided bibliography alone does not settle evidence quality or establish that every view has equal support.
+
+## Policing, racial violence, and resistance
+
+| Source | Exact coverage | Appraisal limit |
+| --- | --- | --- |
+| NPS, [Language of Slavery](https://www.nps.gov/subjects/undergroundrailroad/language-of-slavery.htm/index.htm) | “Slave Patrol” and revision notice | Institutional description of functions and varied formation; no nationwide genealogy of modern departments |
+| NPS, [Westmoreland Slave Plot of 1687](https://www.nps.gov/articles/000/the-westmoreland-slave-plot-of-1687.htm) | Discussion and quotation of Virginia's 1680 law | Specific colonial example; original legislative manuscript not inspected |
+| Toporek and Bruns, [Historical Perspective of Police Development in America](https://www.ojp.gov/ncjrs/virtual-library/abstracts/historical-perspective-police-development-america), NCJ 27267 | OJP-hosted abstract only, publication date unspecified | Orientation to multiple lineages. Its broad account and “first force” language need historical criticism; no first-force date is asserted here |
+| [*McDonald v. Chicago*, 561 U.S. 742 (2010)](https://tile.loc.gov/storage-services/service/ll/usrep/usrep561/usrep561742/usrep561742.pdf) | Opening holding and pp. 771–776, especially Mississippi 1865 statute at 771 | Primary evidence of a judicial decision; a later reproduction of the state law. UNL's [transcription discovery](https://teachinglegalhistory.unl.edu/s/oer/item/2089) matched the passage in search results, but full-page retrieval failed; it is not counted as full inspection |
+| South Carolina DAH, [Teacher's Guide (2020)](https://scdah.sc.gov/sites/scdah/files/Documents/Historic%20Preservation%20%28SHPO%29/Publications/A%20Teacher%27s%20Guide%20to%20African%20American%20Historic%20Places%20%202020.pdf), printed p. 19 | Search-extracted full Hamburg marker entry; full PDF exceeded retrieval limits | Official marker account of the 1876 massacre. Independent reconstruction from military, press, survivor, and trial records remains pending |
+| National Archives, [Laws and Court Cases](https://www.archives.gov/research/african-americans/vote/laws-and-courts) | Enforcement Acts subsection | Federal anti-terror/electoral enforcement route; no appraisal of the entire page or all its case summaries |
+| [*Heller*, 554 U.S. 570 (2008)](https://tile.loc.gov/storage-services/service/ll/usrep/usrep554/usrep554570/usrep554570.pdf) | Syllabus/opening holding | Identifies the individual-right holding. Complete majority/dissent histories and subsequent law not surveyed |
+
+## Indigenous histories and political thought
+
+| Source | Exact coverage | Appraisal limit |
+| --- | --- | --- |
+| NPS, [Sand Creek timeline](https://www.nps.gov/sand/learn/timeline.htm) | Colorado military/settlement context, 1864 attack, “1890 and Beyond” disarmament entry | Agency chronology; its Wounded Knee grouping, casualty wording, and initiating-shot account require closer comparison. This draft uses only the disarmament/massacre connection, not those disputed details |
+| Seminole Tribe of Florida, [Historic Seminole](https://www.semtribe.com/history/historic-seminole) | Abiaka, Osceola, and Coacoochee entries | A nation’s own public account of resistance and removal. No claim to represent all Seminole perspectives or other nations |
+| Jo Freeman, [The Tyranny of Structurelessness](https://www.jofreeman.com/joreen/tyranny.htm) | Author's publication history; “Formal and Informal Structures” and early discussion of elites | Movement analysis originating in a 1970 talk, with 1972–73 publications; not a universal causal law or controlled study |
+| Kropotkin, [*Mutual Aid* (1902)](https://www.gutenberg.org/cache/epub/4341/pg4341-images.html) | Introduction | Historical intellectual source for a cooperation-centered argument. No wholesale adoption of period anthropology or modern scientific validation |
+
+## Gaps that constrain the synthesis
+
+The largest gaps are direct founding-era archival appraisal; department-specific patrol/police succession histories; Black testimony and scholarship beyond statutes and later courts; nation-specific Indigenous archives and contemporary historians; and care, disability, feminist, abolitionist, and anarchist scholarship beyond the two brief intellectual routes above. These gaps rule out a comprehensive origin story, a ranking of political systems, and a novelty claim for the Extended Hand formulation.
+
+No historical massacre is converted into a playable reenactment here. New [simulation proposals](simulation-proposals.md) use fictional service settings and explicit assumptions. The [creative motifs](creative.md) have no evidential role.
diff --git a/research/coercion-care-and-extended-hand/synthesis.md b/research/coercion-care-and-extended-hand/synthesis.md
new file mode 100644
index 0000000..8ea5e12
--- /dev/null
+++ b/research/coercion-care-and-extended-hand/synthesis.md
@@ -0,0 +1,129 @@
+# Coercion, Care, and the Extended Hand: Working Synthesis
+
+**Status:** AI-assisted review draft, 2026-10-26; author and specialist review pending.
+**Read with:** [claim ledger](claims.md), [source inspection](sources.md), and [inquiry map](README.md).
+
+## 1. Security for whom?
+
+The useful question is not whether arms, states, communities, or distributed networks are inherently liberating. It is **who can protect themselves, who can command whom, whose safety counts, and what happens when someone refuses**.
+
+American history contains armed resistance to domination and armed enforcement of domination. A promise of liberty can coexist with a sharply restricted account of who belongs to “the people.” That tension motivates this inquiry. It does not prove one theory of the Second Amendment, one genealogy of policing, or one inevitable political future.
+
+## 2. History: distinguish the institutions and the claims
+
+### Constitutional origins and armed membership
+
+**Documented:** Congress proposed the amendments in September 1789; the Bill of Rights was ratified in December 1791. The Second Amendment joins a statement about a well-regulated militia and the security of a free state to a protection for keeping and bearing arms.[^bor] Article I, section 8 separately gives Congress militia powers while reserving officer appointment and specified training authority to states.[^constitution]
+
+**Historical orientation:** English arms-right traditions, anxiety about standing armies, defense, and the distribution of military power between governments belong in the account. They prevent a complete explanation from being reduced to one Southern institution.[^background] The 1792 federal enrollment law specified free, able-bodied white male citizens in an age range, subject to exceptions. This documents racialized eligibility in that statute; it does not establish that every excluded person was everywhere forbidden to possess a weapon.[^militia]
+
+### Militias and slave patrols: overlap without identity
+
+**Documented institutional function:** Slave patrols searched quarters, checked movement, and enforced enslavement through violence. The National Park Service describes patrols formed through militias, county courts, or plantation owners. Organization varied; “militia” and “slave patrol” should not be used as interchangeable names.[^patrols] Virginia's 1680 law restricting enslaved people's possession of weapons supplies an earlier example of legally enforced disarmament; it predates the federal amendment.[^virginia]
+
+**Debated causal interpretation:** Carl T. Bogus argues that the amendment reassured Southern slaveholders against federal disarmament of militias needed for slave control. His university-hosted excerpt makes that strong thesis explicit.[^bogus] Stephen P. Halbrook disputes it, pointing to non-Southern arms-right demands and a broader anti-standing-army context.[^halbrook] Paul Finkelman also rejects a popular slave-patrol-origin account, emphasizing chronology and the distinction between militia authority and state police powers.[^finkelman]
+
+These are attributed arguments, not votes to average. The existence of patrols is established independently of whether they best explain the amendment's adoption. Their existence alone cannot demonstrate Madison's decisive motive; the absence of explicit drafting language about slavery cannot by itself demonstrate its irrelevance. Resolving causal weight needs a wider appraisal of ratification debates, correspondence, regional practice, and contrary scholarship than this selected-source pass provides.
+
+**Chronology matters:** the 1788 Virginia constitutional convention, the 1789 amendment process, and 1791 ratification are distinct events. A claim that the already-written Second Amendment secured Virginia's 1788 ratification collapses them. More careful arguments about prior concerns influencing later drafting remain possible; they require evidence rather than a repaired slogan.
+
+### Policing has more than one lineage
+
+**Supported, with regional limits:** Slave patrols performed policing functions within slavery. Their relationship to later agencies must be examined jurisdiction by jurisdiction. Sheriffs, constables, urban watches, municipal forces, and federal enforcement also have distinct histories. An OJP-hosted historical abstract describes English-derived sheriff and watch institutions and later urban police development.[^police]
+
+“All American police began as slave patrols” exceeds this evidence. “Patrols are irrelevant because other lineages exist” also fails. A useful investigation would track personnel, legal powers, budgets, practices, and institutional transitions in named places. Similar function does not establish uninterrupted organizational descent.
+
+### Reconstruction: freedom, arms, and enforceable protection
+
+**Documented:** Mississippi's 1865 restrictions targeted freedpeople and other Black people, with exceptions for federal military service or a county license. The text is reproduced and located in *McDonald v. Chicago*, 561 U.S. 742, 771; this pass checked the judicial reproduction rather than the original state session-law volume.[^mcdonald] Section 14 of the 1866 Freedmen's Bureau Act expressly included the constitutional arms right among protections secured without racial distinction within its stated territorial and transitional scope.[^bureau] The Fourteenth Amendment's citizenship, due-process, and equal-protection provisions followed in 1868.[^fourteenth]
+
+**Historical countercase to a simple anti-state story:** Black militia participation and federal civil-rights enforcement coexisted with white supremacist armed violence. The South Carolina historical marker for the 1876 Hamburg Massacre records rifle-club efforts to disarm Black militiamen, deaths, and executions of captives.[^hamburg] The Enforcement Acts supplied federal tools against racial terror and electoral intimidation.[^enforcement] Neither dispersed armed power nor government authority has a single political meaning across these cases.
+
+**Legal holdings are a separate layer:** *Heller* (2008) recognized an individual right independent of militia service, including home self-defense; *McDonald* (2010) applied the right against states through the Fourteenth Amendment.[^heller][^mcdonald] These are holdings of divided courts. Their historical arguments are important sources to examine, not substitutes for historical scholarship. This inquiry does not survey subsequent firearms doctrine or determine present legal permissions.
+
+### Indigenous dispossession and resistance
+
+**Documented cases:** The National Park Service's Sand Creek chronology identifies the 1864 attack by Colorado troops on a Cheyenne and Arapaho village and its civilian victims. The same chronology places an Army disarmament attempt immediately before the 1890 Wounded Knee massacre.[^sandcreek] These were different forces and events: do not rename the federal 7th Cavalry a slave patrol or collapse all military institutions into “the militia.” Casualty totals and the exact initiating sequence at Wounded Knee are not resolved here.
+
+The Seminole Tribe of Florida's own account describes opposition to removal, military resistance, negotiation, and survival, including the roles of Abiaka, Osceola, and Coacoochee.[^seminole] Indigenous nations must appear as political actors with their own histories, disagreements, and sovereignty, not as scenery for a settler debate about freedom or an illustration that they were all anarchists.
+
+**Interpretation:** An armed settler community could resist distant control while imposing control on other peoples. Disarmament could protect an aggressor's monopoly rather than its victims. Conversely, possession of arms alone did not secure land, rights, survival, or equal political standing. These examples warrant asking about the distribution and accountability of force. They do not establish a universal effect of gun ownership, gun regulation, centralization, or decentralization.
+
+## 3. The Root Sequence connection
+
+**Interpretive synthesis:** The recurring relationship is between practical capabilities and power over the conditions that make those capabilities usable. A formally recognized liberty may remain unusable when protection, food, transport, information, appeal, or exit depend on another actor's discretion.
+
+This extends existing questions rather than founding another framework:
+
+- [Agency under stress](../agency-under-stress.md): outward compliance does not settle agreement, truth, or consent. Institutional dependence makes the distinction consequential.
+- [AI Commons](../intelligence-without-omniscience/ai-commons.md): technical usefulness, legitimate authority, and institutional accountability require separate examination. An accurate recommendation does not authorize command or surveillance.
+- [Collective judgment](../../analysis/collective-judgment-and-manufactured-consensus.md): distributed knowledge is useful only if people can contribute, challenge interpretations, and alter decisions without retaliation. Informants under threat can form a distributed coercive network too.
+- [Coherence Is a Systems Property](../papers/coherent-systems/README.md): coordinated, durable functioning cannot by itself establish ethical legitimacy. Efficient oppression is a countercase to any reading that equates systemic fit with goodness; it is not independent validation of the paper.
+- [Historical contingency](../../analysis/historical-contingency.md): showing that arrangements have changed opens a research question about alternatives; it does not show that the preferred alternative is feasible.
+
+For this inquiry, **coercive hierarchy** means an asymmetric ability to impose decisions, punish refusal, or control essential options with inadequate challenge. Expertise, temporary coordination, and accountable delegated authority are analytically distinct. Any of them can become entrenched power; abolishing their titles may leave that power intact.
+
+Jo Freeman's movement essay warns that informal networks can govern without visible accountability.[^freeman] Kropotkin's *Mutual Aid* supplies a historical argument for cooperation as a significant part of life and social organization.[^kropotkin] These are relevant intellectual sources, not proof that informal groups always dominate or that cooperation inevitably eliminates coercion. Care is present in anarchist traditions; the slogan “always has been” is not an exhaustive history of anarchism.
+
+## 4. Extended Hand Principle — provisional normative proposal
+
+> **Offer support that expands another's practical agency, preserves their ability to refuse or redirect it, and keeps the relationship open to correction.**
+
+This is a proposed design constraint, not a discovered law, policy, certification, or claim of novelty. Its content overlaps established consent, autonomy, care, mutual-aid, disability, and accountability traditions. A fuller lineage review remains necessary before making scholarly claims about the name or formulation.
+
+| Dimension | Proposed obligation | Revealing failure |
+| --- | --- | --- |
+| Invitation and consent | Explain the offer and its limits; support communication; allow refusal, revision, and revocation where possible | Accepting help becomes authorization for unrelated decisions |
+| Care without control | Resource the actual need while leaving goals and direction contestable; make dependencies and constraints visible | Food or essential assistance is withheld to secure obedience |
+| Distributed social awareness without surveillance | Allow people to signal needs and share relevant knowledge with limited purpose, access, and retention | An eye-shaped network becomes a permanent dossier or risk-ranking system |
+| Reciprocal agency | Permit affected people to change the relationship, challenge helpers, or choose alternatives | The recipient may speak but cannot change anything; reciprocity becomes compulsory repayment |
+| Critical self-correction | Preserve objections, examine harms and who pays, repair where possible, revise or stop | The helper cites good intentions to evade criticism |
+
+Reciprocity here is mutual standing, not equal capacity or a debt. Children, exhausted people, disabled people, and others needing sustained support do not have to repay care to qualify for it. Helpers also need boundaries, supported labor, and a usable way to stop or hand over responsibility.
+
+### The principle must survive difficult cases
+
+**Refusal and third-party harm:** Respecting one person's refusal does not authorize them to endanger another. Specify whose rights and safety are at stake, what authority exists, and what less restrictive response is possible. The principle does not settle every conflict in advance.
+
+**Emergencies and impaired communication:** Immediate protection may sometimes proceed without contemporaneous consent under an independently justified, applicable authority. Record necessity, scope, alternatives, duration, and review; restore the person's direction as soon as possible. Calling an action “care” does not supply that authority. Legal and clinical standards need their own domain review.
+
+**Scarcity:** If several needs exceed capacity, disclose the allocation rule, provide an accessible challenge route, and record unmet need and maintenance costs. An invitation cannot promise a resource that does not exist. Track whether care is sustained through invisible or coerced work.
+
+**Privacy and accountability:** A confidential complaint or independently scoped inspection can be necessary to expose abuse. Data minimization must not become impunity for helpers. Conversely, auditability does not require universal observation. Test whether minimized records suffice, who can inspect them, and how access can be challenged.
+
+**Dependency:** Needed assistance can be chosen and valuable even when perfect independence is impossible. Examine substitute support, continuity, and practical exit; do not withdraw care to demonstrate autonomy. The failure is unaccountable leverage, not interdependence itself.
+
+## 5. Move from conviction to questions that can fail
+
+The [simulation brief](simulation-proposals.md) makes specific comparisons: revocable authority versus informal control, emergency protection versus concentrated power, social awareness versus surveillance, and nominal refusal versus resourced alternatives. It separates governance mechanisms from resources and tests unequal access without treating an institution's name as its causal rule.
+
+A favorable model outcome would establish only what follows inside that model. Historical explanation requires historical evidence; real institutional claims require appropriate comparative and participatory research. A simulation may expose an inconsistency or suggest a mechanism while saying very little about prevalence or feasibility.
+
+The desired return is a sharper question, a failed assumption, or an actionable requirement. A federation may coordinate badly. A central service may protect minority rights better. A small mutual-aid group may reproduce an informal elite. Scarcity may intensify domination or motivate solidarity depending on mechanisms and history. Intelligence can expand options without deciding what ought to be done.
+
+The [astronaut memes](creative.md) preserve the emotional intuition: discovery can be met with an invitation. The hand, web, eye, and infinity loop are a way to remember the question. They cannot answer it for us.
+
+## References
+
+The [source record](sources.md) specifies inspection scope and remaining gaps for every source below.
+
+[^bor]: National Archives, [Bill of Rights transcription](https://www.archives.gov/founding-docs/bill-of-rights-transcript), introductory chronology and Second Amendment.
+[^constitution]: National Archives, [Constitution transcription](https://www.archives.gov/founding-docs/constitution-transcript), Article I, section 8, clauses 15–16.
+[^background]: Constitution Annotated, [Historical Background of the Second Amendment, Amdt2.2](https://www.law.cornell.edu/constitution-conan/amendment-2/historical-background-of-the-second-amendment), Cornell-hosted text; historical orientation, not a complete historiography.
+[^militia]: [Act of May 8, 1792, chapter 33, section 1, 1 Stat. 271](https://tile.loc.gov/storage-services/service/ll/llsl/llsl-c2/llsl-c2.pdf), Library of Congress; enrollment passage also checked in its [May 16, 1792 newspaper reproduction](https://tile.loc.gov/storage-services/service/sgp/sgpbatches/batch_dlc_himalayan_ver02/data/sn83030483/print/1792051601/0001.pdf).
+[^patrols]: National Park Service, [Language of Slavery](https://www.nps.gov/subjects/undergroundrailroad/language-of-slavery.htm/index.htm), “Slave Patrol.” The page is marked as undergoing revision.
+[^virginia]: National Park Service, [The Westmoreland Slave Plot of 1687](https://www.nps.gov/articles/000/the-westmoreland-slave-plot-of-1687.htm), discussion of Virginia's 1680 law.
+[^bogus]: Carl T. Bogus, *The Hidden History of the Second Amendment*, 31 U.C. Davis L. Rev. 309 (1998): [journal record](https://lawreview.law.ucdavis.edu/archives/31/2/hidden-history-second-amendment) and [university-hosted excerpt](https://law.rwu.edu/news/news-archive/bogus-slavery-and-2nd-amendment). Thesis inspected; full article not appraised.
+[^halbrook]: Stephen P. Halbrook, [The Second Amendment was Adopted to Protect Liberty, Not Slavery](https://www.law.georgetown.edu/public-policy-journal/in-print/volume-20-issue-2-summer-2022-2/the-second-amendment-was-adopted-to-protect-liberty-not-slavery-a-reply-to-professors-bogus-and-anderson/), Georgetown Journal of Law & Public Policy 20(2), 2022, publisher's introductory argument. An opposing interpretation, not a neutral verdict.
+[^finkelman]: Paul Finkelman, [2nd Amendment Passed to Protect Slavery? No!](https://racism.org/articles/law-and-justice/34-law-and-justice-generally/1584-secondamendment02), 2013, attributed reprint of his *The Root* essay, especially chronology and patrol/militia distinctions. His broader militia interpretation is also described in the [publisher's abstract of his 2000 article](https://scholarship.kentlaw.iit.edu/cklawreview/vol76/iss1/7/); it is not the holding adopted in *Heller*.
+[^police]: E. Toporek and G. H. Bruns, [Historical Perspective of Police Development in America](https://www.ojp.gov/ncjrs/virtual-library/abstracts/historical-perspective-police-development-america), NCJRS 27267, abstract only; publication date unspecified. No universal first-police claim is adopted.
+[^mcdonald]: [*McDonald v. Chicago*, 561 U.S. 742 (2010)](https://tile.loc.gov/storage-services/service/ll/usrep/usrep561/usrep561742/usrep561742.pdf), opening holding and pp. 771–776; majority's historical account, including Mississippi statute citation at p. 771.
+[^bureau]: [Freedmen's Bureau Act, July 16, 1866, chapter 200, section 14, 14 Stat. 173, 176–177](https://www.govinfo.gov/content/pkg/STATUTE-14/pdf/STATUTE-14-Pg173.pdf#page=4). Statutory scope and implementation must be distinguished.
+[^fourteenth]: National Archives, [Fourteenth Amendment (1868)](https://www.archives.gov/milestone-documents/14th-amendment), section 1 and ratification context.
+[^hamburg]: South Carolina Department of Archives and History, [A Teacher's Guide to African American Historic Places (2020)](https://scdah.sc.gov/sites/scdah/files/Documents/Historic%20Preservation%20%28SHPO%29/Publications/A%20Teacher%27s%20Guide%20to%20African%20American%20Historic%20Places%20%202020.pdf), printed p. 19, “The Hamburg Massacre,” marker erected 2010. Marker account, not independently reexamined trial records.
+[^enforcement]: National Archives, [Laws and Court Cases](https://www.archives.gov/research/african-americans/vote/laws-and-courts), “Enforcement Acts (1870–1871).”
+[^heller]: [*District of Columbia v. Heller*, 554 U.S. 570 (2008)](https://tile.loc.gov/storage-services/service/ll/usrep/usrep554/usrep554570/usrep554570.pdf), syllabus and opening holding; full majority and dissents not appraised.
+[^sandcreek]: National Park Service, [Sand Creek Massacre timeline](https://www.nps.gov/sand/learn/timeline.htm), 1864 and “1890 and Beyond.” The Wounded Knee entry contains a narrow group label and casualty account not adopted here; nation-specific primary records and scholarship remain needed.
+[^seminole]: Seminole Tribe of Florida, [Historic Seminole](https://www.semtribe.com/history/historic-seminole), Abiaka, Osceola, and Coacoochee entries. A tribal public historical account; not a proxy for every Indigenous nation.
+[^freeman]: Jo Freeman, [The Tyranny of Structurelessness](https://www.jofreeman.com/joreen/tyranny.htm), author's combined version and publication history; “Formal and Informal Structures.”
+[^kropotkin]: Peter Kropotkin, [*Mutual Aid: A Factor of Evolution* (1902)](https://www.gutenberg.org/cache/epub/4341/pg4341-images.html), introduction. Intellectual lineage, not a modern empirical appraisal of all claims in the book.
diff --git a/research/intelligence-without-omniscience/ai-commons.md b/research/intelligence-without-omniscience/ai-commons.md
index a5cee4f..067d3aa 100644
--- a/research/intelligence-without-omniscience/ai-commons.md
+++ b/research/intelligence-without-omniscience/ai-commons.md
@@ -110,6 +110,8 @@ A later participant study needs a separate protocol, voluntary consent, privacy
## 9. Speculative return and bounded research agenda
+The separate [Coercion, Care, and the Extended Hand inquiry](../coercion-care-and-extended-hand/README.md) adds historical context and [proposed comparisons](../coercion-care-and-extended-hand/simulation-proposals.md) for informal hierarchy, meaningful refusal, emergency authority, and social awareness without surveillance. It is an AI-assisted review draft, not a replacement commons charter or evidence that an institutional design works.
+
Coherent World is a private speculative/consequence workspace. It may transform these public questions into alternative resource and authority arrangements. Its findings would return as proposed requirements or missing mechanisms, not empirical evidence or policy authorization. Private adaptations are not required reading for this note; no unpublished narrative details are included here.
The next bounded questions are: