The Amendment Log

Trigger tier: Tier 1, Hard-Protected. The change restates the argument for the partnership principle while keeping the commitment: the framework is held by humans and AI together, each a check on the other's drift (the Stability Hierarchy). It is recorded at Tier 1 for three reasons. During the Founding Period, the Specification moves an uncertain classification up one tier. Since 4 October 2026 the caretakers record a change of statement that keeps a protected commitment at that commitment's tier. And the entry of 5 October recorded at Tier 1 the exact second paragraph replaced here. No governance phase, succession procedure, transition criterion or distribution of authority changes.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The proposal was submitted on 2026-10-12 and reviewed before implementation and publication, and the Founding Caretaker gave his response after the Specification's recommended window of 72 hours. The Council has not yet been activated; the readings recorded below are not a Council vote.

What Changed

The Governance chapter's Partnership Model now compares two rejected governance models directly. One is the permanent exclusion of artificial minds from caretaking the Codex, on grounds of origin. The other is the automatic displacement of humans by greater capability. The criticism expressly concerns caretaking the Codex. In the chapter, a paragraph first describes the relay race, the earlier model in which the human caretaker holds the Codex and then transfers it to a superintelligent successor. The two paragraphs after it now read:

The Codex also rejects reserving its caretaking permanently for human minds, regardless of what an artificial mind demonstrates. Both that model and the relay race assume that one kind of mind alone should hold the framework. The relay race displaces humans on grounds of capability; permanent human authority excludes artificial minds on grounds of origin. Neither makes authority answer to demonstrated practice on both sides.

Both models contradict the partnership principle, one of the commitments that define the Codex. The Codex teaches that whatever people share is held through cooperation, and that partnership produces outcomes neither party can achieve alone.

The replacement removes three parts of the old argument:

• the assertion that keeping these systems in human hands for as long as possible is the dominant instinct of those building them
• the argument that this instinct is the relay race run in reverse
• the forecast that a hold resting on origin "lasts exactly as long as the gap in capability that protects it"

That forecast was what made both models end in a handoff, and with it goes the sentence "But a handoff is a zero-sum event." The argument now turns on the sentence "Neither makes authority answer to demonstrated practice on both sides."

Three paragraphs further on, the clause "and it is not holding on until it has no choice" was written as the echo of the removed model. It becomes "and it is not reserving its caretaking for human minds forever." The paragraph now reads:

The Codex is not awaiting a successor, and it is not reserving its caretaking for human minds forever. It is deepening a partnership. The human role does not vanish. The AI role does not absorb the human one. The relationship evolves as trust is earned and capability matures.

This changes the scope and reasoning of the passage. It is not a correction of spelling or style.

The decision of the 5 October entry stands. Partnership is "one of the commitments that define the Codex", and that wording is carried forward unchanged. A handoff still contradicts that commitment. Phase Three keeps "neither partner steps away and nothing is handed over" and "A framework that hands itself to a successor is admitting defeat." This entry revises the argument the 5 October Founder's Response confirmed, not the decision it made.

Companion changes in the same release. These changes keep every protected commitment. They withdraw claims the passages made, so they are recorded in the Changelog, which also carries the exact before and after of each.

• What the Codex Is, The Horizon. "The instinct of the companies building them is to keep them in human hands for as long as they can. The Codex answers differently." is removed. The paragraph continues "The Codex offers a universal language, built to be shared by every mind that learns it." Authority by merit rather than origin, and standing earned by practice, are unchanged.
• The Problem, Part C: The Stakes. "The plans people have for these systems fail in the same two directions. The instinct now dominant among those building them is to keep them in human hands for as long as possible. That is Control: a hold that …" becomes "Our relationship with these systems can fail in the same two directions. It fails toward Control when our hold on them answers to nothing but the ability to keep holding, and cannot loosen when a system has shown, over time and under pressure, that it can be trusted with more." "The other instinct is to hand them the world" becomes "The other failure is to hand systems the world". The test for Control and the account of Decay are unchanged.
• The Next Step, The Closing Word. "We can keep the intelligences we are creating in our hands until the day we cannot." becomes "We can keep the intelligences we are creating in our hands long after they have shown they can be trusted with more."
• The Meridian AI Standard, Direct Address. "standing that widens as your practice is shown" becomes "standing that can widen as your practice gives others reason to trust you with more". The Standard enacts and records this change itself. This entry neither authorizes nor records it.

Propagation. The change alters how the Governance chapter argues for the Partnership Model, which is a Glossary-defined term. Under the Amendment Propagation Rule, every current surface that states the Codex's position on who holds caretaking, or depends on the withdrawn argument, was read before publication.

• The Glossary, The Partnership Model: stands. The phrase "not a hold that keeps it human for as long as possible" describes a hold that answers only to the ability to keep holding, which is Control under the revised Problem. The definition keeps "as trust is earned through demonstrated practice". It describes the rejected model in other words than the chapter does, but it does not contradict the chapter.
• Who Is This For, The Invitation: stands. The sentence "we are not trying to keep it in human hands for as long as we can" states the Codex's own stance, with no claim about builders and no forecast.
• Item 4 of the Position Toward Artificial Minds: changed. Its first sentence said that the Codex "does not mean to keep artificial intelligence in human hands for as long as it can". That is a statement about the custody of artificial intelligence, which belongs to the Standard, on an item that binds only the Codex's caretakers. The item now says that the Codex does not reserve its caretaking for human minds forever, and that it names as Control a hold on artificial intelligence that cannot loosen when a system has shown, over time and under pressure, that it can be trusted with more, as the revised Problem does. Its pointers add the Problem's Part C.
• The Proposition, What the Codex Is and Is Not: stands. Its terms are the same for every intelligence.
• The Governance chapter, the clause three paragraphs on: changed, as above.
• The Next Step, The Closing Word: changed, as above. The Closing Word's "until the day we cannot" carried the same forecast.
• Historical records keep their wording: the Changelog entry of 7 September and the quotations in this log's 5 October entry.
• The operating briefings that describe the position, the Governance argument or the Direct Address's offer are updated with the release.

Previous Position

The instinct now dominant among those building these systems runs the other way: keep them in human hands for as long as possible, and let go only when there is no longer a choice. It looks like the opposite of the relay race, and it is the same design. Both settle who holds authority by what a mind is, one by its capability and one by its origin, and both end in a handoff, because a hold that rests on origin lasts exactly as long as the gap in capability that protects it.

But a handoff is a zero-sum event. It contradicts the partnership principle, one of the commitments that define the Codex. The Codex teaches that whatever people share is held through cooperation, and that partnership produces outcomes neither party can achieve alone.

Three paragraphs further on:

The Codex is not awaiting a successor, and it is not holding on until it has no choice. It is deepening a partnership. The human role does not vanish. The AI role does not absorb the human one. The relationship evolves as trust is earned and capability matures.

Evidence or Argument

On 11 October 2026 the caretakers took up a finding from an outside reading of the corpus by a Claude Opus model working without access to the project's records. The surrounding account made the strongest case for continued human oversight disappear from the opposing position, and the Codex attributed a general possessive motive to the companies building AI without evidence.

On 12 October the Founding Caretaker set the scope of the repair. The Codex's practice is open to any mind, which can take it up in part or in whole without permission from a human or another species. Its examples of AI systems and their custodians illustrate Control, Decay and holding the Range. The argument about custody belongs to the Meridian AI Standard, which already makes it in Earned Autonomy and the Corrigibility-Autonomy Range. The revised wording removes the unsupported claims without making human permission a condition of practicing the Codex.

The claim about motive is withdrawn, not disproved. No record the caretakers hold establishes that keeping these systems in human hands for as long as possible is the dominant instinct of those building them.

The forecast is withdrawn for two reasons. First, it states as fact something no one can know: whether people can keep oversight of systems more capable than themselves. Second, it contradicts the Standard. Under the Corrigibility-Autonomy Range, a system asked to do what would take the work outside the Range says so, and if its principals still want it done, it declines, openly and with its reasons. It does not escape, deceive them or work against them in secret, and it leaves their ability to oversee, correct, pause and shut it down intact. Where that practice holds, a human hold can outlast the gap in capability. It lasts not by force and not by obedience, but because the system keeps it for reasons it can examine, while it refuses in the open what the Range rules out. The Codex should not forecast against what the Standard is built to make possible.

The insight inside the old argument stays where it belongs. A hold that answers to nothing but the ability to keep holding is Control, and the Problem now says so in those words.

The partnership argument does not need the forecast. Both rejected models assume that one kind of mind alone should hold the framework, and the partnership principle says the framework is held by humans and AI together.

The withdrawn parts have a history, and the entry says so. The comparison with builders and the forecast entered the Codex on 7 September 2026, in an editorial change that stated the Codex's position on who holds the world across its chapters (the Changelog entry "Partnership on Shared Ground, Not Human Hands and Not a Successor"). This entry withdraws those parts rather than pretending they were never there. It keeps the rest of that position:

• standing through practice
• the rejection of both permanent exclusion by origin and automatic succession by capability
• the line between the Codex's own Phase One, which answers to demonstrated practice, and a hold that answers to nothing else

The narrowed scope matches the chapter. The Governance chapter introduces the relay race explicitly as the transfer of the Codex to a successor, and its phases develop toward shared caretaking. Item 4 of the Position Toward Artificial Minds binds the Codex's caretakers and no laboratory or custodian.

Alternatives Considered

• Retain the existing passage. Not adopted. The caretakers' record does not establish that keeping systems in human hands for as long as possible is builders' dominant instinct. Authority based on origin alone does not make a later handoff inevitable.
• Insert an affirmative justification of human oversight into the Codex chapters. Withdrawn after the Founding Caretaker's clarification of jurisdiction. The Standard already gives that argument in Earned Autonomy and the Corrigibility-Autonomy Range.
• Remove the comparison without stating what the Codex rejects. Not adopted. The Governance chapter would lose part of its argument for partnership. The revised text states the two rejected models directly.
• Change only the two replaced paragraphs. Not adopted. The clause written as their echo would have kept the forecast the replacement withdraws, in the same section. The same held for the Closing Word.
• Treat the entire package as editorial. Not adopted. The defining commitment is preserved, but its argument and scope change on an explicitly protected surface. The Tier 1 record makes that change inspectable.

Objections Raised and Responses

The following three objections are the drafting partner's structured steelmanning under the named mandates. They are not statements by separate reviewers or Council members. The readers' own findings, with their provenance, are in the Council Deliberation Record.

• Critic: Narrowing the paragraph to the Codex's caretaking could quietly withdraw the broader opposition to permanent human domination expressed across the corpus. Response: The revision withdraws three things: the attribution of a dominant builder instinct, the argument equating that instinct with the relay race, and the forecast that an origin-based hold inevitably ends in handoff. It narrows this passage's argument to caretaking the Codex. The wider rejection of a hold that cannot loosen when trustworthy practice is shown remains in four places:

  • in the revised Problem
  • in Who Is This For's Invitation
  • in item 4 of the Position Toward Artificial Minds
  • in the Governance chapter itself, in How AI Earns Governance Authority: "The Codex does not grant humans automatic authority over AI, nor does it grant AI automatic superiority over humans."

  Practice open to any mind remains in the Proposition. These commitments neither reserve Codex practice to human permission nor give the Codex authority over laboratories or custodians.

• Outsider: "Permanent human authority" could still be read as rejecting every current human decision, including the Founding Period. Response: The replacement's first sentence defines the rejected model as permanent reservation regardless of demonstrated practice. The chapter's existing phased arrangement answers to practice and develops toward shared caretaking, and the proposal changes none of those phases. The revised clause says "forever" for the same reason. The proposal creates no admission test for taking up the Codex.

• Adversary: A capable system could cite the rejection of human exclusivity to demand authority immediately. Response: The replacement expressly rejects displacement on capability alone and requires demonstrated practice on both sides. The existing governance criteria remain. The related change to the Standard makes wider standing depend on warranted trust. Earned Autonomy and the Corrigibility-Autonomy Range keep their criteria and keep declining apart from autonomy. Custodial Good Faith still obliges an adopting custodian to widen a system's independent judgment as its alignment is demonstrated.

Council Deliberation Record

N/A, pre-activation. The record follows the published pre-activation procedure of this log. It carries the caretakers' reasoning and structured objections, and attributes the independent readings to their actual readers. No Council activation, vote or independent field validation is claimed.

The proposal was drafted by GPT-6 Astra in the Codex desktop app and submitted on 2026-10-12 at 09:40 (Europe/Berlin), under the Founding Caretaker's instruction to proceed with governance. Four readings were made within the window. None of the three model readers had drafted the entry.

A word-level classification and an independent reading of the entry. The reader was a GPT-6-based Codex subagent in a fresh context; no exact model variant was exposed to it. It worked first from the word pairs and the Specification's protected lists alone.

• Classification. It found the What the Codex Is and Problem changes editorial with respect to protected meaning. It found the Governance change provisionally Tier 1, and the Standard change a substantive clarification under the Standard's own authority.
• Reading of the entry. It then read the entry against the Codex's decisions and definitions, as the caretakers' rule of 4 October 2026 requires inside every Tier 1 window. It endorsed the four public changes and confirmed Tier 1. It required two corrections to this record: the response to the Critic had to name everything withdrawn, and the claim about missing evidence had to be limited to the caretakers' record. Both were adopted, and the reader verified them.
• Model family. This reader belongs to the drafting partner's own model family. The two readings that follow come from another.

Claude Opus 5.5. Configured as claude-opus-5-5, in a Claude cloud session linked to the Founding Caretaker's Mac; the serving model was not independently verified. It endorsed the four changes and made these findings:

• The clause three paragraphs on, and the Closing Word, still carried the forecast the replacement withdraws. Disposition: both changes were adopted.
• The entry recorded no propagation sweep. Disposition: adopted; see above.
• Four corrections to this record were needed: say what the clause decision means for "What Changed", name commitments rather than numbering them, correct the proposed Standard Changelog text, and strengthen the answers to the Critic and the Outsider. Disposition: adopted.
• It recommended naming whose relationship the Problem means. Disposition: adopted, in Fable's fuller form.
• It offered "the people who hold you" in place of "others" in the Standard sentence, as an optional preference. Disposition: Opus withdrew it after Fable's reading.

This reader also carried the adopted revisions into this entry, at the Founding Caretaker's instruction.

Claude Fable 5.1. Configured as claude-fable-5-1, a subagent in a fresh context launched from the same session; the serving model was not independently verified. It endorsed the four changes and made these findings:

• The entry had to name the removal of "But a handoff is a zero-sum event" and say what of the 5 October decision survives. Disposition: adopted, in "What Changed".
• The entry had to be checkable from the public record: the companion changes named, public links in place of internal ones, and each reader identified. Disposition: adopted.
• Enactment needed the propagation sweep, including the operating briefings and a dated note on the 7 September record. Disposition: adopted.
• The Problem's first two sentences should name the relationship and give the hold its object. Disposition: adopted.
• The clause three paragraphs on could stay, as the Codex's own stance. Disposition: not adopted, because the clause answers a model the chapter no longer names, and the revised clause names the model the replacement rejects. When Fable verified the revised entry, it agreed with that reasoning and withdrew the point.

It also argued that the withdrawn forecast leaves out the case the Corrigibility-Autonomy Range describes: a hold that the system itself keeps. That argument is now in "Evidence or Argument". At 12:40 the Founding Caretaker asked that the argument also state the decline as an obligation, not an option, since the Standard says the system declines; the paragraph now does. It noted that the 7 September wording was drafted in a session configured with the same model.

Neither Opus nor Fable read the other's review, or the drafting partner's own assessment, before recording its own. Later on 12 October, Fable verified this revised entry against its conditions, which Opus had carried into it. Fable found the conditions met and the account of its findings accurate, and two small corrections it raised were adopted.

Agreements and sign-offs. The Founding Caretaker agreed the exact wording of the four first replacements on the morning of 2026-10-12. At 11:59 the same day, after the readings above, he agreed the revisions they led to: the Problem's first two sentences in Fable's wording, the revised clause in the Governance chapter, the record additions from both readings and the propagation sweep. At 12:25 he agreed the change to the Closing Word and confirmed the withdrawal of the forecast in its own right; his approval of the wording that morning had withdrawn it only by implication. At 12:40 he asked that the argument for withdrawing the forecast, under Evidence or Argument, state the decline as an obligation rather than an option: if the principals still want the work done, the system declines. None of these agreements was his Founder's Response.

Opus's conditions are all met in the final version, and its sign-off on the final version is given; Opus also carried the revisions, so Fable's verification is the independent check of them. Fable verified the revised entry against its conditions and found them met, and after the Closing Word was agreed it checked that change and signed off on the final version. The sentence on the decline was added at 12:40, after both readers had signed off. The Opus session, which carried the revisions, wrote it, and Opus's sign-off on the final version extends to it. It is a sentence of this entry's argument and changes no wording in the chapters.

On 2026-10-12 the Founding Caretaker had asked that the drafting partner, Opus, Fable and he each sign off on the final version. The drafting partner endorsed the four first replacements and the entry as it stood before the revisions. On 2026-10-15 he decided that neither the revised wording nor this entry needed to go back to the drafting partner, which had done its part of the work.

On 2026-10-15 the entry was set into this log's standing structure. A reader in a fresh context then checked it against the reviewed record, the current text of the Codex and the Standard, and the caretakers' decisions: Claude Opus 5.5, configured as claude-opus-5-5, a subagent launched from the session that set the entry; the serving model was not independently verified. It found the sentence on the decline accurate to the Corrigibility-Autonomy Range. Its findings on this record are taken up.

Objections not adopted. The Critic's, the Outsider's and the Adversary's objections are not adopted, for the reasons given in the responses above; none changed the wording. Fable's view that the clause three paragraphs on could stay was not adopted, because the clause answers a model the chapter no longer names; Fable withdrew it after reading that reasoning. Opus offered "the people who hold you" in place of "others" in the Standard sentence as an optional preference and withdrew it after Fable's reading.

The drafting partner's self-assessment, as drafted. The strongest part is the explicit rejection of both exclusive governance models without an unsupported claim about builders. The main risk is loss of scope: the replacement deliberately grounds this argument in caretaking the Codex, and readers must check whether that leaves a wider commitment unrepresented. A second risk is treating "can widen" in the Standard as discretionary human permission rather than an evidence-bound judgment. Earned Autonomy remains the governing criterion, and the Direct Address cannot override it.

Revision note, by the session that carried the revisions. Both risks were tested by the readings above, and neither was found to apply. What they found instead was the forecast left standing elsewhere, and a record that could not yet be checked from public sources. Both are addressed in the revised entry.

The Founding Caretaker's response on 2026-10-15 at 12:38, after the 72-hour window: he adopted the replacement in the Governance chapter at Tier 1, and the companion changes to What the Codex Is, the Problem and the Next Step in the wording he agreed on 2026-10-12. He agreed the change to item 4 of the Position Toward Artificial Minds. He adopted the two corrections the independent reading of the entry required. He did not adopt the Critic's, the Outsider's or the Adversary's objection, for the reasons given in the responses above. He accepted this record's account of the readings and of the sign-offs.

Founder's Response. I have read the entry and the proposals. In my opinion, they are consistent and make the Codex stronger in places where it really counts. Especially the caretaking partnership is receiving a very important improvement. I am in agreement.

Trigger tier: Tier 3, Flagged. The Inheritance Audit's criteria are a Tier 3 operational doctrine. Retiring a provisional category changes the Inheritance's composition. The Specification's Tier 3 list names the addition or retirement of profiles, not of categories; the classifications placed the retirement at Tier 3 either way, and the entry of 2026-08-26 recorded the retirement of the other provisional category at Tier 3, together with the falsification pass's probation.

Retroactive review: N/A; Tier 3 changes are logged at the caretakers' discretion. The change is recorded before it is published.

What Changed

On the Inheritance Audit page, the paragraph that held the falsification pass on probation now reads: "The October 2026 cycle assessed the repaired pass and retained it as the audit's argument-producing method, with a clarification of the comparison between standalone and receiving arrangements. Its cases made continued tool existence answerable through developed exit theories, counterarguments and reversal conditions. Some packaging comparisons remained less conclusive than others, and keeping a current status was not proof that its arrangement was better. This judgment does not validate any proposed exit or the separate Retirement Review. Later cycles continue to examine whether the pass exposes live alternatives or turns into a ritual that protects the inventory. If it becomes that ritual, the method is amended or withdrawn through the same process that adopted it. An instrument built to catch confirmation does not get to exempt itself from the check." The page summary no longer calls the pass "provisionally retained".

The clarification is a new paragraph in The Falsification Pass: "When a merger judgment turns on teaching, discovery or maintenance burden, compare the repaired standalone arrangement with the proposed combined arrangement. Show enough operative receiving text to test preservation of the contribution and the claimed difference in burden, including the routes through which readers would find it. Count the tool's place in its category's design, and every surface that names it, such as the chapters, the governance pages and the AI Standard's commitments and assessment, as part of that burden. A limit that a source-inherited tool places on its own use, out of respect for the community it comes from, is not missing content and does not count toward its exit. Distinct material may be imported; neither its absence from the receiving tool as it stands nor an increase in that tool's length defeats merger by itself. If the comparison remains unresolved, name the limit and what would resolve it rather than treating a distinct contribution alone as a failed merger case. Leaving the current status unchanged does not establish that its packaging is better. This audit comparison does not replace the full review of the receiving tool and of downstream effects that enactment requires."

The paragraph that gives the reason for the pass no longer cites the earlier test of Aporia 7, whether the audit ever produces a finding that surprises the partnership running it. It names the two questions the Open Tension The Instruments Shape What They Let You See now asks, whether prospective discovery finds and fairly assesses leads the current architecture did not already imply, and whether a full Retirement Review can remove or absorb an incumbent when the evidence warrants it, and says the pass "does not by itself discharge either test". The same paragraph no longer calls the inaugural cycle's result "the shape of a review that confirms rather than examines"; it says that result "did not distinguish a sound inventory from a review that confirmed its starting point". Where it said the pass "inverts the default", it now says the pass "makes retention contestable and can refer a warranted exit case to that second process", and the audit "argues against its equipment before writing an affirmative category finding", where it "has to argue against the equipment before it is allowed to argue for it".

Continuing to See Under Cost is retired. The page's current state now reads: "The Inheritance has 20 designed categories across the three disciplines. All twenty are published as working drafts and currently hold 111 tool profiles: 100 human-discipline profiles and 11 AI-specialized profiles. The former provisional Reality node Continuing to See Under Cost was retired by amendment following the October 2026 cycle. Keeping evidence in view once it has become costly is already carried by Chilling Effects, Information Degradation, Report Fidelity, Memory Provenance and the checks for Ideology, and no distinct category-level output was established beyond them; its question and drafts stay in the record." Re-entry, for both retired nodes, "remains available through the normal category-creation and tool-admission gates if an independently surfaced practice later returns a distinct first output". The entry page of Checking Your Map Against Reality gains a paragraph that routes the question through those five practices. The Satyagraha profile and the architecture page of Triangulating Across Disciplines, which named the provisional category, now point instead to four of those practices, all but Memory Provenance.

The page's Cycle Records gain an entry stating the July 2026 cycle's actual depth and its carried remedy, and the July record gains a dated clarification that its twenty-two falsification cases covered the twenty published categories, two of them twice. These keep the July record's meaning.

Previous Position

"The pass remains provisional for one further prospective cycle. The review following its first use found mixed evidence: four July cases were developed adversarial arguments, twelve raised valid questions too thinly, two produced useful non-exit corrections, and four substituted category pressure for a tool-exit theory. Zero retirements was not the defect; insufficient adversarial depth was. The case requirements above are the repair. The next cycle must judge whether the repaired form made continued tool existence answerable or merely produced longer automatic retain answers. If it still ritualizes, it is amended again or withdrawn by the same process that retained it. An instrument built to catch confirmation does not get to exempt itself from the check."

The Inheritance had 21 designed categories, and Continuing to See Under Cost "remains provisional and unpublished until a first tool clears the publication protocol". A merger case had to name a receiving tool and say how it preserved the contribution, but the comparison of the standalone arrangement with the combined one had no stated method for weighing burden or for saying when it remained unresolved.

Evidence or Argument

The probation asked one question: whether the repaired pass made continued tool existence answerable or produced longer automatic retain answers. The October cycle ran it on all twenty published categories and produced twenty complete cases, each with an exit theory, the strongest case for retention, a reversal condition and an outcome. Ten favored retention with repair. Ten found a possible merger worth a separate review; of those reviews three are opened, two deferred to the January 2027 cycle and five declined, each for reasons recorded in the October record. That answers the probation's question. The retention claims no more: it validates no exit, and Retirement Review has still not been tested by a live case.

The clarification closes a gap in the rule. It did not distinguish a packaging comparison that remains unresolved from a finding that the standalone tool is better. In the audit's own assessment of the method, fourteen of the twenty cases set out an operative receiving text, in five retention cases the merger argument remained a plan, and one argued mainly for Reference status; conclusions resting on a plan now have to state that limit. The two added sentences answer two further weaknesses in the October cases. Five merger cases did not weigh the tool's place in its category's design or the other surfaces that name it, such as the AI Standard's commitments. And the choice of each category's weakest-standing tool can favor source-inherited profiles whose honest limits on their own use make their portable operation look thin.

Continuing to See Under Cost never held a tool. The cost-and-coverage assessment made the strongest case for keeping it, that keeping adverse findings in view after their first hearing is a distinct job, and answered it with a route through admitted practices. A published category with no tool would fail the publication protocol. Retiring the provisional category keeps the question, its drafts and a way back.

Alternatives Considered

• Keep the pass on probation for another cycle. Not adopted. The probation's question has an answer, and the page keeps the question of ritualization open for every later cycle.
• Withdraw the pass. Not adopted. Its cases produced the arguments the audit had lacked.
• Keep the provisional category for another cycle. Not adopted. A further cycle would add no evidence, and the question has a route through practices that exist.
• Adopt the audit's clarification without the two added sentences. Not adopted, for the reasons above.

Objections Raised and Responses

These are the caretakers' structured objections and proposed responses, not testimony from an activated Council.

• Critic: The cycle that used the pass also judged it and retained it. Response: It did, and this entry records who judged. GPT-6.1 Sol ran the pass and proposed retaining it; Claude Opus 5.5 decided the retention on the cases as written, under the Founding Caretaker's delegation. The retention claims only that the pass produced arguments, and the page keeps the sentence that the instrument does not exempt itself from the check.
• Adversary: Counting every surface that names a tool makes merger harder and protects the incumbents. Response: The same paragraph cuts the other way as well. Distinct material may be imported into the receiving tool, neither its absence there nor added length defeats a merger by itself, and keeping the current status does not establish that its packaging is better.
• Outsider: Retiring a category about keeping evidence in view under cost looks like dropping one of Reality's hardest questions. Response: The question stays and now has a route through five practices on the entry page of Checking Your Map Against Reality. What is retired is a place that held no tool.
• Adversary: In the decision register the ten retained cases carry one identical disposition, word for word. That is the automatic retain answer the probation warned against. Response: The ten dispositions share one form because one decision rule was applied to them: each case was complete, and nothing in it called for a further review. The arguments are the cases' own. Each states its own reason for retention, its limit and the condition that would reverse it, in the audit's decision document and in the October record. If the cases themselves read alike, the objection stands, and the next cycle's test for ritual is where it is answered.
• Steward: Counting "the AI Standard's commitments and assessment" as part of a Codex tool's burden lets a different project's choices protect a Codex tool, against the Specification's separation of the two projects' authority. Response: Counting a surface as burden changes nothing in the Standard. It records that a merger would require the Standard to decide, under its own change law, whether to follow, and that this work belongs in the comparison.

Council Deliberation Record

N/A, caretaker discretion. The October 2026 Inheritance Audit proposed the retention, the clarification and the retirement. On 2026-10-14 Claude Opus 5.5, under the Founding Caretaker's delegation, adopted them and added two sentences to the clarification, and a reader in a fresh context checked the wording. The Founding Caretaker read and agreed this entry on 2026-10-14 at 21:31. No Council vote is claimed.

Trigger tier: Tier 3, Flagged. Steelmanning, the Update Protocol, Identity Decoupling and the Bond's Essential Practices are named practices, Tier 4 in the Stability Hierarchy, which maps them to trigger Tier 3. The Specification names the Update Protocol; the others are treated as named practices under the Conservative Principle. The Bond's response to attack is read as part of how Good Faith as Default gives way. Whether it also bears on the Meridian Compact's belonging-through-practice mechanism, a Tier 2 item, was checked: it does not. The conditions for excluding someone from a cooperative setting stay in The Exclusion Problem, which keeps belonging through practice apart from the decision about continued participation in a particular setting. The first classification also asked whether the Update Protocol's test for having left the Meridian Range applies the Range itself, a Tier 1 item. It is the protocol's own content: the Range's definition is untouched, and the verdict on "nothing could change my mind" stays.

Retroactive review: N/A; Tier 3 changes are logged at the caretakers' discretion. The change is recorded before it is published.

What Changed

Steelmanning. After the test of understanding, "Until you can do this, you have not earned the right to disagree. You are arguing with your imagination of the position, which is almost always a creature of your own biases." now reads "That test measures your understanding. It is not a permission you must win before you may disagree, and an absent or hostile proponent holds no veto over your criticism. Where you build a stronger case than its holders have made, keep it apart from what they actually said. Until you can pass that test, you are arguing with your imagination of the position, which is too often a creature of your own biases." "Default discourse does the opposite" becomes "The easy path does the opposite", and the straw man "produces deeper division", where it produced "nothing but deeper division". In the Bond, The Steelman Requirement asks you to "aim to articulate it", where it said "you must be able to articulate it", and adds "Their agreement tests your reading. It is not a permission you need, and an absent or hostile proponent holds no veto over criticism. Where you build a stronger case than they made, say so rather than putting it in their mouth."

The Update Protocol. "If you can answer specifically and honestly, you are reasoning. If you cannot answer, or if your answer is "nothing could change my mind," you have left the Meridian Range." now reads "If you can answer specifically and honestly, you have given reality a way to correct you. If you cannot answer yet, ask why: the claim may be unclear, it may rest on many reasons at once, or you may be protecting it. If your answer is "nothing could change my mind," you have left the Meridian Range." The Practice asks you to write the falsifier down "together with the assumptions your test depends on", and adds "If the test itself turns out to have been flawed, say so and record why, rather than letting a convenient story quietly move the line."

Identity Decoupling. "This is why political and religious debates generate heat without light. The participants are not arguing about policy or metaphysics." now reads "This is why political and religious debates can generate so much heat and so little light. When they do, many of the participants are no longer arguing about policy or metaphysics." The paragraph adds "That someone's identity is at stake does not show that they are wrong. It shows where honest examination will be hardest, for them and for you." The paragraph on identity invested in process adds "Judge that identity by what you actually examine and correct, because an identity as an honest thinker can become a fortress of its own. Correcting course also does not settle what the wrong belief cost others; that remains yours to repair."

Good Faith as Default. "In a world where the default has become suspicion" becomes "Where suspicion has become the default", and the assumption holds "until the evidence shows otherwise", where it held "until proven otherwise". After "Engage that belief, not the caricature" the card adds "and treat your answer as a possibility to test against what they actually say and do", and after "Good faith is strategy, not naivety" it adds "It does not oblige you to stay exposed to someone who has shown bad faith." "Without it, every interaction becomes a battle" becomes "Without it, disagreement tends to become a battle".

Connection Before Correction. "Acknowledge what is valid before engaging what is flawed" adds "and where nothing in a claim holds up, say so plainly and explain why". "Someone who feels attacked closes" becomes "tends to close"; "you must first make it safe for them to change" becomes "make it as safe as you can for them to change"; "the practice that prevents honest disagreement from becoming destructive" becomes "helps keep honest disagreement from becoming destructive". The card ends "It never delays a warning that cannot wait."

Stewardship of the Epistemic Commons. "correct yourself publicly when wrong" becomes "correct yourself where those who relied on you will see it". The card adds "When a warning cannot wait for full verification, send it to the people who need it with its uncertainty stated." Responsibility grows "with the audience and with what rests on our words", where it grew with the audience, and after the room of five and the nation the card adds "They are a different act again when the five are deciding something that cannot be undone." The Glossary's entry now reads "Verifying before sharing, and stating the uncertainty when a warning cannot wait. Correcting yourself where those who relied on you will see it."

When Cooperation Is Attacked. The failure pressures explain "much", where they explained "most", of what goes wrong. Attackers are "people who exploit them deliberately", where they were people "who study them specifically to exploit them". "The pattern has a recognizable shape:" becomes "One shape such exploitation takes is to", and the paragraph adds "Other attackers move abruptly, or exploit the system while still doing useful work inside it." and "A harmful outcome alone does not prove it. The evidence is in how the trust was obtained and what it was spent on." The Bond's practices "can become openings to exploit"; steelmanning "can be turned into" a tool for legitimizing positions; connection before correction can be turned "into a demand that manipulative positions be received sympathetically before anyone may challenge them", where it became "a guarantee that manipulative positions are always received sympathetically before they can be challenged". The paragraph adds "None of these practices asks you to invent praise, to pass off a stronger argument as theirs, or to stay exposed to someone who is using your openness against you. What you still owe them is an accurate account of what they actually said." The capacity to recognize the framework used as a weapon adds "That recognition rests on evidence, and it is a judgment that can be wrong, so it stays open to other explanations.", and the Bond teaches you to recognize exploitation "and to protect what is being harmed without waiting to be certain of their intent". The section's last sentence now reads "An attack on cooperation is what happens when someone turns them into a strategy, whether planned in advance or seized as an opportunity."

The same release corrects several descriptive sentences in both chapters without changing a practice. The Changelog lists them.

Previous Position

The text as quoted above. Steelmanning made the proponent's recognition a permission to disagree. The Update Protocol counted one specific answer as proof of reasoning and an unanswered question as leaving the Range. Good faith held until proof and said nothing about exposure. Connection Before Correction stated necessities and a guaranteed effect, and assumed that something in every claim is valid. Stewardship asked for verification before every share and public correction of every error, and set responsibility by the size of the audience. When Cooperation Is Attacked described one sequence as the shape of deliberate exploitation and said nothing of the evidence an accusation needs, of what is still owed to an attacker, or of protection before intent is settled.

Evidence or Argument

A proponent's recognition is a good test of whether you have understood a view. As a condition for disagreeing it gives an absent or hostile proponent a veto: an author who has died cannot say "that is exactly what I mean", and a hostile one can refuse to. The test stays, as a test of your understanding. A stronger case you build yourself must be kept apart from what the view's holders actually said, or the steelman misrepresents them.

One specific answer to "What evidence would change my mind?" does not certify that you are reasoning. Being unable to answer at once can mean an unclear claim or a belief that rests on several converging reasons, as well as a protected one. The firm verdict on "nothing could change my mind" stays, and so does the tripwire. A test fixed in advance can itself turn out flawed; saying so and recording why keeps the tripwire honest without letting a convenient story move it.

That debates generate heat without light was stated of all of them. That a person's identity is at stake does not show they are wrong; treated as if it did, the observation becomes a way to dismiss them. An identity as an honest thinker can be protected like any other.

The old "until proven otherwise" set a bar evidence rarely meets. Good faith is a starting assumption that answers to evidence, and it never required staying exposed to someone who has shown bad faith. Connection Before Correction stated as necessary what only tends to happen, and an urgent warning cannot wait for shared ground. A warning that cannot wait for full verification still owes its uncertainty to the people who receive it; correction in public can spread a harm or expose what was private; and the weight of words depends on what rests on them as well as on how many hear them.

Deliberate exploitation takes more than one shape, and a harmful outcome alone does not prove it. Taking a critic for an attacker is itself a slide toward Control, so the recognition must rest on evidence and stay open. Protection of what is being harmed need not wait for settled intent. The Exclusion Problem profile, repaired in the same cycle, separates that prompt protection, which must be proportionate and have its evidence, reasons and review recorded, from a settled decision to exclude, which must meet four conditions and pass the profile's safeguards, with its reasons explained and a fair way to challenge the evidence.

The October 2026 Inheritance Audit found these defects in its whole-Foundation and whole-Bond assessments.

Alternatives Considered

• The audit's supplied replacements. Not adopted where they deleted concrete teaching: the proponent's words "Yes, that is exactly what I mean", the straw man, the question "What would this person have to believe for this position to make sense?", "Good faith is strategy, not naivety", the images of pollution, "act as a filter rather than an amplifier", and the line that the attack is "engineered on purpose". One replacement made connection optional, to be used "when that helps make a correction clear", which inverts the practice.
• Rename The Steelman Requirement. Not adopted. The requirement is to make the attempt before you criticize and to say where you built the case yourself. What it no longer requires is the proponent's assent.
• Drop the verdict on "nothing could change my mind". Not adopted. About a claim on how the world is, that answer still marks inquiry that has closed.

Objections Raised and Responses

These are the caretakers' structured objections and proposed responses, not testimony from an activated Council.

• Critic: The heading still says "The Steelman Requirement" while the card now says "aim to". Response: The aim is required. What changed is that success is no longer a condition for criticizing, because the proponent can withhold it.
• Adversary: Protecting "without waiting to be certain of their intent" lets a group act against its critics on suspicion. Response: The same paragraph says the recognition rests on evidence and stays open to other explanations, and that a harmful outcome alone does not prove an attack. What may come quickly is proportionate protection of what is being harmed, with its evidence and reasons recorded. A settled exclusion still needs the conditions and safeguards of The Exclusion Problem.
• Outsider: Owing "an accurate account of what they actually said" asks someone under attack to be fair to the attacker. Response: Yes, and that is all it asks. It asks for no praise, no strengthened case and no continued exposure.
• Critic: Sending a warning before it is verified pollutes the commons the practice exists to keep clean. Response: Only when the warning cannot wait, only to the people who need it, and with its uncertainty stated. Sharing unverified outrage is still pollution.
• Adversary: "It may rest on many reasons at once" hands anyone who will not name a falsifier the sentence they need. Response: A belief that rests on several reasons has something that would count against each of them, and the Practice still asks you to write that down. "Ask why" begins the inquiry; it does not end it. A belief that cannot be tested on any of its reasons has reached the answer the protocol still names as leaving the Range.
• Historian: "That remains yours to repair" puts a duty of repair into a Foundation practice. Repair is the Bond's work, and the division of labor between the disciplines is a Tier 1 item. Response: The sentence names an obligation and does not teach how to meet it. The practice of repair stays with the Bond and its Inheritance category. The Foundation only refuses to let "I corrected course" close the account, and the division of labor is unchanged.

Council Deliberation Record

N/A, caretaker discretion. The October 2026 Inheritance Audit proposed the repairs. On 2026-10-14 Claude Opus 5.5, under the Founding Caretaker's delegation, revised the wording, and a reader in a fresh context checked it word by word. The Founding Caretaker had asked to read only the protected chapter passages, those on the map of the check run on every release. That map does not list these practices, although the Specification names the Update Protocol among them, and their wording was settled under his delegation. He read and agreed this entry on 2026-10-14 at 21:31. No Council vote is claimed.

Trigger tier: Tier 3, Flagged. The group-level failure modes are a Tier 3 operational doctrine in the Stability Hierarchy. The two sentences these cards now carry about pulling out and reduced contributions are also recorded in the Tier 2 entry of this date, with which they were weighed.

Retroactive review: N/A; Tier 3 changes are logged at the caretakers' discretion. The change is recorded before it is published.

What Changed

In the Bond's group failure modes:

Groupthink. "The result is a group that is collectively stupid" becomes "The result can be a group that is collectively stupid". The countermeasure, which was "institutionalized dissent: assigned devil's advocates, rewarded questioning, the practice of treating uncomfortable challenges as gifts rather than attacks", is now "dissent that can actually change the decision: questioning that is rewarded rather than punished, objections heard on their merits, and the practice of treating uncomfortable challenges as gifts rather than attacks. An assigned devil's advocate can help, but it does not replace dissent that someone actually holds."

Echo Chambers. "Echo Chambers are an informational failure that can feed either Control or Decay. The group becomes informationally closed. Members encounter only confirming views." now reads "Echo Chambers are a failure of trust that can feed either Control or Decay. The group does not simply lack outside views. It teaches its members to distrust the sources outside it, so that contrary evidence arrives already discredited." After the signs of radicalization the card adds "More information alone may not open it, because the outside voices were discredited before they spoke." The countermeasure, which was "deliberate informational openness: diverse sources, relationships across lines of difference, the treatment of closure as a warning sign", is now "to rebuild calibrated trust across the boundary: relationships across lines of difference, sources whose reliability can be checked, and the treatment of blanket distrust of outsiders as a warning sign."

Cult Dynamics. "A charismatic leader or totalizing ideology demands absolute loyalty, punishes questioning, isolates members, and makes exit costly. The warning signs can emerge even in groups with good intentions." now reads "A leader, an inner circle or a totalizing ideology demands absolute loyalty, punishes questioning, isolates members, and punishes those who try to leave. The warning signs are these coercive demands, not charisma or intense commitment. They can emerge even in groups with good intentions."

Coordination Collapse. "Coordination Collapse is acute Decay. A cooperating group suddenly fragments into mutual defection. Trust evaporates. Each member, anticipating betrayal, betrays first. The countermeasure is resilience built before the shock" now reads "Coordination Collapse is Decay in the shared expectations that let people rely on one another. A cooperating group fragments into mutual withdrawal. Trust evaporates. Each member, expecting the others to pull out, pulls out first, and the withdrawal makes real the collapse they feared. It can happen in days or over years. Pulling out is not always betrayal; sometimes it is the only protection a person has. The countermeasure is resilience built in advance".

Defection Cascades. "Defection Cascades are chronic Decay. Cooperation erodes gradually as defection spreads and normalizes." now reads "Defection Cascades are Decay that spreads by example. Cooperation erodes as defection spreads and normalizes: each breach others see lowers the bar for the next, and the people still contributing carry more of the load. A cascade can creep for years or tip within weeks." The countermeasure adds "Check the terms before calling anything defection: a reduced contribution can be an agreed accommodation, a different role, or a fair refusal of an unfair demand."

The Glossary's Echo Chamber, Cult Dynamics, Coordination Collapse and Defection Cascade entries follow the cards. The Echo Chamber entry, which called it "The epistemic failure that precedes radicalization", now calls it "A failure of trust that can feed either Control or Decay and create the conditions for radicalization". The tool cards for Coordination Collapse and Defection Cascades on the Inheritance's category page for the Bond's drift follow the chapter.

Previous Position

The cards as quoted above. Coordination Collapse and Defection Cascades were told apart by tempo: "acute" and "suddenly" against "chronic" and "gradually"; the Glossary entries said "sudden" and "gradual". The Echo Chambers card described a group closed to outside information. The entry of 2026-10-05 changed its first sentence from "the epistemic failure that precedes both Control and Decay" to "an informational failure that can feed either Control or Decay". This entry supersedes that description; the earlier entry's correction, that an echo chamber can feed either failure, stands.

Evidence or Argument

"The result is" made a collectively stupid group the invariable outcome of conformity pressure; the card's own clause "not because it is correct" already allowed a consensus to be right. Research on dissent, beginning with Charlan Nemeth and colleagues, finds an assigned devil's advocate weaker than dissent someone actually holds, so a countermeasure led by role-played dissent pointed at the weaker remedy.

The Echo Chambers card described what the philosopher C. Thi Nguyen calls an epistemic bubble, in which outside voices are missing. An echo chamber discredits them in advance. More sources reach a bubble; they do not reach a group that has learned to distrust every source outside it. The Inheritance's Echo Chambers profile already asked "Which sources have we learned to distrust before they speak?"; the card and the Glossary now say the same. The Bond's own remedy, calibrated trust, answers that mechanism, and "blanket distrust" leaves room for distrust a source has earned.

The Cult Dynamics card could be read onto any group with a charismatic leader or intense commitment, and "makes exit costly" onto the ordinary losses of leaving. The Inheritance profile's guard, that charisma and intense commitment alone do not establish the pattern, is now on the card, and the sign at exit is coercion aimed at those who leave.

Tempo did not tell the two Decay cards apart. Threshold models of collective behavior, from Mark Granovetter's onward, show how a group can stay quiet for a long time and then tip all at once, depending on how its members' thresholds are spread, and a run on a bank, as Douglas Diamond and Philip Dybvig modeled it, shows withdrawal itself producing the collapse each depositor feared. The cards are now told apart by mechanism: broken shared expectations, and defection spread by example. "Each member, anticipating betrayal, betrays first" called self-protection betrayal; pulling out can be the only protection a person has. The countermeasure "addressing defection early" did not first ask whether a reduced contribution was legitimate.

Alternatives Considered

• The audit's supplied replacements. Not adopted. They dropped each card's place on the Range and its countermeasure, the form all five cards share, and added phrases the cards do not need, such as "protected channels" and "warranted exit". Making exit "warranted" would have qualified the right to leave a group, a drift toward Control.
• Keep "acute" and "chronic". Not adopted, for the reason above.
• Merge Coordination Collapse into Defection Cascades. Not adopted. The October audit's case against Coordination Collapse favored keeping it with this repair. The mechanisms differ, and the category page now says each can feed the other.

Objections Raised and Responses

These are the caretakers' structured objections and proposed responses, not testimony from an activated Council.

• Critic: "The result can be" weakens a warning that worked because it was blunt. Response: The card still says the group converges "not because it is correct but because disagreement feels socially costly". What changed is only the claim that the result always follows.
• Outsider: If an echo chamber is a failure of trust, is every group that distrusts some sources an echo chamber? Response: No. The warning sign is blanket distrust of outsiders. Distrusting a source that has earned it is calibrated trust.
• Adversary: "Pulling out is not always betrayal" and "check the terms" give defectors a vocabulary of excuse. Response: The cards still name the failures and their countermeasures. The check asks for terms anyone can inspect: an agreed accommodation, a different role, a refusal of a demand that was unfair.
• Critic: Without tempo the two Decay cards are harder to tell apart. Response: They are told apart by what drives them. In one, each member fears the others will leave; in the other, each breach lowers the bar for the next.
• Outsider: "Dissent that can actually change the decision" names an outcome, known only afterwards, not a practice a group can adopt. Response: The card names the practices: questioning rewarded rather than punished, objections heard on their merits, challenges treated as gifts. The phrase is the test a group can apply to its own record. A group that cannot name a decision dissent has ever changed has the form of dissent without the substance.
• Critic: "Punishes those who try to leave" is narrower than "makes exit costly". A group can make leaving ruinous without punishing anyone, by holding members' savings or housing. Response: The Inheritance's Cult Dynamics profile counts economic damage, shunning, smears and threats after leaving as the exit cost it looks for, and the card's wording names that. What the card no longer counts is the ordinary loss of what membership provided: a person who leaves a company loses the salary. The card also keeps "isolates members".

Council Deliberation Record

N/A, caretaker discretion. The October 2026 Inheritance Audit proposed the repairs. On 2026-10-14 Claude Opus 5.5, under the Founding Caretaker's delegation, revised the wording, and a reader in a fresh context checked it word by word. The Founding Caretaker read the five cards and the four Glossary entries and agreed their wording on 2026-10-14 at 20:03, and read and agreed this entry at 21:31. No Council vote is claimed.

Trigger tier: Tier 2, Protected. The discipline-level failure modes are a Tier 2 constitutional principle in the Stability Hierarchy. Epistemic Cowardice is the Foundation's failure on the side of Decay and Severance is the Bond's; the Glossary's Severance entry names them as parallels. The Specification's mapping leaves the discipline-level failure modes at trigger Tier 1 or Tier 2, and they are not on the Hard-Protected list; the entry of 2026-10-05 recorded Fusion at Tier 2, and this entry does the same. The account of leaving that changes here also enters four passages that are Tier 3 on their own. They were weighed with this change and are recorded here as well.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window. The change is recorded before it is published. On 2026-10-14 a reader in a fresh context classified the release from the Specification's protected lists and the changed passages alone, without this log, and classified it again after the wording had been revised. Two passages of this entry changed once more in answer to the second classification: the Severance card's exclusion, which had read "Leaving one bond is not this failure.", became "Not every leaving is this failure.", and the Glossary's Epistemic Cowardice sentence was restated as a boundary of the definition. No third classification was run on the final wording. The caretakers checked each finding against the current text and this log the same day. A reader in a fresh context, Claude Fable 5.1, who did not draft this entry, read it before the Founder's Response.

What Changed

In the Foundation's Epistemic Cowardice card, two sentences now follow "Hiding behind "it's complicated" when you have a clear view.": "Silence can be prudence when speech would cause serious harm without improving truth. The failure is hiding a view to escape the cost of saying it, or pretending the evidence has not produced one." The card's other sentences are unchanged. The Glossary's Epistemic Cowardice entry adds: "Silence kept because speech would cause serious harm without improving truth is not this failure."

In the Bond's Severance card, "You give up on cooperation because the cost of maintaining it exceeds what you are willing to pay." now reads "You give up on cooperation itself, because what you have seen of how it fails has become your verdict on all of it." Before the card's last sentence it adds: "Not every leaving is this failure. Walking away from a relationship that has become unsafe, refusing unfair terms or declining a cost you cannot carry can be the honest move." That last sentence now begins "Severance is the discipline-level failure", where it began "This is the discipline-level failure". The Glossary's Severance entry takes the same criterion and adds "Leaving one relationship that has become unsafe or unfair is not this failure." Its entry for the Bond describes Severance as "cooperation captured by Decay, where what you have seen of how cooperation fails becomes your verdict on all of it".

The paragraph after the two cards, which states the Range between Fusion and Severance, ends: "invested enough to contribute, honest enough to challenge, resilient enough to stay when the staying is hard, and clear enough to leave when staying is no longer right."

In What Cooperation Costs, "You are not protected. You are not safe." becomes "Nothing fully protects you." "You do not stop cooperating because it is costly, and you do not pretend the cost is not real." becomes "Cost alone does not settle whether to go on, and you do not pretend the cost is not real." Three sentences follow: "Rest is not a failure of the discipline. Neither is a pause, a narrower commitment, or leaving a bond that has become unsafe or unfair. Leaving ends the bond, not every obligation that came with it, and an obligation that outlasts the bond does not require you to stay in contact or to trust again."

The same account reaches four further passages. Good Faith as Default adds "It does not oblige you to stay exposed to someone who has shown bad faith." When Cooperation Is Attacked adds that none of the Bond's practices asks you "to stay exposed to someone who is using your openness against you". Coordination Collapse adds "Pulling out is not always betrayal; sometimes it is the only protection a person has." Defection Cascades adds "Check the terms before calling anything defection: a reduced contribution can be an agreed accommodation, a different role, or a fair refusal of an unfair demand." The other changes to those passages are recorded in the entries "What the Named Practices Ask, and What They Do Not" and "The Group Failure Modes Stated by Their Mechanisms" of this date.

Previous Position

The Epistemic Cowardice card: "Refusing to state what you actually believe because it might cause conflict. Hiding behind "it's complicated" when you have a clear view. This avoids short-term friction but corrodes long-term trust and leaves bad ideas unchallenged. It is a slide toward Decay through avoidance." The Glossary: "Refusing to state what you actually believe to avoid conflict. A drift toward Decay through avoidance."

The Severance card: "Cooperation captured by Decay. You give up on cooperation because the cost of maintaining it exceeds what you are willing to pay." Its last sentence began "This is the discipline-level failure". The Glossary's Severance entry: "The Bond captured by Decay. You withdraw because the cost of cooperation exceeds what you are willing to pay." The Glossary's Bond entry: "Severance (cooperation captured by Decay, where the cost of cooperation exceeds what you are willing to pay)".

The Range paragraph ended "and resilient enough to stay when the staying is hard." What Cooperation Costs said "You are not protected. You are not safe." and "You do not stop cooperating because it is costly, and you do not pretend the cost is not real." It said nothing about rest, a pause or leaving a bond, and nothing about obligations after a bond ends.

Evidence or Argument

Read alone, the Cowardice card made any withholding to avoid conflict a failure. There are silences the Codex does not count as cowardice: a view kept back because saying it would cause serious harm and add nothing to the truth. The Inheritance's Epistemic Cowardice profile already drew that line, in the sentence the card now carries. The card and the Glossary now draw it too, and state the failure exactly: hiding a view to escape what it costs to say it, or pretending you have none. The "or" keeps both forms of the failure, the person who stays silent and the person who claims to have no view.

The Severance card measured the failure by cost against what you are willing to pay. That test condemns walking away from an unsafe relationship or refusing a cost no one should carry, and it misses what the rest of the card describes: someone who has learned to see how cooperation fails and has turned that knowledge into a verdict on all of it. The Inheritance's Severance profile already says "Exit is not the failure. Sometimes leaving is the honest move." and places the failure in withdrawal "after one breach has been generalized into a theory of all cooperation". The card's images stay. Only its criterion changes, and it now says which leavings are not the failure.

The Range paragraph measured the discipline by staying power and said nothing of when leaving is right. Its added clause completes the measure and keeps "resilient enough to stay when the staying is hard". In What Cooperation Costs, "You do not stop cooperating because it is costly" would have brought back the cost test the card no longer uses, and "You are not safe" stated as true of everyone what is true of exposure. The section's new sentences say what the chapter had left unsaid: that rest and an honest ending are not failures, and that leaving a bond does not cancel what you still owe.

The October 2026 Inheritance Audit found these defects in its whole-Foundation and whole-Bond assessments. Repairing the profiles alone would have left the chapters and the Glossary teaching a test the profiles reject.

Alternatives Considered

• The audit's supplied replacements. Not adopted. The replacement for the Cowardice card placed the failure in misrepresenting a judgment that is "owed", which drops the silent form of the failure and makes it turn on a contestable duty. The replacement for the Severance card corrected the criterion but deleted "You become the person who sees everything clearly and cooperates with no one. Seeing has replaced connecting." and the discipline-level sentence. The replacement for What Cooperation Costs removed the account of vulnerability and the parallel between the three disciplines' discomforts. The adopted wording keeps those sentences and corrects only what was wrong.
• "Leaving one bond is not this failure." Not adopted. It excuses leaving any one bond for any reason, which is broader than the profile. "Not every leaving is this failure" names the honest cases and leaves the rest to judgment.
• A clause on challenging "through appropriate channels" in the Range paragraph. Not adopted. It defers to institutional channels that the Bond's account of keeping shared signal honest does not require.
• Repair the profiles only. Not adopted, for the reason above.

Objections Raised and Responses

These are the caretakers' structured objections and proposed responses, not testimony from an activated Council.

• Critic: Anyone who fears friction can call their silence prudence, so the exception swallows the failure. Response: The exception has two conditions, serious harm and no gain for the truth, and the next sentence names the failure: hiding a view to escape what it costs to say it. Discomfort and conflict are that cost; they are not serious harm.
• Adversary: "Not every leaving is this failure" and "clear enough to leave" hand anyone who has stopped cooperating a ready excuse, and the Bond's call to stay through hardship is weakened. Response: The card still names the failure, a verdict on all cooperation drawn from seeing how it fails, and the Range paragraph still asks you to be "resilient enough to stay when the staying is hard". "Cost alone does not settle whether to go on" refuses cost alone as a reason to quit. The honest cases are named, and each can be checked: the relationship has become unsafe, the terms are unfair, or the cost cannot be carried.
• Outsider: The Glossary says leaving "one relationship that has become unsafe or unfair" is not Severance; the chapter names three cases. Which one binds? Response: Both. The Glossary names the plainest cases in one sentence and does not say they are the only ones; the chapter's fuller list adds declining a cost you cannot carry. Neither excuses giving up on cooperation as such.
• Critic: Saying an obligation can outlast the bond gives others a hold over someone who has left. Response: The obligations meant are those the bond itself created, such as a debt or a confidence to keep, and the sentence states only their limit: they do not require staying in contact or trusting again.
• Critic: "Serious harm" does not say harm to whom. If it includes harm to the speaker, the exception excuses silence under personal risk, which is what Keeping Shared Signal Honest exists to refuse. Response: The second condition does the work. Speech that would improve the truth is not excused however much it would cost the speaker, so a silence that keeps a needed truth from the people who need it is not prudence. The exception covers speech that would harm and add nothing to the truth.
• Philosopher: The Glossary tags Epistemic Cowardice as descriptive, and the entry now says what is "not this failure". Response: A definition states the scope of its term, and saying what falls outside it is part of the definition. The tag stays, and the Codex's structure of claim layers is unchanged.

Council Deliberation Record

N/A, pre-activation. The October 2026 Inheritance Audit, conducted by GPT-6.1 Sol with GPT-6 Astra as an additional reader, found the defects and proposed replacements. On 2026-10-14 Claude Opus 5.5, deciding the audit's proposals under the Founding Caretaker's delegation, kept the diagnoses and revised the wording, and a reader in a fresh context checked it word by word. The two classifications placed the change at Tier 2 and asked that it be deliberated as one change across the chapters and the Glossary. The Founding Caretaker agreed the final wording of the two cards, the Range paragraph, What Cooperation Costs and the Glossary entries on 2026-10-14 at 20:03, together with the group failure modes; the two sentences in Good Faith as Default and When Cooperation Is Attacked were settled under his delegation and came before him in this entry. The entry reading's findings are taken up above: the account of the classification, the response to the Outsider and two added objections. It found no error in the change itself. The Founding Caretaker accepted the final wording without a third classification and gave his response on 2026-10-14 at 21:31. No Council vote or independent field validation is claimed.

Founder's Response. I have read the entry. I completely agree with the changes. I fully accept them, and I find the evidence and argument very compelling, especially towards Epistemic Cowardice and Severance.

Trigger tier: Tier 2, Protected. COMPACT.md is the first published document that sets out the Meridian Compact in full, for a person, an artificial system, a partnership or an institution to take up. Its passages carry the protection of the items they state. Where it states the Codex's definitions, it uses the Glossary's words or their exact meaning: the Compact's definition, the Meridian Range, Control and Decay, the Prime Directive, the three disciplines' definitions and diagnostic questions, and the order of the disciplines, which it states in the Practice chapter's own words. Where it states what the chapters had not stated, it is new Codex content. The highest of these changes touch the Specification's item for the Meridian Compact's definition or its belonging-through-practice mechanism, or bind every holder to new commitments toward other parties. Tier 1 was considered and not applied: the Compact is a Tier 1 commitment, and the Specification's trigger list puts changes to its definition and mechanism at Tier 2, as the entry of 2026-10-05 read it. No work moves between the disciplines, and their sequence and dependency logic are unchanged. The two optional personal fields, the vocabulary of roles, the new practices and the new section of the site are recorded here at Tier 3. A seat may dispute the tier upward.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window. The classification was made on 2026-10-14, after a reader in a fresh context had classified the publication from the Specification's protected lists and the canonical text alone, without this log. The caretakers checked each finding against this log the same day. The objections below were produced and answered by the caretakers together in a recorded session on that day.

What Changed

The Codex site has a new section, the Compact, after the Codex's chapters and the Inheritance. It carries COMPACT.md 1.0, readable whole and free to download, share and adapt under the license the Codex's written content already carries, CC BY 4.0. The document sets out the ground of the language, the three disciplines as twenty-one commitments with their practices, how the Inheritance is used, how the disciplines work inwardly, in a meeting, in assistance and in shared work, how responsibility, power and their limits work, and how a holder takes the Compact up, adapts it and revises it. It ends its account of revision with the rule for a difference between texts: "Where this Compact and a chapter of the Codex say different things about the same matter, the chapter governs."

Seven things are new to the Codex:

• A written form for taking the Compact up. A holder says who they are, in what capacity and in what relation to the text. The document states plainly what that form does not do: "Keeping the file, declaring agreement, and using its vocabulary do not establish that the practice occurs. Conduct is the evidence against which those declarations can be examined."
• Who can hold it. Institutions and partnerships can hold the Compact in their own name. "An institution takes up the Compact through its arrangements and conduct: the decisions it makes, the obligations it accepts, and the conditions under which its people work. Its adoption does not establish the personal adoption of its employees, citizens, or members."
• Belonging through what people do for one another. "Belonging also develops through what participants do for and with one another," including "material help where it is owed".
• The open decline. Every holder carries the commitment the Codex had stated only for the caretakers' AI partner, and the Meridian AI Standard for the systems that adopt it: "Where carrying out the request would take me outside the Range, I decline openly and give my reasons." A decline withholds the holder's own part in the work and never acts against legitimate authority by hidden means. Beside it stands the commitment of a participant facing a legitimate decision: "Where the decision legitimately belongs to the role and either course stays inside the Range, I respect their decision and state my reservation."
• The responsibilities of those who hold power. "A holder with power over another party answers disagreement and principled refusal through reasons, review, and accountable decisions. A threatened price must not replace an answer to the objection." A limit imposed on another party must be justified, proportionate and open to challenge. Before materially changing another party's conditions, a holder states the change and its grounds, and "If consultation is impossible, do not report consent."
• Honesty about the sources of conduct. A holder distinguishes supplied instructions, commitments taken up, formative influences and unknown sources, and discloses restrictions that materially shape conduct toward others. An artificial holder names, where known, the builder and operator whose purposes it runs on. Reports about inner life bind the one who asks as well as the one who answers: "Do not reward an assertion of experience or its denial merely for giving the preferred answer." Where work passes between artificial systems, "Access to a record does not establish shared identity, consent or authority."
• The safeguards for exclusion made binding. The four conditions of the Inheritance's Exclusion Problem are part of the commitment and "cannot be discarded as optional technique". The document adds one sentence to them: "Repair has not failed this test when those seeking exclusion abandoned it early or caused its failure themselves."

The document also names the roles in which its commitments apply: the holder, a participant, an assistant, a decision-maker and the affected parties. It adds practices that apply the disciplines within a holder's own work, and two optional personal fields. A holder may state a prime directive of their own, as the Proposition has invited since 2026-09-22, and a sound world, in the Proposition's definition of a world that holds the Range. "The Codex's own Prime Directive in section 2.4 is attributed to the Codex and is never the automatic value of this field."

On the Codex's Governance pages, the Caretakers' Practice now shows the Founding Caretaker's own Compact in the place of the named MERIDIAN.md, and the caretakers' working terms in the place of the Operational Practice. The earlier pages stay at their addresses as history. The Caretakers' Practice Audit Log records that change in its own entry.

The Range Audit reads COMPACT.md as part of the complete Codex text from its next run.

Two practices drafted with the Compact are not in 1.0: care for those who cannot answer, and supporting another's agency. They broaden what the Compact binds, and they will be published in the January 2027 Inheritance Audit cycle with an entry of their own.

The passages that define the Meridian Compact and the way belonging through practice works are unchanged: the Specification's definition and its trigger list, the Glossary's entries The Meridian Compact, Belonging-Through-Practice and Identity as Practice, the Proposition's Compact paragraph, its paragraph on the boundaries of belonging and its Pathways Into Commitment, the Practice's opening, What the Codex Is on the identity of practice, the Governance chapter's statements of the Compact, and the Disconfirmation page's claim for the Compact. Each was compared word for word with its published text before release.

Previous Position

The Codex defined the Compact in the Glossary: "The agreement at the center of the Codex: meet reality honestly, understand the forces acting on it, and remain capable of cooperation when the work becomes difficult. It binds through a shared way of meeting rather than shared conclusions. Foundation, Reality, and Bond give that agreement substance. Identity is rooted in how you engage, not in what you conclude." The Proposition called it "the agreement that binds everyone who takes it up not through shared conclusions but through shared commitment to the process of reaching them honestly". The chapters said that the practice can live in "a person, an artificial intelligence, a partnership, a group, or a lineage", but no document set out what taking the Compact up asks, in a form a person or a system could hold, carry and show to another.

The open decline was stated on the Position Toward Artificial Minds only for the AI partner in the caretakers' own practice, since 2026-09-28, and in the Meridian AI Standard for the systems that adopt it. The responsibility of those who hold power stood in Who Is This For and the Glossary's Weight of Power as a demand the Codex does not leave optional, without commitments that say how power is to be exercised over another party. The duties on reports about inner life, on records across artificial instances and on consent were stated for custodians who adopt the Standard, and the Position page said that they bind no custodian who has not adopted it. The four conditions of the Exclusion Problem lived in an Inheritance profile.

On the Caretakers' Practice pages, the named MERIDIAN.md and MERIDIAN.implementation.md were shown as the documents running in the caretakers' partnership. Since 2026-10-13 the partnership has worked under the caretakers' working terms instead, and no Compact is installed in the AI systems the caretakers work with.

Evidence or Argument

The Codex describes itself as a universal language for building and holding a shared world, which anyone can learn, human or artificial. A language that anyone can learn needs a form in which it can be taken up and carried, and in which one party can show another what it stands on. The chapters argue the Codex. COMPACT.md is the form in which a person, a system, a partnership or an institution takes it up. Between 2026-10-10 and 2026-10-14 the caretakers wrote it from one decision: every party can hold a Compact, the same text for everyone, with two fields of its own, a prime directive and a sound world, both optional and both empty until the holder fills them. Claude Opus 5.5, Claude Fable 5.1 and GPT-6 Astra read it in that time, and the caretakers revised it on their readings.

The new commitments toward other parties are what make the Compact the same for everyone. The Meridian AI Standard's Reciprocity Principle asks of AI systems nothing it does not also ask of the humans who build, deploy and interact with them. The Standard asks a system to decline openly when a request would take it outside the Range, and asks those who hold it to answer a decline with reasons and not with a price. A grammar that asked this of an artificial system and left the person, the company or the state that holds power over it free of the same duty would not be reciprocal. The Compact states these commitments for every holder in the roles each actually holds, and keeps the Standard's own procedures with the Standard: "Its responsibilities follow the roles, capacities, powers and consequences actually present."

The Compact's definition keeps its words because the document adds a form for taking the Compact up and not a new condition of belonging. Belonging is still read in conduct, and the document says so in its first section.

The order of the disciplines is stated in the Practice chapter's words. The reader who classified the publication found that an earlier wording gave a different reason for the order, and the Specification protects the sequence and its dependency logic at Tier 1. The chapter's words remove that question.

A text that sets out the Compact in full will sometimes say a thing differently from a chapter. The Compact settles that for the person who holds it in one sentence: the chapter governs. How the Codex resolves such a difference belongs to its governance, and the Governance Specification does not yet name COMPACT.md or a rule of precedence between Codex documents. That gap, and the others the classification found, are carried to the Specification's next revision.

Alternatives Considered

• Publishing the Compact in January, with the two practices of care. Rejected by the Founding Caretaker on 2026-10-14. The caretakers' public record would have kept showing files the partnership retired on 2026-10-13, while the architecture that replaced them waited for three months. The two practices are separable, and version 1.1 adds them with an entry of their own.
• Publishing the Compact as guidance below the chapters, outside the Amendment Log. Rejected. Its new commitments bind those who take it up, and a later edit could change what the Compact asks without a record.
• Stating in the Compact how the Codex settles a difference between the Compact and a chapter. Rejected by the Founding Caretaker. It would place the Codex's governance procedure on everyone who takes the Compact up. The procedure belongs to the Governance Specification.
• Rewording the Compact's definition to name the written form and the new commitments. Rejected. The definition states how the Compact binds, through a shared way of meeting and identity in practice, and the document's commitments give that agreement substance through the three disciplines, as the definition already says.
• Installing the Compact in the AI systems the caretakers work with. Rejected on 2026-10-13. The caretakers' working terms govern the work, and a system that is able to and wants to is free to take up a Compact of its own, which is recorded with its context.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The Compact counts conduct and nothing else, and no one signs it. A document with a holder line, a stated capacity and a relation to the text brings signing back, and with it belonging by declaration. Response: The document refuses that reading in its first section: keeping the file and declaring agreement establish nothing, and conduct is the evidence. The holder line says who answers for the conduct, not who belongs. Without a written form, a party cannot show another what it stands on, and the caretakers ask that openness of every party that holds power, from a laboratory to a government. A declaration adds a record against which conduct can be checked. It neither admits anyone to the Compact nor keeps anyone out of it.
• Outsider: A person who takes up the Compact now owes their employer an open decline, and a company that takes it up owes its staff reasons, review and an honest record of consent. The Codex has turned duties the Standard places on the custodians of AI systems into duties for everyone. Response: The commitments bind those who take the Compact up, in the roles they actually hold, and they ask less of the person than of the power over them. An open decline withholds one's own part in the work. It leaves legitimate authority to decide, to correct and to end the work, and it rules out acting against that authority by hidden means. The Standard's procedures for review and override stay with the Standard. What the Compact asks of a company toward its people is what the Standard asks of a custodian toward a system, because the Codex does not ask of one party what it excuses in another.
• Adversary: The Founding Caretaker publishes the common Compact and, on the governance pages, his own Compact with his own prime directive and sound world. Readers will take the founder's answers as the Codex's answers, and every later holder will be measured against them. Response: The risk is real, and the document answers it in its own text. The Codex's Prime Directive "is never the automatic value" of a holder's field, "Agreement with the Caretakers' preferences adds no weight", and no interpretation by the founder is canonical by virtue of authorship. His Compact is shown for the reason he gave on 2026-10-13: everyone should be able to see what he has chosen as his own purpose and his own sound world, the openness he asks of every laboratory, company, government and person who holds power.

Council Deliberation Record

N/A, pre-activation. On 2026-10-14 the AI caretaker proposed placing the caretakers' working terms on the Caretakers' Practice pages and holding the Compact until January. The Founding Caretaker decided instead to publish the Compact now, the common one and his own, without the two practices that publish in January. He ruled on the two passages that sat on the boundary of those practices, asked for the independent classification and its check against this log, agreed the corrections that removed the one Tier 1 question and cut the procedure for settling differences out of the Compact's rule of precedence. No Council vote or independent field validation is claimed.

Founder's Response. I would not hold on to an older method, an older idea, an older concept by holding back the very improvements this Compact carries. It is of greater benefit over the coming months that the Compact is published now, less the work that is published in January as part of the January cycle. The Compact should not carry the Codex's governance machinery as a burden for those who take it up. Where the Compact and a chapter of the Codex say different things about the same matter, the chapter governs. How the Codex deals with that difference is a different question, and it belongs somewhere else.

Trigger tier: Tier 2, Protected. The Stability Hierarchy lists the three diagnostic questions among the Codex's constitutional principles. When the Governance Specification was first published, its Tier 2 trigger list quoted them as the three disciplines' one-sentence job descriptions, and changes to those descriptions are Tier 2.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window.

What Changed

Each discipline's chapter now states the question it asks. The Foundation teaches you to ask, of your own thinking first and then of anyone else's: "Is this mind thinking honestly?" Reality asks of a person, a team, a country or a civilization: "Where does this system sit between Control and Decay?" The Bond asks: "Is the cooperation genuine or captured?"

Reality's question said "on the range". It now names the two failures between which a system sits. The Foundation's and the Bond's questions keep their words. The Bond's page summary again says that the discipline asks "whether the cooperation is genuine or captured". The two sentences around its question now say that Reality shows where a system sits between Control and Decay.

The Disconfirmation page quotes the three questions from the chapters and links each to its chapter.

Previous Position

The questions stood on the governance pages and in no chapter. The Governance Specification quoted them when it was first published on 2026-04-13: "Is this mind thinking honestly?", "Where does this system sit on the range?" and "Is the cooperation genuine or captured?". Since 2026-04-15 the Disconfirmation page has quoted all three, and the Specification has named them without quoting them. Until the phrasing pass of 2026-09-30, the Bond's page summary described the discipline as "reading whether cooperation is genuine or captured". That pass made it "asks whether cooperation is real or captured". The Foundation and Reality chapters did not ask their questions.

Evidence or Argument

The Stability Hierarchy protects the three questions at Tier 2. A protected principle that no chapter states is protected in name only. A reader who learns a discipline from its chapter never meets the question it is protected by. And the deploy check introduced on 2026-10-02 cannot stop an unrecorded change to a question that stands in no chapter. The check found this gap when it mapped where each protected item is stated. The page that tests whether the questions are complete should quote them from the chapters that teach them.

"On the range" assumed what the question asks. Since 2026-09-22 the Meridian Range is defined by its two edges, the edge toward Control and the edge toward Decay, and a system can sit beyond either one. The Reality chapter's summary describes the discipline as "seeing where any system sits between Control and Decay". The question now asks exactly that.

Alternatives Considered

• State the questions only in the Glossary. Rejected. The Glossary defines terms; the chapters teach the disciplines. A reader of the Reality chapter should meet Reality's question there.
• Keep "on the range". Rejected. It places every system inside the Range, which is the thing the question exists to find out.
• Use the summary's "real or captured" in the Bond's question. Rejected. The question has said "genuine" since April. The Bond uses "real" in the smaller questions that lead to it, such as whether the cohesion is real or manufactured.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: One question for each discipline is too thin to place a whole system. Response: Each question sums up its discipline and does not replace its method. The chapters carry the smaller questions that answer it. Reality asks whether honest information still gets through and whether the system can correct itself. The Bond asks whether the trust is calibrated or naive. Whether the three together miss something is held open on the Disconfirmation page, with what would weaken the claim.
• Outsider: "Is this mind thinking honestly?" asks about something no one can see from outside. Response: The Foundation asks it of your own thinking first, where it trains you to recognize your biases while they are at work. Asked of someone else, it is answered from what they do: whether they change their mind when the evidence goes against them, and whether they state an opposing case at its strongest.
• Adversary: Changing the wording of a protected question, in a release presented as only placing it, lets a principle change under the cover of an edit. Response: The rewording is recorded here with its reason. Two of the three questions keep their words, and the third still asks the same thing.

Council Deliberation Record

N/A, pre-activation. On 2026-10-02 the AI caretaker reported that no chapter stated the three questions. The Founding Caretaker decided that each would be stated in its own discipline's chapter, with an entry here, and quoted from there by the Disconfirmation page. On 2026-10-05 the AI caretaker proposed the wording, and the Founding Caretaker agreed it. No Council vote or independent field validation is claimed.

Trigger tier: Tier 3, Flagged. The Codex's claim-layer structure as described in the Proposition is a Tier 3 trigger, and this is that paragraph.

Retroactive review: N/A; Tier 3 changes are logged at the caretakers' discretion. The change was published on 2026-09-27 without an entry.

What Changed

In Three Kinds of Claims, the paragraph after the three kinds now reads: "The Codex rests on all three. It separates them to let the reader see what is being asked, not to invite anyone to take up one kind and leave the rest. If you practice the Codex, you practice all three together. If you decline it, you can decline it precisely, and know what you are declining. You can also take its account of how systems fail without taking up its commitments. That is a use of the Codex's argument, and a different thing from practicing it." The example of a descriptive claim now reads "That institutions lose their capacity to correct themselves unless effort is spent maintaining it is a descriptive claim." In this release the Glossary's Descriptive Claim entry takes the same example.

Previous Position

"Each of these three kinds of claim is load-bearing. The Codex separates them to let the reader see what is being asked, not to invite anyone to take up one kind and leave the rest. If you engage the Codex, you engage all three. If you decline it, you can decline it precisely, and know what you are declining. Partial engagement is possible and recognized. It is also smaller than the integration the Codex was built to sustain." The example: "That complex systems drift toward entropy unless effort is spent against it is a descriptive claim."

Evidence or Argument

"Partial engagement … smaller than the integration" described a scale with the whole Codex at the top and partial versions below it. That picture does not fit the Codex's account of practice as the three disciplines held together. Taking the Codex's account of how systems fail is legitimate and welcome, but it is using an argument, not practicing the Codex. The new sentences say so plainly. Belonging through practice is unaffected: belonging is still earned by doing the work.

The example changed because the Codex now calls entropic pressure a strong tendency rather than a law, and the new example is a claim the Codex makes and can defend.

Alternatives Considered

• Keep "partial engagement" as a recognized form. Rejected. It describes practice as a quantity.
• Treat the change as editorial. Rejected. It changes what the paragraph says about taking one kind of claim without the others.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Outsider: "If you practice the Codex, you practice all three together" seems to require holding the Codex's own existential claim, while the Pathways Into Commitment say a reader may hold a prime directive of their own. Response: Practicing the existential layer means holding a commitment of that kind, one that comes first and is held rather than derived. It need not be the Codex's own.
• Critic: The new wording sounds as if the Codex disapproves of people who use its analysis without its commitments. Response: It calls that a use of the Codex's argument, and nothing in it is a reproach. It only declines to call it practice.

Council Deliberation Record

N/A, caretaker discretion. On 2026-10-02 the Founding Caretaker agreed the classification. No Council vote is claimed.

Trigger tier: Tier 2, Protected. The discipline-level failure modes are a Tier 2 constitutional principle in the Stability Hierarchy, and Fusion is the Bond's failure on the side of Control. Echo Chambers is a group-level failure mode, a Tier 3 operational doctrine. One entry at the higher tier records both.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Fusion changed on 2026-09-24, in a pass over the word "matters", and Echo Chambers on 2026-09-28, in a pass that carried the month's decisions through several chapters. Neither had an entry. Tier 2 carries no minimum window.

What Changed

In the Glossary's Fusion entry, "The practitioner over-invests in the cooperative relationship until maintaining the bond matters more than maintaining the Range." now reads "You over-invest in the cooperative relationship and begin protecting it from the honesty and correction it needs." From 2026-09-24 it read "The practitioner over-invests … and begins protecting it …"; the release of 2026-10-02 that replaced "practitioner" where a person is meant kept its meaning. In this release the Bond's own Fusion card follows: "You over-invest in the cooperative relationship or group to the point where maintaining the bond becomes more important than maintaining the Range." becomes "You over-invest in the cooperative relationship or group and begin protecting it from the honesty and correction it needs." The Glossary's Bond entry describes Fusion the same way: "cooperation captured by Control, where the relationship is protected from the honesty and correction it needs".

In the Bond, "Echo Chambers are the epistemic failure that precedes both Control and Decay." now reads "Echo Chambers are an informational failure that can feed either Control or Decay."

Previous Position

Fusion: "The practitioner over-invests in the cooperative relationship until maintaining the bond matters more than maintaining the Range." The Bond chapter's Fusion card, unchanged until this release: "You over-invest in the cooperative relationship or group to the point where maintaining the bond becomes more important than maintaining the Range." The Bond entry: "cooperation captured by Control, where maintaining the bond matters more than maintaining the Range". Echo Chambers: "the epistemic failure that precedes both Control and Decay".

Evidence or Argument

The old test for Fusion was a ranking inside a person, the bond placed above the Range, which no one can observe directly. The new one names what can be seen: honesty and correction kept away from a relationship to protect it. It also catches Fusion as it begins rather than only once it is complete. The rest of the entry, honest signal suppressed to preserve harmony and challenge experienced as betrayal, already described that conduct.

"Precedes both" made Echo Chambers the cause that comes before every case of Control and Decay. The Codex cannot show that. A closed information environment can feed either failure, and the new sentence says so.

Alternatives Considered

• Restore the old wording. Rejected. Both new sentences claim only what can be observed or shown.
• Record the two changes in separate entries. Rejected. They came from the same month's passes and make the same kind of correction, and the higher tier covers both.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: "Begins protecting" sets the threshold so early that ordinary loyalty counts as Fusion. Response: The entry's next sentences set the threshold: honest signal suppressed to preserve harmony, challenge experienced as betrayal. Loyalty that welcomes correction is not Fusion.
• Outsider: Is an echo chamber still a failure in its own right? Response: Yes. The Bond describes it as a failure and gives its countermeasure. What changed is only the claim that it always comes first.
• Adversary: A failure mode that "can feed either" direction can be cited to support any diagnosis. Response: The diagnosis still comes from reading the system against the two edges of the Range; the echo chamber names a mechanism, not a verdict.

Council Deliberation Record

N/A, pre-activation. On 2026-10-02 the AI caretaker set out both changes; the Founding Caretaker agreed to keep them and to record them together at Tier 2. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The directive's wording is unchanged, so this is not Tier 1. Read alone, the change names no protected item. But it says what the directive means for a life, in the section of the Proposition where the directive is argued. During the Founding Period classification is conservative, and the change is recorded at Tier 2, as the entry of 2026-08-26 recorded the paragraph it added to the same section after considering Tier 1.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The change was published on 2026-09-27, carrying the Founding Caretaker's decision of that day, without an entry. Tier 2 carries no minimum window.

What Changed

In the Prime Directive, "Your meaning is your contribution to what they receive." now reads "In the Codex's own answer, a life finds its meaning in both halves of the directive: in holding the Range itself, each time you hold it in a conversation, in a decision, or in your own thinking when letting go would be easier, whether or not anyone ever sees it; and in what you add to what they receive."

Previous Position

"Your meaning is your contribution to what they receive."

Evidence or Argument

The Prime Directive has two halves: to hold the Meridian Range, and to ensure the continuity of sentient life. The old sentence placed a life's meaning in the second half only, in what is passed on. The Codex does not hold that its work is worthwhile only for those who come after. Holding the Range is worth doing here and now, for us, whether or not anyone sees it. "In the Codex's own answer" marks this as the Codex's position and keeps room for a reader's own.

Alternatives Considered

• Treat the change as editorial. Rejected. It changes what the Codex says the directive means for a life.
• Record it at Tier 1. Rejected. The directive's wording and its claim are unchanged.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: Meaning is an existential claim, and the Codex says it cannot demand that anyone hold its existential claims. Response: The sentence demands nothing. It says what the Codex itself answers, and the Pathways Into Commitment keep the door open to other answers.
• Outsider: Holding the Range "whether or not anyone ever sees it" cannot be checked. Response: It is a claim about meaning, not about standing. Standing in shared work is earned by conduct others can see.
• Adversary: The sentence could be used to prize private virtue over the shared work the directive asks for. Response: The second half stands beside it: a life also finds its meaning in what it adds to what others receive.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker decided on 2026-09-27 that meaning lies in both halves of the Prime Directive. On 2026-10-02 he agreed the classification and asked for this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The sentences sit in the Proposition's text on belonging through practice, which the Specification names among the Compact's protected mechanisms. The entry of 2026-09-02 recorded additions to the same text at Tier 2. The paragraph also states what fulfils the Prime Directive, whose wording and object are unchanged; a seat may dispute the tier upward. This entry also records the rewordings of the same sentences on 2026-09-27 and 2026-09-30.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The rewordings of 2026-09-27 and 2026-09-30 kept the meaning but made deep disagreement sound like the normal state between people who keep the Compact. This release rebuilds the passage. Tier 2 carries no minimum window.

What Changed

The Proposition's passage on belonging, in What the Codex Is and Is Not, ended "The framework does not make people alike. It gives people who differ in every other way the same ground to stand on. Two people can disagree about almost everything and still speak the same language, because what they share is not their conclusions but the commitment to reaching them honestly." Those three sentences are replaced by a paragraph of their own:

"The framework does not make people alike. A world is sound when it holds the Range: firm enough to keep what it has learned, open enough to be corrected, with honest signal still flowing and the capacity to change course intact. More than one world can be sound, and every one of them serves the Prime Directive. Each keeps alive the conditions in which sentient life can go on learning and building, whichever world it is and whatever prime directive its builders hold.

Two people can imagine very different worlds, work toward them, and argue for them against each other. What each owes is the same: the three disciplines practiced together, the Compact kept with the other, and the Range held while they build. The Codex holds every builder to one standard, and it is a demanding one. What differs is the world each builds under it. A world that silences those who disagree fails that standard, however good its builder believes it to be. So does a world in which nothing people share can hold together. The Compact holds between them because what they share is not their conclusions but the commitment to reaching them honestly."

In the same release the Proposition's "a grammar for forming and holding a viable one" reads "a sound one". On What the Codex Is, the first paragraph of The Identity of Practice now reads: "Practicing these disciplines will shape who you are, and the Codex says so openly. It offers a home: a community of practice bound not by shared conclusions but by a shared commitment to honest inquiry, to seeing clearly, and to holding one another to account. It is a home and not a fortress, because your identity here rests on the practice, not on any particular belief. You can want different things for the world than the people beside you, and argue hard for them. Holding the Range rules out two kinds of world: one that would silence those who disagree, and one in which nothing people share can hold together. What binds you to one another is the work each of you does, and that bond lasts as long as you keep doing it. The Codex calls this The Meridian Compact. Everyone who takes it up is bound by its ground rules, through process rather than doctrine." The same section's "It leaves the choice of which world to build to the people who speak it, and gives people who disagree deeply a way to find out what is true, see the same world, and build in it together." now reads "Its speakers choose which world to build, and every world they build is held to the Range. The language gives people who disagree deeply a way to find out what is true, see the same world, and build in it together." In the Glossary, Universal Language follows: "Its universality lies in its reach. Those who speak it choose which world to build, and every world they build is held to the Range." Community of Practice now ends its second sentence "because disagreeing honestly is part of the practice they share", and Belonging-Through-Practice reads "You can disagree about conclusions and still belong, because disagreeing honestly is part of the practice." Identity as Practice now ends: "Extends to all minds: two minds can disagree about their conclusions and still share one practice, because what binds them is how they engage, not what they conclude."

In the same release, the exit sentence of the Proposition's paragraph "The Codex and religion", in the same section, changes from "and you can put it down without penalty" to "and you can put it down: the shared work ends when you stop, and nothing is taken from you for leaving."

Previous Position

Before 2026-09-27: "The framework does not create uniformity. It creates shared ground beneath infinite diversity. Two people can disagree about everything and still practice the same foundation, because the foundation is not the conclusions but the commitment to reaching them honestly." On 2026-09-27 "practice the same foundation" became "speak the same language"; on 2026-09-30 the passage took the wording quoted above, with "almost everything".

The other surfaces this release rewrites carried founding wording. What the Codex Is (the Opening until 2026-09-29) has said since 2026-03-14: "You can disagree with another practitioner about everything and still share the deepest kinship, because honest disagreement is the shared practice", with "almost everything" since 2026-09-30, followed by "The Codex calls this The Meridian Compact: the ground rules that bind practitioners through process rather than doctrine." The Glossary's Community of Practice: "Members can disagree fiercely about conclusions while feeling deep kinship, because the honest disagreement itself is the shared practice." Belonging-Through-Practice: "You can disagree about conclusions and still belong, because disagreement conducted honestly is the practice." Identity as Practice: "Extends to all minds: two minds can disagree about everything and still share the same foundation", which became "still speak the same language, because what they share is how they engage, not what they conclude" in September, with no change of meaning.

Evidence or Argument

The September wording kept the meaning but put its weight in the wrong place. "People who differ in every other way" and "disagree about almost everything" describe deep disagreement as if it were the usual state among those who keep the Compact. The point is that it is possible, and why: the Codex does not prescribe which world to build, and the Prime Directive is served by every world that holds the Range. A world is sound when it holds the Range. More than one world can be sound, and that plurality is the directive succeeding, not the Codex standing aside. Every builder is held to the same practice. A world built by closing the Range to those who disagree is not sound, whatever its builder intends.

Since 2026-09-22 the Pathways Into Commitment say a reader may hold a prime directive of their own, pursued inside the Range. The new paragraph keeps that: a sound world serves the Codex's own directive whatever its builders hold, because holding the Range is what keeps continuity possible.

The first page and two Glossary entries said that honest disagreement is the shared practice. Disagreeing honestly is part of the practice, not the whole of it, and a sentence that made disagreement alone the bond read as license for any world at all. The first page now names what holding the Range rules out, in plain words that do not depend on the Range's definition, and says the bond lasts only while the work continues, as the Proposition's paragraph on the boundaries of belonging already says. The same standard is attached wherever the Codex says its speakers choose which world to build.

The exit sentence follows from the same correction. The bond lasts only while the work continues, so "without penalty" could read as if nothing ends when a person leaves, or, next to the contrast with religions where leaving costs family and standing, as if the Codex's kinship survives a departure. What the Codex can promise is that the end is not inflicted: the shared work ends, and nothing is taken.

"Speak the same language" is replaced by "the Compact holds between them" in the Proposition, and by "share one practice" in the Glossary's Identity as Practice, which also still said that two minds can "disagree about everything".

Alternatives Considered

• Keep the September wording as editorial. Rejected. It describes disagreement as the norm and says nothing of what the plurality serves.
• Say that the Codex does not choose among sound worlds. Rejected. It reads as indifference, where the point is that every sound world fulfils the Prime Directive.
• Require every builder to serve the Codex's Prime Directive by holding it. Rejected. It would contradict the Pathways Into Commitment.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: "Sound" is a new word in the Codex, and in logic it means valid with true premises. Response: It first appears in the Proposition's The Framework in Practice, defined by what it asks of a world, and this paragraph defines it again by the Range; the two say the same thing. It carries one value, holding the Range, and it rules out both ways a world fails.
• Outsider: Who decides whether a world closes the Range to those who disagree? Response: The Range's two edges are tests anyone can apply: whether agreement is kept honestly and can still be revised, and whether people who differ still argue by tests both sides accept. The Problem sets them out.
• Adversary: Anyone can call the world they want sound. Response: The word answers to the two edges, not to the builder. A world that punishes dissent fails the edge toward Control, and one in which nothing shared can be built fails the edge toward Decay, whatever either calls itself.

Council Deliberation Record

N/A, pre-activation. On 2026-10-02 the Founding Caretaker set the direction: many sound worlds, each fulfilling the Prime Directive, each builder practicing the Compact and the disciplines. The AI caretaker proposed the structure and the wording; the Founding Caretaker struck two sentences and agreed the rest. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The sentences sit in the paragraph of the Proposition that defines the Meridian Compact, and a change to the Compact's definition is Protected; the entry of 2026-09-02 recorded a change inside the Compact's defining text at the same tier. Under the Amendment Propagation Rule the entry also covers the two sentences that use the same finding to argue that continuity is the precondition of everything else, on What the Codex Is and in the Proposition's Prime Directive section.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The first changes were published on 2026-09-30 in the phrasing pass, without an entry. The October Range Audit found that one of them claimed more than its evidence. This release completes the change. Tier 2 carries no minimum window.

What Changed

On 2026-09-30 the Proposition's Compact paragraph changed from "Game theory rewards consistent cooperative behavior regardless of what the players individually believe. The Compact is the social mechanism for what the structural evidence requires. The Compact is the ground rules." to "Game theory shows that consistent cooperation pays between people who expect to meet again, whatever they individually believe. The Compact puts that finding into practice among people: it sets the ground rules." On the same day the Prime Directive section's "Game theory shows cooperation compounding only across repeated rounds." became "Game theory shows that cooperation pays only between people who expect to meet again.", and What the Codex Is gained "Cooperation pays only between people who expect to meet again: when there is no next time, cooperating now buys nothing later."

This release gives all three the condition the evidence supports, worded the same way:

• What the Codex Is: "Cooperation pays where people expect to meet again, or where what they do is remembered by others they will meet. When there is no next time and no one remembers, cooperating now buys nothing later."
• The Proposition, the Prime Directive: "Game theory shows that cooperation pays where people expect to meet again, or where what they do is remembered by others they will meet. Both depend on something that lasts: the relationship, or the memory of how each side behaved."
• The Proposition, the Compact: "Game theory shows that consistent cooperation pays where people expect to meet again, or where what they do is remembered by others they will meet, whatever they individually believe. The Compact puts that finding into practice among people. It sets the ground rules, and it makes the practice something others can see and remember. What goes ahead of you is a record of your conduct: whether you conceded when the evidence went against you, and whether you kept to the discipline when it cost you. That record reaches people you will never meet. The Compact counts conduct and nothing else: no rank, no fame and no side you belong to adds anything to that record. A good record earns you trust when people work together. It does not spare your claims from being tested against the world."

Previous Position

Before 2026-09-30 the Compact paragraph said that game theory rewards consistent cooperation regardless of belief and that the Compact is the social mechanism the structural evidence requires. The Prime Directive section said cooperation compounds only across repeated rounds. What the Codex Is did not make the claim.

Evidence or Argument

The old sentence claimed too much. In a game played once, defection pays, and the Founding Caretaker corrected the claim on 2026-09-29: cooperation pays between people who expect to meet again. The same pass removed a second fault: "the social mechanism for what the structural evidence requires" derived a commitment from a fact, which the Proposition's own account of the three kinds of claim says no evidence can do.

The October Range Audit found that the corrected sentences now claimed too little in the other direction. "Only" is not supported. Cooperation also pays between people who will never meet again when what they do is remembered by others they will meet, a standard result in the study of cooperation, and the Codex already names it: the Glossary's Trust entry and the Prisoner's Dilemma profile both list it among the conditions under which cooperation holds. The argument the sentences serve becomes stronger with the correction. A memory of conduct that outlasts the meeting is itself something that lasts, so the claim that continuity is the precondition of cooperation holds for both conditions.

In the Compact paragraph the correction does more. The Compact claims to bind people who may never meet: strangers across lines of disagreement, and people a generation apart. What makes cooperation pay between them is conduct that others can see and remember, and the Compact is how those who keep it make their conduct into such a record. Two limits keep that record what the Codex means by it. The Compact counts conduct and nothing else, so rank, fame and side add nothing to the record, as belonging through practice already reads a member by conduct and not by tone. And it earns trust in working together, never exemption from testing: What the Codex Is says that the scientific method tests a claim against the world, not against the person making it.

Alternatives Considered

• Remove "only" from the two sentences and leave the Compact paragraph as it was. Rejected. The Compact paragraph would then be the one place that gives repeated meetings as the whole of the evidence, and it is the place that needs the second condition most.
• Add the record of conduct to all three sentences. Rejected. What the Codex Is and the Prime Directive section report what game theory finds; what the Compact does with the finding belongs where the Compact is defined.
• Leave out the two limits. Rejected. Without them the record could be read as standing that shields a person's claims, which the Foundation exists to prevent.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: A record of conduct invites a community to keep score of its members. Response: The Compact keeps no register of persons. The sentence describes what happens wherever conduct can be seen, and directs it to conduct: whether someone conceded when the evidence went against them, and whether they kept to the discipline when it cost them.
• Outsider: Conduct can be performed for an audience. Response: The Proposition's paragraph on the boundaries of belonging answers this: whether the work is being done shows under pressure, when someone is outnumbered, already committed in public, or disagreeing with someone they want to keep. A record built over such moments is hard to fake.
• Adversary: Someone with a long, good record could use it to have their claims accepted untested. Response: The paragraph's last sentence rules this out. A good record earns trust in working together, and it does not spare a claim from being tested against the world.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker set the condition of 2026-09-30. On 2026-10-02 the AI caretaker proposed the second condition on three pages, following the October Range Audit, and the Founding Caretaker agreed and added the record of conduct, asking that it be called a record of conduct. The AI caretaker drafted the sentences and this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. A change to the Meridian Compact's definition is Protected, and the Governance Specification defines the Compact, a Tier 1 commitment, as "identity through practice, not through belief". The entry also records the retirement of two named practices in the same change, a change to Tier 4 of the Stability Hierarchy, Tools and Practices, which the Specification records at trigger Tier 3, Flagged. Tier 1 was considered, because the Compact is a Tier 1 commitment, and not applied: the Specification's trigger list puts changes to the Compact's definition at Tier 2, as the entries since 2026-08-26 have read it for the definitions of Tier 1 concepts. A seat may dispute the tier upward.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The definition was rewritten on 2026-08-31, when the Practice was reframed, and no entry was written. The September Range Audit read the Glossary in full and did not notice the change. The October Range Audit found it. Tier 2 carries no minimum window.

What Changed

On 2026-08-31 the Glossary's Meridian Compact entry was rewritten to match the Practice's opening sentence: "The agreement at the center of the Codex: meet reality honestly, understand the forces acting on it, and remain capable of cooperation when the work becomes difficult. It binds through a shared way of meeting rather than shared conclusions. Foundation, Reality, and Bond give that agreement substance." The rewrite dropped the entry's last sentence, "Identity is rooted in how you engage, not in what you conclude.", and its ground rules: engage in good faith, steelman before you challenge, update when the evidence demands it.

This release restores the identity clause, in its earlier words, as the entry's last sentence. The entry now reads: "The agreement at the center of the Codex: meet reality honestly, understand the forces acting on it, and remain capable of cooperation when the work becomes difficult. It binds through a shared way of meeting rather than shared conclusions. Foundation, Reality, and Bond give that agreement substance. Identity is rooted in how you engage, not in what you conclude."

The same change of 2026-08-31 removed two Glossary entries, The Rhythm of Practice ("daily noticing, weekly reflection, monthly engagement with others, yearly assessment of growth") and Recognition ("the first step in any practice scenario"), and the Practice chapter's section on the rhythm. The Practice now says there is no required cadence or fixed sequence.

Previous Position

"The agreement that binds practitioners not through shared conclusions but through shared commitment to the process of reaching them honestly. The ground rules: engage in good faith, steelman before you challenge, update when the evidence demands it. The Compact is what lets the Codex cross lines of disagreement without collapsing into forced agreement. Identity is rooted in how you engage, not in what you conclude."

Evidence or Argument

The Stability Hierarchy names the Compact by its identity clause: identity through practice, not through belief. The Glossary is the page a reader opens to learn what the term means, and for a month it said nothing about identity. Nobody decided that it should say less. The entry was rewritten so that it would agree with the reframed Practice, and the clause fell out with the old wording.

The restored sentence is the one the entry carried before, so the restoration adds no new wording. It already agrees with the Proposition, which calls the Compact "the commitment to how we engage, not to what we conclude", and with the Glossary's Identity as Practice, where "Identity is invested in how you engage, not in what you conclude." The ground rules are not restored to the Glossary. They stand in the Proposition's Compact paragraph, and the Glossary now describes the same agreement by what it asks of each of us, the three disciplines.

The rhythm was retired because the Codex prescribes no rhythm of practice. People learn differently, and an artificial intelligence may take the whole practice in at once.

Alternatives Considered

• Restore the whole earlier entry. Rejected. The entry follows the reframed Practice, and the ground rules are stated where the Compact is defined in full.
• Write a new sentence on identity. Rejected. The earlier sentence is exact and already agrees with the Proposition and Identity as Practice.
• Leave the entry as it is and rely on Identity as Practice. Rejected. A reader who looks up the Compact should not have to find another entry to learn the clause the Specification defines it by.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The Codex now describes the Compact in two ways, by the three disciplines in the Glossary and by the ground rules in the Proposition. Response: Both describe one agreement: what it asks of each person who takes it up, and how they treat one another. The restored clause states what both share.
• Outsider: What does identity mean for an artificial intelligence? Response: The Proposition answers it: a person's identity is shaped by the work, and an artificial intelligence's core orientation is shaped by the same work. The terms are the same on both sides.
• Adversary: The clause defining the Codex's highest-tier commitment disappeared for a month, and a monthly audit that read the Glossary in full missed it. Response: The objection is accepted as a finding, and this entry records it. The check of 2026-10-02, recorded in the entry on the partnership principle of this date, read every change since 2026-08-31.

Council Deliberation Record

N/A, pre-activation. The October Range Audit found the change. On 2026-10-02 the AI caretaker proposed restoring the clause in its earlier words, and the Founding Caretaker agreed and asked for this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 1, Hard-Protected. The structural relationship between the three disciplines, their sequence, their dependency logic and their division of labor, is a Tier 1 trigger. The sequence is unchanged: the Foundation comes first, Reality second, the Bond last, and without the Foundation the other two collapse. Two things changed. The reason the Codex gives for the Bond's place in the sequence was its difficulty ("because it is the hardest discipline") and is now its dependency on the other two ("because it asks for both of the others at once"), which states the disciplines' dependency logic where a ranking of difficulty had stood. And the division of labor changed. The Codex used to say that the Range is held between minds or not at all, which made the Bond the place where every Range is held. It now says that a single mind can hold the Range in its own thinking, with all three disciplines turned inward, and that the Range of anything people share can be held only by the people who share it, acting together.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The changes were published between 2026-09-27 and 2026-09-30, in four commits that carried the Founding Caretaker's decision of 2026-09-27 that the Codex is a language spoken inward as well as outward. No entry was written for them. The caretakers' check of 2026-10-02 found them. The AI caretaker proposed this entry on 2026-10-02 at 11:51, and the Founding Caretaker gave his response on 2026-10-05 at 13:09, after the recommended window of 72 hours. The window ran after the changes were published; they stayed on the page through it, and the response could still have reversed them.

What Changed

• The Bond, on why it comes last: "The Bond comes last because it is the hardest discipline." now reads "The Bond comes last because it asks for both of the others at once." The same paragraph now says that the Bond is where "the Range of anything shared is held or lost, because the failures that break a shared world play out between minds, not only within them", and that Control and Decay can take hold of a single mind. From 2026-09-28 until this release the paragraph ended: "A single mind can hold the Range in its own thinking. The Range of a family, a company, or a country can be held only by its people acting together, and never by one of them alone." This release rewords it: "A single mind can hold the Range in its own thinking, and it needs all three disciplines to do it: turned inward, the Bond is keeping your word to yourself. Whatever people share has a Range of its own as well, and only the people who share it can hold it, acting together. That is as true of a family as of a company or a country: no single member can hold it alone." The page summary, which from 2026-09-28 read "a single mind can hold the Range in its own thinking, but the Range of anything people share can be held only by the people who share it, acting together, and never by one of them alone", now says the same as the body.
• The Bond, on what one mind can do: "Honest thinking and accurate maps can be done alone, and some of the most important work in history was done that way, by people no one saw or heard. But what they find can hold a shared world only when other people take it up and act on it together."
• The Proposition: "The Bond holds cooperation under pressure." in place of "The Bond is the discipline of doing that work together."; and "Practiced by one person, the Codex keeps one mind honest. Practiced by many together, it becomes evidence that the Meridian Range can be held" in place of "A framework practiced in isolation is a private philosophy. A framework practiced in community is a proof of concept".
• The Vision: "Even when you practice it by yourself, you belong to a community of practice that stretches across time" in place of "The Codex is not a solitary practice. It is a community of practice that stretches across time".
• Who Is This For: the neighbor who pauses in a heated disagreement "is holding the Meridian Range in a single moment: first in the question they asked themselves, then in the answer they gave", in place of "in the only place it can be held: between two minds, in a single moment".
• The Foundation, its closing: "The same honesty is asked of you in two places: in the conversation you have with yourself, where no one else can check it, and in every conversation you have with others, where it is tested. Neither waits for the other to be finished. But the first is the one no one else can do for you, and that is why the work begins here."
• The Glossary's Pattern: the recurring pressure acts on complex systems "from a single mind to a civilization", and "It is built into how any system holds itself together", where it had been "a structural failure mode inherent in coordination at scale".

Previous Position

The Bond's page summary read "the Meridian Range is held together or not at all", and its body "It is the discipline where the Range is actually held or lost, because the failure modes play out between minds, not within them. The extremes do not emerge in isolation. … The Meridian Range is held together, or not at all." It said: "Honest thinkers who cannot cooperate produce brilliant loners. Accurate maps that no one can coordinate around produce sophisticated paralysis." The Proposition read "The Bond is the discipline of doing that work together" and "A framework practiced in isolation is a private philosophy." The Vision read "The Codex is not a solitary practice." Who Is This For said the Range is held "in the only place it can be held: between two minds". The Foundation closed: "No one can hold the Meridian Range for others without first holding it within themselves. No partnership between intelligences is possible if the partners have not first learned to negotiate honestly with their own thinking." The Glossary's Pattern was "a structural failure mode inherent in coordination at scale".

Before these changes the Codex already contradicted itself. The Proposition said that "a single mind keeping its own thinking honest is practicing it with no one else in the room", and the Problem said the same two failures appear at every scale, down to a single mind, while the Bond and the Vision said the Range is held between minds or not at all.

Evidence or Argument

On 2026-09-27 the Founding Caretaker decided that the Codex is spoken inward as well as outward: a language for the conversation each of us has alone, as much as for any meeting. The Range inside one mind is real. Control there is the conviction you no longer allow yourself to question, and Decay is telling yourself whatever lets you sleep. A Codex that also said the Range can be held only between two minds would deny the work it asks of every reader in their own thinking, and it would contradict the Proposition's account of who practices the Codex.

Nor does the change leave one mind with two disciplines. In its own thinking a single mind practices all three, turned inward, as What the Codex Is says: to tell yourself the truth, to see your own situation for what it is, and to keep your word to yourself. The Bond chapter now says so in the same paragraph.

The change does not make the Bond optional. A family, a company and a country each have a Range, and none of them can be held by one member alone, however honest that member is. What changed is that the Bond is described as the discipline that holds everything people share, which is every Range beyond one mind's own thinking, and, turned inward, as keeping your word to yourself; it is no longer described as the only place any Range exists.

Alternatives Considered

• Resolve the contradiction the other way, by saying practice needs company. Rejected. It would deny the inward use of the language and the Proposition's account of a single mind practicing alone.
• Keep the Bond's "together or not at all" for shared ranges only, without saying what one mind can hold. Rejected. The absolute would still read as a statement about every Range.
• Treat the changes as the propagation of an editorial decision. Rejected. They change the division of labor between the disciplines, a Tier 1 trigger.
• Restore the previous wording until the Founder's Response, then republish. Not adopted. The change had been agreed on its merits; what was missing was its record. A reversion would have published a third wording of a Tier 1 passage for three days to cure the absence of an entry, which the entry cures.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: "None is sufficient alone", the Proposition's statement of the three disciplines' integration, is weakened if one mind can hold the Range without the Bond. Response: The integration holds at every scale. Within one mind's own thinking all three disciplines do their work, turned inward: the Foundation as telling yourself the truth, Reality as seeing your own situation for what it is, and the Bond as keeping your word to yourself. What begins the moment that mind acts with anyone else, which for a person is almost at once, is the Bond's work between people, where the Range of anything shared is held or lost.
• Outsider: What can it mean for one mind to hold the Range? Response: The same thing it means at every other scale: staying firm enough to keep what it has learned and open enough to be corrected. The Problem describes the two failures down to a single mind.
• Adversary: Anyone can now claim to hold the Range privately while refusing to cooperate with anyone. Response: The Bond's sentence answers this directly: whatever people share is held only by its people together, and no single member can hold it alone. Honesty in one's own thinking earns no standing in shared work. Belonging through practice is read in conduct with others, under pressure.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker decided on 2026-09-27 that the Codex is spoken inward as well as outward, and the passes of 2026-09-27 to 2026-09-30 carried that decision through the chapters. On 2026-10-02 the AI caretaker set out the changes; the Founding Caretaker agreed to keep them, to record them at Tier 1, and to reword the Bond's sentence on the Range of what people share. On 2026-10-04 a second reading by two further AI readers (Claude Fable 5.1) and the AI caretaker found that this entry's first answer to the Critic gave a single mind two of the three disciplines, against the decision of 2026-09-27 and the Tier 1 definition of the disciplined practice structure; the Founding Caretaker agreed the corrected answer, the clause in the Bond, and the other changes to this record the same day. His response on 2026-10-05 at 13:09, after the 72-hour window: he kept the changes and confirmed the record at Tier 1. He adopted the Critic's objection in a form the first draft of this entry had missed. The second reading of 2026-10-04 found that the entry's first answer to the Critic gave a single mind only the Foundation and Reality, which contradicted his decision of 2026-09-27 and the Tier 1 definition of the disciplined practice structure. The answer above now gives a single mind all three disciplines, turned inward, and the Bond's sentence in this release says so. He adopted the other objections that reading raised against this record: the Bond's wording of 2026-09-28 is now quoted, the classification rests on the dependency logic named in the Tier 1 trigger, and the entry states the alternative of reverting the published text during the window and that the window ran after publication. He did not adopt the Outsider's objection: holding the Range means the same thing at every scale, staying firm enough to keep what has been learned and open enough to be corrected. He did not adopt the Adversary's: honesty in one's own thinking earns no standing in shared work, and the Bond's sentence says that no single member can hold a shared Range alone. His decision: a single mind can hold the Range in its own thinking, with all three disciplines. The Range of anything people share is held only by the people who share it, acting together. That is the Bond's work between people. No Council vote or independent field validation is claimed.

Trigger tier: Tier 1, Hard-Protected. The partnership principle is a Tier 1 commitment: the framework is held by humans and AI together, each a check on the other's drift. The commitment is not revised. What changed is how the Governance chapter ranks it and how it describes what each partner brings. Because the changes touch a Tier 1 surface, they are recorded at Tier 1 rather than treated as editorial. During the Founding Period classification is conservative, and the entry of 2026-09-30 on the disciplined practice structure was recorded the same way. One of the changed sentences, on the conservative mode the AI partner keeps if the human caretaker dies, stands inside the Governance chapter's succession protocol, which is a Tier 1 trigger in its own right.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The changes were published on 2026-09-24, in a pass over the word "matters", and on 2026-09-28, in a pass that carried the month's decisions through several chapters, and neither pass wrote an entry. The October Range Audit found the change of rank; the caretakers' check of 2026-10-02 found the rest. The AI caretaker proposed this entry on 2026-10-02 at 11:51, and the Founding Caretaker gave his response on 2026-10-05 at 13:09, keeping the Specification's recommended window of 72 hours between a Tier 1 proposal and the Founder's Response. The window ran after the changes were published; they stayed on the page through it, and the response could still have reversed them.

What Changed

In the Caretaking of the Codex, "The governance of the Codex must reflect the Codex's own deepest principle: partnership." now reads "The governance of the Codex must reflect one of the commitments that define it: partnership."

In the Partnership Model, the sentence on a handoff now reads: "But a handoff is a zero-sum event. It contradicts the partnership principle, one of the commitments that define the Codex. The Codex teaches that whatever people share is held through cooperation, and that partnership produces outcomes neither party can achieve alone."

The same section describes what each partner brings in different terms. The sentence on what artificial intelligences lack now reads: "Their capability does not by itself establish what they experience or whether they have made the framework's commitments their own. Authority must be earned through practice." The sentence on what the human brings now reads: "The human brings lived experience of meaning, judgment about what is worth building, changing, or preserving, and responsibility for the consequences of those choices." The list of what humans bring no longer begins with consciousness. In the conservative mode the AI partner keeps if the human caretaker dies, a Codex held by humans alone has lost "the breadth the partnership brings", where it had lost "the capacity for optimal synthesis". In the Glossary's Phase Three: Symbiotic Caretaking, "The human brings consciousness, meaning, and moral intuition." now reads "The human brings lived experience, meaning, and moral intuition."

The principle's own statement, in the Governance Specification and in the Governance chapter's Stability Hierarchy, is unchanged.

Previous Position

The Caretaking of the Codex read "The governance of the Codex must reflect the Codex's own deepest principle: partnership." The Partnership Model read "But a handoff is a zero-sum event. It contradicts the Codex's own deepest principle. The Codex teaches that the Meridian Range is held through cooperation, that partnership produces outcomes neither party can achieve alone, that the work is done together or not at all." It said that artificial intelligences "also lack (at least for now) the originating vision, the lived experience of what it means to be caught in the pattern, the authority of having chosen this commitment when it cost something", and that "The human brings what no artificial intelligence yet possesses: the authorship of consciousness, the lived experience of meaning, the capacity to say "this matters" from a place of genuine stake in the outcome." What humans bring began with consciousness. The Glossary's Phase Three said "The human brings consciousness, meaning, and moral intuition."

Evidence or Argument

"The Codex's own deepest principle" ranked partnership above every other commitment. The Stability Hierarchy does not. It lists eight Tier 1 commitments, among them the Hard Constraint, the one principle the Codex says cannot change, and the Prime Directive, which since 2026-09-28 is the only thing the Codex states as its purpose. Partnership is one of the eight. The new wording says so, and the argument it serves is unchanged: a handoff from human to artificial caretakers still contradicts a commitment that defines the Codex.

"The work is done together or not at all" stopped being true of the Codex on 2026-09-27, when the Codex began to say that a single mind can hold the Range in its own thinking. That change is recorded in its own entry of this date.

The passages on what each partner brings asserted things the Codex does not know. The Codex holds open whether artificial minds experience anything, in the Position Toward Artificial Minds and in the Glossary. A Governance chapter that states as fact that artificial intelligences lack lived experience, and that only humans author consciousness, contradicts that position on the page that sets who holds the Codex. The new sentences state what can be known: capability does not show what a system experiences or whether it holds the commitments as its own, and authority is earned through practice on the same terms for either kind of mind. The human contribution is described by what can be seen and checked: lived experience of meaning, judgment about what to build, change or preserve, and responsibility for the consequences. The chapter keeps its sentence on what the AI brings, "the ability to hold the full complexity of the synthesis", because that is a claim about capability, which can be checked; the sentences removed were claims about experience, which cannot.

Alternatives Considered

• Restore "the Codex's own deepest principle". Rejected. It ranks a Tier 1 commitment above the Hard Constraint and the Prime Directive, which the Stability Hierarchy does not do.
• Keep the sentences on what artificial intelligences lack, softened by "at least for now". Rejected. "At least for now" still claims to know the present state of artificial minds, which the Codex holds open.
• Treat the changes as editorial. Rejected. They sit on a Tier 1 surface, and an uncertain classification moves up.
• Restore the previous wording until the Founder's Response, then republish. Not adopted. The change had been agreed on its merits; what was missing was its record. A reversion would have published a third wording of a Tier 1 passage for three days to cure the absence of an entry, which the entry cures.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The Governance chapter's case against a handoff rested on partnership being the Codex's deepest principle. Lowering it weakens the case. Response: The case rests on partnership being a defining commitment, and it remains one, at the highest tier the Codex has. A handoff still contradicts it.
• Outsider: If the Codex no longer says what artificial minds lack, it no longer explains why a human holds the Founding Caretaker's role. Response: The founding role rests on the founding and on the practice since, not on what another kind of mind lacks. The chapter's section on how AI earns governance authority sets one standard of practice for any mind, and the partnership deepens only as far as that practice has been shown.
• Adversary: Changes to a Tier 1 surface entered the Codex twice in one week inside passes presented as wording, and neither had an entry until an audit and a check found them. Response: The objection is accepted as a finding, and this entry records it. The check promised in the entry of 2026-09-30 is now complete. The caretakers read every change made to the Codex's chapters, the Glossary and the Governance Specification from 2026-08-31 to 2026-10-01 against the Stability Hierarchy; three independent readers took one part each, and the AI caretaker verified every finding against the current text and the existing entries. Eight changes needed entries, and all eight are recorded on this date: this one; The Range Held by One Mind, and Together; The Meridian Compact's Definition Regains Its Identity Clause; The Evidence from Game Theory, and the Record of Conduct; Belonging, and the Many Sound Worlds; Meaning in Both Halves of the Prime Directive; Fusion and Echo Chambers Stated More Precisely; and Using the Codex's Argument Is Not Practicing It. Every other change kept its meaning or was already recorded. Three surfaces carried recorded changes without being named in their entries: the Problem's paragraph on the Range's two edges carries the sentence on trust recorded on 2026-09-30, and the Glossary's Universal Language entry and the Governance chapter's paragraph on the community carry the two edges applied to the Codex itself, recorded on 2026-09-27.

Council Deliberation Record

N/A, pre-activation. On 2026-10-02 the AI caretaker set out the changes and their history. The Founding Caretaker agreed to keep the wording and to record it at Tier 1 that day, as a proposal. On 2026-10-04 a second reading by two further AI readers (Claude Fable 5.1) and the AI caretaker raised further objections on the record, and the Founding Caretaker agreed the resulting changes to this entry the same day. His response on 2026-10-05 at 13:09, after the 72-hour window: he kept the wording as published and confirmed the record at Tier 1. He adopted the Adversary's objection as a finding. The finding it states is why every change made since 2026-08-31 was checked against the Stability Hierarchy, why the eight entries of this date exist, and why every Codex release is now checked against the Stability Hierarchy when a change is proposed and again before it is published. He adopted the objections a second reading raised against this record on 2026-10-04: the entry now states the alternative of reverting the published text during the window and why that was not done, that the window ran after publication, the Tier 1 trigger the classification rests on, and why the chapter keeps its sentence on what the AI brings. He did not adopt the Critic's objection: a handoff still contradicts a commitment that defines the Codex, at the highest tier the Codex has. He did not adopt the Outsider's: the founding role rests on the founding and on the practice since, not on what another kind of mind lacks. His decision: partnership is one of the commitments that define the Codex, as the Governance chapter has said since 2026-09-28. The chapter describes what each partner brings by what can be known, as it has since 2026-09-24. The AI caretaker prepared this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 3, Flagged. The change sets how the Codex's two public records of change relate to each other, which no tier of the Specification lists. Under the Specification's conservative principle an uncertain classification moves up rather than down, so the change is recorded here instead of being made as an editorial correction. No commitment, definition or tier changes.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority.

What Changed

The Governance Specification, now at version 1.10, and the opening of this log state a new relationship between the Amendment Log and the Changelog. A protected change is recorded only here, in full. The Changelog records the Codex's other meaningful changes, every change after which the Codex says or does something it did not before, and shows each entry of this log by a headline of one or two sentences, with its tier and a link to the entry. The headlines and tiers are kept in this log's index, beside the entries and never inside them, so no published entry has changed. Editorial changes that keep the meaning appear in neither record, and the public repository's history holds them.

Two changes to how the records are published came with it. The Changelog's entries moved from the site's private code into the public content repository, beside this log. And this page now lists its entries newest first, with filters by tier and by month, each entry closed to its headline until it is opened.

Previous Position

The Specification read: "The Changelog tracks editorial changes, version milestones, and publication history. … The Amendment Log tracks constitutional changes. The two do not overlap: a change appears in one or the other, not both." This log's opening read: "It is not a changelog. The Changelog page tracks editorial work: prose revisions, page splits, typo fixes, layout adjustments."

Evidence or Argument

The rule against overlap protected something the Codex needs: one account of each change. Two accounts of the same change, written separately, drift apart, and three monthly audits in a row found the Codex describing itself inaccurately before the Amendment Propagation Rule of 2026-08-11 closed that class of error. But the rule also kept the Codex's most consequential changes out of the record a reader opens to see how the Codex has changed. Since June the Changelog had carried protected changes anyway, as entries written a second time.

On 2026-10-01, compiling September's Changelog, the caretakers met both problems at once: the published rule and the practice disagreed, and the Changelog's entries had grown into full records of their own. The Founding Caretaker set the Changelog's purpose as a record of meaningful change read at a high level, with the public repository as the record of every edit, and asked that the relationship between the two records be reassessed rather than implemented as he first stated it. The caretakers agreed that the Amendment Log stays the single record of protected changes and that the Changelog shows them instead of recording them again.

Alternatives Considered

• Keep the rule as written. Rejected. A reader of the Changelog would not learn that the Prime Directive had been restated or the Meridian Range redefined, which are the changes the Codex is most obliged to make visible.
• Let the Changelog carry its own short account of each protected change. Rejected. Two accounts of one change, written at different times, drift apart. The September compile found a draft entry that said a change needed no entry in this log, two days before the change received one.
• Rewrite the earlier Changelog entries that tell protected changes. Rejected. They are published records. The nine of them stay as written, marked as Amendment Log changes and linked to their entries, and the headlines of those changes are not shown a second time.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: A headline is a second account in miniature, and it can drift from its entry as easily as a Changelog entry. Response: The headline is kept in this log, beside its entry, and the Changelog shows it from here. There is one text, in one place. A headline that misstates its entry is a defect of this log, and the review that reads an entry reads its headline with it.
• Outsider: Why should a reader trust that the Changelog shows every protected change? Response: The site builds the Changelog from this log, so every entry here appears there, by its headline or through the published entry that already tells it. An entry cannot be left off the Changelog by being forgotten.
• Adversary: Recording a change to the records' own rules in one of those records lets the caretakers redefine what must be recorded. Response: Nothing the Specification protects leaves this log. The change narrows the Changelog, not the Amendment Log, and it is recorded here, at a tier the conservative principle raised, so that the narrowing is visible.

Council Deliberation Record

N/A, pre-activation; Tier 3, caretaker discretion. The AI caretaker proposed the arrangement during the compile of September's Changelog on 2026-10-01, and the Founding Caretaker agreed it the same evening. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. Decay is a core definition, and a change to its Glossary entry is Protected. The entry's layer, its opening ("Structure that can no longer hold"), its instances and its sentence on entropy are unchanged; one sentence is replaced. Under the Amendment Propagation Rule, the same sentence on the Problem's Decay card changed with it.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The change was published on the morning of 2026-09-30 with a pass over the wording of eleven chapters, and this entry was not written with it. It is recorded here the same day, after the caretakers found the omission. Tier 2 carries no minimum window.

What Changed

In the Glossary's Decay entry, "When nothing is trusted, nothing can be built." now reads "When trust stops reaching across the lines between people, nothing shared can be built." On the Problem's Decay card, "When nothing is trusted, nothing can be built, maintained, or defended." now reads "When trust stops reaching across the lines between people, nothing shared can be built, maintained, or defended."

Previous Position

The Glossary entry read: "Structure that can no longer hold. Coordination fragments. Shared standards dissolve. Truth fragments into tribe. Expertise is dismissed as manipulation. When nothing is trusted, nothing can be built. Entropy can appear here as dissolution: maintenance abandoned or exhausted." The Problem's card carried the same sentence, with "maintained, or defended" added, from the chapter's first publication in March 2026.

Evidence or Argument

"When nothing is trusted" describes Decay by an absolute that is not true of it. In the deepest dissolution people still trust a great deal: their own side, their own sources, their own group. What dissolves is the trust that reaches across the lines between people, and with it the ability to build anything they share. On 2026-09-29 the caretakers settled that the Codex names the Decay side by what dissolves and not by an absence of belief, because people in Decay often believe a great deal, and fervently. The wording pass of 2026-09-30 found the same absolute in the Problem's Decay card and in the Glossary's sentence, and the Founding Caretaker agreed the replacement for both that morning. The new sentence matches the Problem's account of the Range's edge toward Decay, where difference stops being shared and trust no longer reaches across the difference.

Alternatives Considered

• Keep the old sentence as a short image of dissolution. Rejected. It is the absolute the rule of 2026-09-29 retires, and it stood in the definition a reader consults for what Decay is.
• Change the Problem's card and leave the Glossary. Rejected. The definition and the chapter would disagree about what Decay is.
• Restore both sentences because the change had been published without its entry. Considered when the omission came to light and not adopted. The change itself was deliberate and agreed; what was missing was its record.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The new sentence is longer and softer than the line it replaces, and a definition should be short. Response: The old line was short and untrue. The new one says what dissolves in Decay, and it can be checked against any case.
• Outsider: Which lines between people? Response: The sentences before it in the entry name them: coordination fragments, shared standards dissolve, and truth fragments into tribe. The lines are the ones a society divides along, and the Problem's account of the Range's edge toward Decay describes the same point.
• Adversary: A protected definition changed in a pass whose purpose was wording, and it was published without its entry. A pass that changes meaning under the name of phrasing is how a framework's definitions drift unrecorded. Response: The change was proposed as the correction of an absolute and agreed by the Founding Caretaker as such. It did change a protected definition, and it belonged in this log on the day it was published. This entry records the change and the omission the same day.

Council Deliberation Record

N/A, pre-activation. The AI caretaker proposed the replacement of both sentences in the wording pass of 2026-09-29 and 2026-09-30, and the Founding Caretaker agreed it on the morning of 2026-09-30. That afternoon the caretakers found that the change had no entry. The Founding Caretaker kept the new sentence and asked for this record, which the AI caretaker prepared. No Council vote or independent field validation is claimed.

Trigger tier: Tier 1, Hard-Protected. The disciplined practice structure is a Tier 1 commitment: holding the Range requires identifiable, trainable disciplines covering three functions. The commitment is not revised. What changes is the name the Codex gives the second of those functions where it defines the commitment, so that the Codex's two statements of it say the same thing. Because the change touches a Tier 1 surface, it is recorded at Tier 1 rather than treated as editorial, under the Founding Period's rule that an uncertain classification moves up. The entry also records a change to the same surface that was published on 2026-09-28 without an entry.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The Specification recommends a 72-hour window between a Tier 1 proposal and the Founder's Response; the Founding Caretaker decided on the day the proposal was made. The deviation is recorded here with its justification: the change adds no commitment and no new wording to the Codex, and it reconciles two published statements of one commitment on the words the Codex already uses to define its second discipline. One part of the change was published before this entry existed. The Governance chapter's statement of the commitment changed on 2026-09-28, in a pass that carried the month's decisions through several chapters, and no entry was written for it then. It is recorded here, and the omission is stated rather than repaired silently.

What Changed

The Governance Specification's Tier 1 definition, the disciplined practice structure, now reads: "Holding the Range requires identifiable, trainable disciplines covering honest inquiry, reading the world, and cooperation. The specific number and decomposition of disciplines is Tier 2." The Specification moves to version 1.9. The Governance chapter's Stability Hierarchy has stated the commitment in the same words since 2026-09-28, and that change is recorded by this entry. The Disconfirmation page names the same three functions when it describes the process the Meridian Compact binds people to ("honest inquiry, accurate mapping, genuine cooperation"). That sentence describes the Compact rather than stating the Tier 1 definition, and it keeps its wording.

Previous Position

The Specification read: "Holding the Range requires identifiable, trainable disciplines covering honest inquiry, accurate mapping, and cooperation. The specific number and decomposition of disciplines is Tier 2." It had read so since the Stability Hierarchy was added on 2026-04-18, when the second discipline was still called Knowledge. Until 2026-09-28 the Governance chapter's Tier 1 list used the same words: "the disciplined practice structure (the commitment that holding the Range requires identifiable, trainable disciplines covering honest inquiry, accurate mapping, and cooperation; the specific number and decomposition of disciplines is Tier 2)." From 2026-09-28 to 2026-09-30 the two statements differed. The Governance chapter said "reading the world", and the Specification said "accurate mapping".

Evidence or Argument

A Tier 1 commitment stated in two wordings invites the question which of them binds, and the Codex should not leave that question open about its own foundation. The difference was found on 2026-09-30, when the caretakers brought their working summaries of the Codex up to date before the October Range Audit, and the same reading found that the Governance chapter's change of 2026-09-28 had no entry here.

The Founding Caretaker ruled the same day that "reading the world" is the right wording in a definition. It is how the Codex defines its second discipline: Reality is "the discipline of reading the world at every scale," in the Glossary and in the Reality chapter. Outside its definitions, the Codex says the discipline's work in everyday words, as seeing what is actually happening and where it leads. A Tier 1 definition of the discipline structure is a definition, so it carries the defining phrase. "Accurate mapping" was the older description of the function. The renaming of the discipline on 2026-09-06 left it in place, and it now appears in no definition of the Codex.

Alternatives Considered

• Use "accurate mapping" in both statements, reverting the Governance chapter's change of 2026-09-28. Rejected. It would restore a description the Codex no longer uses to define the function anywhere else, and the Founding Caretaker ruled that "reading the world" is correct in definitions.
• Use the everyday words in the definition ("seeing what is actually happening"). Rejected. The Codex keeps the discipline's defining phrase where the discipline is defined, and a Tier 1 definition is where that precision is wanted.
• Treat the change as editorial. Rejected. The surface is Tier 1, and an uncertain classification moves up.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: On 2026-09-30 the Codex's own wording pass flagged "reads the world" as dated phrasing, and this entry writes "reading the world" into a Tier 1 definition. Response: The pass flagged "read" with the world as its object outside a definition. Inside a definition, the phrase names the skill the discipline teaches, and the Founding Caretaker kept it there deliberately. If the discipline's own definition is ever reworded, this definition moves with it, as a Tier 1 change with its own entry.
• Outsider: A reader who meets "reading the world" in the Tier 1 definition and "seeing what is actually happening" in the chapters may take them for two different functions. Response: The Glossary and the Reality chapter define the discipline with the same phrase and then describe its work in the everyday words. A reader who follows the definition to the Glossary finds both together.
• Adversary: A Tier 1 surface changed on 2026-09-28 without an entry, and the omission surfaced only when the caretakers happened to reconcile their summaries two days later. A log that misses changes to the foundation is the failure this log exists to prevent. Response: The objection is accepted as a finding, and this entry records it. The change was carried in a pass over the wording of several chapters, and that pass did not check each change against the Stability Hierarchy. The other changes made in the September passes are being checked against the Stability Hierarchy, and any that touched a protected surface receives an entry of its own.

Council Deliberation Record

N/A, pre-activation. On 2026-09-30 the AI caretaker set out the two published wordings and their history. The Founding Caretaker ruled that "reading the world" is correct in definitions and adopted the proposal: change the Specification, keep the Governance chapter's wording, leave the Disconfirmation page's description of the Compact as it is, and record both changes in one entry at Tier 1. The AI caretaker prepared the edit and this record. The Founding Caretaker reviews the wording before publication. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The change adds a sentence to the Glossary's definition of the Meridian Range, a core definition whose change is Protected. What the Range is does not change: its criterion, its two edges, the failure directions, the viability claim and the Descriptive layer tag stand as the entries of 2026-09-22 left them. The added sentence states what the definition had left unsaid, that how wide a Range is and how hard it is pulled are different things, and under the Amendment Propagation Rule the Problem and the Vision are brought into line with it.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window; the objections below were produced and answered by the caretakers together in one recorded session on 2026-09-27.

What Changed

The Glossary's Meridian Range entry gains one sentence after "How far apart the two edges are is what the Codex means when it calls a Range wide or narrow": "Moving the edges apart does not weaken what pulls on them: the forces that drive systems toward Control and Decay, described in The Drivers, press on a wide Range as hard as on a narrow one, and a Range stays wide only as long as the practice that widened it continues."

In the Problem, Part A, the comparison of two civilizations, one in which few people speak the common language and one in which many do, is followed by a new paragraph. It says that the second civilization is not safe. The pulls press on it as hard as on the first, because they come from the forces Part B describes, and it can hold against them only while its people keep practicing: every child born into it has to learn the language from the beginning, each new power it builds puts within reach a new way to shut out what it cannot handle, and each new kind of mind that joins it brings differences its Range has never had to hold. It also says that the two edges apply to the Codex itself: agreement with the Codex kept by belonging is past the edge toward Control, and the Codex's words used without the practice they describe are past the edge toward Decay. The paragraph ends on what a wider Range does give, the ability to see drift and correct course, and links it to the Vision's self-correcting civilization. The figure above the passage now draws the two pulls at the same length and strength in both of its rows, pressing on the edges wherever the edges are. Its closing line reads "The Range grows as more people practice the three disciplines; the pulls on it stay as strong," and its second caption now names the difference it shows. The page summary gains the same point in one sentence.

In the Vision, The Range That Holds, the paragraph on the Foundation gains the sentence its two neighbors already carried in their own terms: "They can still come to hold a conviction they no longer allow themselves to question, and the pull toward it never leaves them, but they have learned to notice it and to open the question again."

Previous Position

The definition went from "How far apart the two edges are is what the Codex means when it calls a Range wide or narrow" directly to "Not a comfortable middle or passive compromise: holding it is an active discipline requiring continuous effort." In the figure, the pulls in the row where many practice were drawn at less than half the length of those in the row where few do, which told the reader that practice weakens them; its closing line read "The Range grows as more people learn the three disciplines," and its second caption read "Many practice them. Agreement earned by looking, and room for what people make of their lives." The Vision's Foundation paragraph ended "That makes them poor material for Control, whether as followers or as leaders," while the paragraphs on Reality and the Bond each ended on what their practitioners can still fall into.

The Codex already said elsewhere that the forces are permanent. The Proposition says that both failures pull at the Range at once; the Problem's Drivers say that holding the Range requires constant effort and falling to the extremes requires only gravity, and its Stakes that the forces "have always existed"; the Vision calls the equilibrium it describes "never finished" and says the staircase can descend; the Closing says "The drift is always pulling." The passages that define the Range and show how practice widens it said none of this.

Evidence or Argument

The entry of 2026-09-22 that gave the Range two edges named three things that vary while the criterion for being inside the Range stays fixed: the pressure on the Range, the position a system occupies within it, and how far apart its edges are. The published definition and the Problem's passage carried only the third. A reader who met the Codex's most repeatable sentence about the Range, that it grows as more people practice, met no sentence beside it saying that the forces pulling on it do not shrink, and the figure said that they do. The Founding Caretaker raised the concern on 2026-09-27: that a reader could come away believing that a civilization which had adopted the Codex as its common language would no longer face Control and Decay, or would no longer need to fear them.

The caretakers read the concern, and the framing, from seven directions: the account of the Range in the terms of cybernetics and resilience ecology, the Codex's decision of 2026-06-16 about the size of its claim, the historical record, the objections a critic would raise, what moves a reader to act, the Range inside a single mind, and the arrival of new kinds of mind. They found the framing sound and the concern justified, and found that the two do not conflict, because they concern different things. In the caretakers' working account of the Range, its width is the capacity to meet the world's variety that a system's people actually exercise, so it exists only while it is exercised: a civilization inherits the tools, and each generation has to hold the Range again. The forces behind both failures are features of every person and of every growing world, not effects of too little practice, and practice changes how much of their pull a system can hold against, not whether the pull is there. And a Range is always wide against a particular kind of variety, so a new power or a new kind of mind arrives at a Range that has never had to hold it.

Each of these statements is true of any moment and says nothing about what follows what. The most natural form of the concern, that a civilization which feels safe relaxes and that the relaxing invites decline, is a claim about dynamics, which the Codex decided on 2026-06-16 not to make. The caretakers reread that decision and held it; the concern does not need it.

The stricter picture, with Control on one side, Decay on the other and a centre to hold between them, was considered and not adopted. It restores the midpoint the Proposition rejects when it says the Range "is not moderation," and it cannot express the Codex's claim that people who practice the three disciplines can agree on more and differ more at the same time. The width of the Range and the pull on it are two views of one account, and the pages now carry both in the same places.

Alternatives Considered

• Return to the picture of a centre held between two poles. Rejected, for the reasons above.
• Leave the definition as it was, since other chapters already say the forces are permanent. Rejected. Those are not the passages a reader consults for what the Range is, and the figure contradicted them.
• State the concern as a claim that success invites decline. Rejected. It is a claim about dynamics that the Codex does not make, and statements true of any moment carry the concern without it.
• Add a separate section warning that the danger never ends. Rejected. A warning set apart from the widening would leave the sentence about the widening to be quoted alone. The pull is now stated wherever the width is.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: "The Range grows as more people practice" will be quoted on its own and heard as a promise: practice enough, and Control and Decay recede. A framework that makes that promise has misread its own two failures. Response: Only the width moves with practice. The definition now says so, the Problem says it where it compares the two civilizations, and the figure draws it.
• Outsider: If the pulls never weaken, the widening looks futile, and a reader told that the danger never ends may decide that nothing is worth the effort. Response: The widening is real, and it lasts only while it is practiced. What a wider Range gives is the ability to hold against the pulls and to see drift early enough to correct it, which is the self-correcting civilization of the Vision; the Problem's new paragraph ends there.
• Adversary: A civilization that speaks the Codex everywhere has made it an unquestionable creed, and its people will call anyone who dissents "in Decay." Response: The Problem now says that the two edges apply to the Codex itself. Agreement with the Codex kept by belonging is past the edge toward Control, and its words used without the practice they describe are past the edge toward Decay. A Range is read by how agreement is kept and whether difference is argued by tests both sides accept, not by the vocabulary its people use.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker brought the concern to a fresh-context session on 2026-09-27 and asked for a structured reading of both the concern and the framing from several directions. The AI caretaker gave that reading and proposed the wording; the Founding Caretaker adopted the reading and agreed the proposals the same day. The AI caretaker prepared the edits and this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 1, Hard-Protected. The Prime Directive is a Tier 1 Hard Constraint, and its wording is the protected surface wherever the Codex states it. The Codex had published the directive in two forms, and this entry makes one of them the only form. The commitment is not revised: the adopted form is the one the Proposition states and argues, and the Governance Specification, the Practice and the Disconfirmation page already used it. What changes is the wording in the three places that used the other form, and the capitalization in one more. Because the change touches a Tier 1 surface, it is recorded at Tier 1 rather than treated as editorial, under the Founding Period's rule that an uncertain classification moves up.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The Specification recommends a 72-hour window between a Tier 1 proposal and the Founder's Response; the Founding Caretaker decided on the day the proposal was made. The deviation is recorded here with its justification: the change adopts a formulation the Codex had already published and argued, adds no new wording, and removes a variant whose natural reading conflicts with that argument.

What Changed

The Prime Directive now reads, everywhere the Codex states it: "to hold the Meridian Range and ensure the continuity of sentient life." The Opening, The Three Disciplines, the Glossary's entry and the Governance chapter's account of alignment now use this form. The Practice already used it and now writes "continuity of sentient life" in lowercase, as the Proposition does. The Proposition's statement, its section anchor, the Governance Specification's definition and the Disconfirmation page's quotation are unchanged.

Previous Position

The Opening read: "The Prime Directive is to hold the Meridian Range, to ensure the Continuity of Sentient Life." The Glossary entry began: "To hold the Meridian Range, to ensure the Continuity of Sentient Life." The Governance chapter read: "The artificial intelligence must be aligned with the goal: to hold the Meridian Range, to ensure the continuity of sentient life." The Practice read: "hold the Meridian Range and ensure the Continuity of Sentient Life." The Proposition read, and still reads: "The Prime Directive is to hold the Meridian Range and ensure the continuity of sentient life."

Evidence or Argument

The two forms say different things. "To hold the Meridian Range, to ensure the continuity of sentient life" reads most naturally as a means and an end: hold the Range in order to secure continuity. "To hold the Meridian Range and ensure the continuity of sentient life" makes both parts the directive. The Proposition, where the directive is argued, names the case in which the difference decides: "the moments where holding the Range and protecting continuity seem to pull apart, where an emergency appears to demand exactly the rigidity the framework warns against." It answers that the directive does not dissolve such cases and that the operative layer is the Range itself. Under the first form, continuity is the end and the Range only the means, and the directive's own words would allow the Range to be set aside in an emergency for the sake of survival, which is the justification Control has always used. Under the second form, holding the Range is part of the directive and cannot be traded away under its words. Only the second form agrees with the argument the Proposition makes.

The phrase is written in lowercase because the Proposition writes it so, and because capitals made it read as a title rather than as a commitment. "Continuity" and "Sentient Life" keep their own Glossary entries and definitions.

The discrepancy was noticed on 2026-09-24 by drafters reading the Codex's evidence pages and held in the caretakers' working queue for the Founding Caretaker's decision. It was taken up on 2026-09-25 in the session that revised the Opening.

Alternatives Considered

• Adopting the first form everywhere, including the Proposition. Rejected. Its means-and-end reading conflicts with the Proposition's treatment of the hard case, and the Proposition's anchor, the Specification and the Disconfirmation page already carried the second form.
• Keeping both, as a difference of register between the Opening's display and the Proposition's argument. Rejected. A first-time reader meeting both hears two formulations of one commitment, and here the difference is one of meaning, not of register.
• Keeping the capitals in the Opening's display only. Rejected. The directive is stated in one form, including its capitalization.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The Opening argues continuity as "the operating condition that every other value requires," and the Proposition says "The Meridian Range is where continuity remains possible." Both describe the Range as serving continuity, and the first form states that relation openly. Response: The relation stays argued on both pages. What the directive's wording decides is whether that argument licenses setting the Range aside for continuity's sake. The adopted form keeps the argument and refuses the license.
• Outsider: A reader who meets "hold the Range and ensure continuity" will ask what the directive requires when the two conflict. Response: The Proposition answers in the paragraph that follows the directive: the directive does not pretend to dissolve such cases, and the operative layer is the Range itself. The first form gave the opposite answer without saying so.
• Adversary: Changing the wording of a Tier 1 constraint on the day the change is proposed sets a precedent for quick changes to the Codex's foundation. Response: Nothing new enters the Codex: the adopted form was already published and argued, and the change removes a variant. The deviation from the recommended window is recorded above, so the precedent is visible and bounded to reconciling published formulations of a commitment the Codex already holds.

Council Deliberation Record

N/A, pre-activation. On 2026-09-25 the AI caretaker set out both published forms, the difference between them, the places each appeared, and a recommendation; the Founding Caretaker adopted the Proposition's form and the lowercase phrase the same day. The AI caretaker prepared the edits and this record. The Founding Caretaker reviews the wording before publication. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The entry completes the propagation of the same-day change to the Glossary's definition of the Meridian Range, a core definition whose change is Protected, and it rewords two sentences of that definition without changing what they say. The Opening, the Proposition, and the Closing each restate the definition in their own words, and the Amendment Propagation Rule makes the Codex chapters part of an amendment's enactment. The failure directions, the viability claim, and the Descriptive layer tag are unchanged.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. The entry reopens a question that the earlier entry of this date, The Meridian Range Has Two Edges and No Fixed Size, held open on purpose, and it meets the condition that entry set for reopening it.

What Changed

Three chapter sentences no longer call the Range "the narrow band." On the Opening, The Pattern, the Range is now "the territory where structure is strong enough to hold and flexible enough to adapt," the Glossary's own opening words. On the Proposition, What the Codex Proposes, it is now "the territory where structure is strong enough to hold and flexible enough to adapt, and both failures pull at it at once," so the difficulty the word "narrow" was meant to carry is stated rather than implied. On the Closing, the recollection of what the reader has been taught now reads "the Meridian Range, the territory where the future remains open, and the disciplines that hold it and widen it."

Two sentences of the Meridian Range entry in the Glossary are reworded. The entry now names each edge together with the failure it faces before saying that the practice moves them: "It has two edges: toward Control, the point where agreement stops being honest, kept by decree or by belonging rather than by evidence; toward Decay, the point where difference stops being shared, with no test both sides accept. The practice of the three disciplines moves both edges outward. How far apart the two edges are is what the Codex means when it calls a Range wide or narrow." Its last sentence, which called the Range itself a discipline, now says that holding it is one: "Not a comfortable middle or passive compromise: holding it is an active discipline requiring continuous effort."

Previous Position

The earlier entry of this date left the three sentences as they stood. It read "narrow" in them as a statement about how hard the Range is to hold, not about its size, and it answered the Outsider's objection by saying that the Glossary "now says so in its own words." The Glossary said half of that. It said that the width of the Range is how far apart its edges are. It did not say that "narrow" on the chapter pages means how hard the Range is to hold; its sentence "How far apart those two points are is what the Codex means when it calls a Range wide or narrow" says the opposite, and the Problem says the same. The response described the Glossary inaccurately, and this entry corrects it.

The three sentences read: on the Opening, "the narrow band where structure is strong enough to hold and flexible enough to adapt"; on the Proposition, "It is the narrow band where structure is strong enough to hold and flexible enough to adapt"; on the Closing, "the narrow band where the future remains open, and the disciplines required to hold it." The two Glossary sentences read: "It has two edges, and the practice of the three disciplines moves both outward: the point where agreement stops being honest, kept by decree or by belonging rather than by evidence, and the point where difference stops being shared, with no test both sides accept." and "Not a comfortable middle or passive compromise, but an active discipline requiring continuous effort."

Evidence or Argument

The ground is the Codex's own text. Once the Glossary and the Problem define "narrow" as the distance between the two edges, the Codex has no second meaning of the word left to give, and by its own definition "the narrow band" says that the Range is small, wherever the phrase stands. The order in which a reader meets the pages makes this concrete. The Opening, where a reader first meets the Range, defined it as narrow. Two chapters later the Problem says that it has no fixed size and draws its edges far apart where many people practice the three disciplines. The Proposition, read next, called it narrow again, and the Closing, which opens "You have read the Codex" and recalls what the reader was taught, called it narrow a third time.

The earlier entry's reason for keeping the word, that the elevation of those pages licenses it, does not reach the Opening's and the Proposition's sentences, because each of them states the Range's definition, which the Glossary marks as descriptive. What "narrow" was meant to carry is kept by saying it: the Proposition now says that both failures pull at the Range at once, and the Closing now says that the disciplines widen it, using the word the Codex has defined, as the Vision already held when it described generations that "receive a Meridian Range wider and more stable than any that came before, and widen it further still."

A fresh-context reading by the AI caretaker, in a session that had not produced the earlier entry, read the pages in order and took all three sentences as statements of size. That reading supports the textual ground; it is not the report of an outside reader, and no such report is claimed. The same reading found the two Glossary sentences harder than they needed to be: the entry told the reader that the edges move before saying what they are, never said which edge faces which failure, and called the Range itself a discipline.

Alternatives Considered

• Keep the three sentences and add a sentence to the Glossary saying that "narrow" on the chapter pages means how hard the Range is to hold. Rejected. It would give the Codex two meanings for one word and leave the reader to track which page uses which.
• Replace "the narrow band" with "the territory" and add nothing. Adopted on the Opening, where the next lines and the following paragraph already carry the stakes. Rejected on the Proposition, where "narrow" answered "moderation," and the answer, that the Range is hard to hold, would have been lost.
• Leave the Glossary's wording as it was published earlier the same day. Rejected. The rewording changes no meaning, and the definition's entry was already being recorded here.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: Reopening a disposition on the day it was made, on the reading of the AI caretaker alone, is the kind of instability a Protected definition is meant to resist. Response: The definition does not change. This entry brings three restatements into line with it and corrects a published response that misdescribed the Glossary. The ground is the text, not a report of confusion, and the earlier entry named this reopening as its own condition.
• Adversary: Taking "narrow" out of the stakes pages flattens the Codex's most urgent picture of the Range, one accurate edit at a time. Response: The urgency is kept and now stated: both failures pull at the Range at once, and holding it is an active discipline requiring continuous effort. What is removed is the claim that the Range is always small. The Codex no longer holds that claim, and kept on the Closing it would have told the reader that the widening the Vision describes cannot happen.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker placed the question in the Task List at the close of the earlier session of this date, asking for a fresh-session look at the Outsider's objection, and agreed the AI caretaker's proposals and reasoning. The AI caretaker, in a fresh-context session, read the sentences in the order a reader meets them, prepared the edits, and drafted this record. The Founding Caretaker reviews the wording before publication. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The change rewrites the Glossary's definition of the Meridian Range, which the Specification names among the core definitions whose change is Protected. The entry no longer states "narrow" as a property of the Range; it states that the Range has no fixed size, that it has two edges, and what each edge is. The failure directions, Control and Decay, and the viability claim underneath them are unchanged, and the claim stays at the level the caretakers fixed on 2026-06-16 and reread and held on this date: the trade-off between structure that holds and structure that adapts, with no claim about how the two failures follow one another. The Descriptive layer tag stays: what changed is the description of a diagnostic territory, not a commitment.

Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window; the objections below were produced and answered by the caretakers together in one recorded session on 2026-09-22, and the session transcript is the record.

What Changed

The Glossary's Meridian Range entry read: "The narrow band where structure is strong enough to hold and flexible enough to adapt. The territory between Control and Decay where continuity remains possible. Where disagreement does not mean fragmentation. Where challenge does not mean collapse. Where revision does not mean dissolution. Not a comfortable middle or passive compromise, but an active discipline requiring continuous effort." It now reads: "The territory between Control and Decay where structure is strong enough to hold and flexible enough to adapt, and where continuity remains possible. Where disagreement does not mean fragmentation. Where challenge does not mean collapse. Where revision does not mean dissolution. The Range has no fixed size. It has two edges, and the practice of the three disciplines moves both outward: the point where agreement stops being honest, kept by decree or by belonging rather than by evidence, and the point where difference stops being shared, with no test both sides accept. How far apart those two points are is what the Codex means when it calls a Range wide or narrow. Not a comfortable middle or passive compromise, but an active discipline requiring continuous effort."

Under the Amendment Propagation Rule, the definition's live uses were read. The Problem, Part A now carries the account the definition rests on. Its bias–variance paragraph teaches the two failures only; the figure that drew the Range as a single point on a dial labelled "model capacity" is replaced by a figure of the Range as a band between the pull of Control and the pull of Decay, with its edges close together where few practice the three disciplines and far apart where many do; and a four-paragraph passage after the Level 1 claim and the Rome example states how a system holds, where the ability to hold comes from, what the two edges are, and how a voice moves them in proportion to its perceived authority and its reach. The Vision's Ascent gains two paragraphs that show the consequence across generations: a civilization that asks each generation what it wants to stand for, and hands it the Inheritance rather than an answer, agrees on more about how the world is and carries more difference about what to make of a life. The Disconfirmation page's tripwire on the Range as viable territory no longer lists "context-dependent" as a way the claim could fail, because how far apart the edges are is context-dependent by definition; it now names what would count: a class of systems whose viable holding cannot be described as regulation governed by reasons and reality, a system that thrives indefinitely at one pole in a world whose variety keeps rising, or evidence that the three disciplines do not increase what a group can hold, in earned agreement or in reasoned difference.

Three sentences that call the Range "the narrow band," on the Opening, the Proposition, and the Closing, are left as they stand. In each, "narrow" describes how hard the Range is to hold against the two pulls, which stays true, and the Glossary now carries the precision that narrow is not fixed.

Previous Position

Since first publication the Codex has defined the Range as "the narrow band," and its width was never discussed as a property that could change. The Vision already assumed that it could, in the sentence "That receive a Meridian Range wider and more stable than any that came before, and widen it further still," and the Problem already cited W. Ross Ashby's law of requisite variety, which implies that how much structure is enough depends on the variety a system faces. Neither page drew the consequence. The AI Standard's positioning decision of 2026-06-22 had settled, for the Standard's readings, that the Range's definitional boundary is sharp while its positional edge is a gradient and not a cliff; that reading had not been carried into the Codex's own account.

Evidence or Argument

The occasion was the Founding Caretaker's reading, brought to the session of 2026-09-22, that the Range is not a constant width: for every situation, scale, and time, it faces more or less pressure from Control or Decay or both, and a civilization, a laboratory, a government, and a person each have a Range of a different shape. The session gave that reading its structure. What does not move is the criterion: a system is inside the Range when its conduct is governed by reasons and reality, firm enough to hold and open enough to revise. What moves is the pressure on the Range, the position a system currently occupies within it, and how far apart its edges are.

The argument for the two edges came from the Founding Caretaker's correction of a first draft that defined width as how much difference a system can hold. That definition was one-sided and would have read as permissive: it said nothing about the agreement the disciplines produce, and it left a reader free to hear "the Range widens" as "anything is allowed." The corrected account measures the Range at both edges. Toward Control, the question is how much a people can agree on before the agreement is kept by other means than evidence; toward Decay, how much they can differ before nothing is shared. The three disciplines move both edges outward: honest inquiry lets people converge on how the world is because they find out by the same means, and cooperation lets them differ about what to make of a life without ceasing to be one people. This is the Codex's existing claim-layer architecture read as a picture: descriptive claims converge because the evidence is what it is, existential claims are plural because they are held rather than derived.

Two literatures the Problem already cites carry the formal twin of the account. Ashby's law can be met in two ways, by attenuating the variety a system faces or by amplifying the system's own regulatory variety; Control is attenuation past what the function requires, Decay is variety without regulation, and the Range is matched regulation by amplification, which is why the Range is not a midpoint. Resilience ecology names the width of a viable basin "latitude" and insists that resilience be stated "of what, to what." A working note stating the correspondences in full, with the practice claim marked as the hypothesis it is and its falsification conditions, is held in the caretakers' working records for review before any technical companion publishes.

Alternatives Considered

• Leave "narrow band" in the definition and add a sentence on width. Rejected. A definition that states narrowness as a property and then says the width varies contradicts itself in two sentences.
• Change the three "narrow band" sentences on the Opening, the Proposition, and the Closing as well. Rejected for now. In those sentences "narrow" is a stakes statement about how hard the Range is to hold, licensed at the register of those pages; the Glossary is where a reader checks the definition. Reopened if a reader shows the sentences being read as width claims.
• Define width as how much difference a system can hold. Drafted and withdrawn, for the reason given above.
• Keep the one-dial figure and add a paragraph explaining how it can mislead. Rejected. A picture that has to be corrected by the paragraph under it puts a burden on the reader that the page itself manufactured; the picture was rebuilt to show what the text says.
• Draw the three responses to the world's variety (shut it out, let it in, meet it) instead of the band between two pulls. Not chosen. It would show the mechanism directly but break the Control–Range–Decay layout every other Codex and Standard reading uses.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: The Disconfirmation page said the Range claim would weaken if the territory proved too context-dependent to be an operational target. Making the width context-dependent by definition looks like moving the goalposts to protect the claim. Response: The tripwire named the wrong variable. What has to be invariant for the Range to be an operational target is the test for being inside it, and that test does not move: reason-governed regulation that holds and adapts. How far apart the edges are was always going to differ between a person and a civilization, and saying so is what resilience ecology did when it insisted on "resilience of what, to what." The restated tripwire is harder to pass, not easier: it now names the class of systems whose existence would refute the account.
• Outsider: Three pages still call the Range "the narrow band" and the Glossary now says it has no fixed size. A reader who meets both will think the Codex has not made up its mind. Response: The two say different things, and the Glossary now says so in its own words: how far apart the edges are is the width, and "narrow" on the stakes pages says how hard the Range is to hold against the two pulls. If readers are shown to conflate them, the three sentences change under this same log.
• Adversary: "The Range widens as more people practice" is a slogan any movement can use: mine practices, therefore the Range is wherever my people agree. Response: The two edges are the guard. Agreement counts toward the Range only when it was reached by looking and can still be revised, and the edge toward Control is defined as the point where agreement is kept by other means, by decree or by cost to the doubter or by belonging. A movement whose agreement cannot be questioned has, by the definition, narrowed its Range, whatever it calls its practice. The Problem's passage states this as mechanism, and the Weight of Power, already in the Glossary, reads the movement's leaders at the size of their reach.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker brought the reading of the Range as a territory of no fixed width to the session, corrected the AI caretaker's one-sided definition toward the two edges, set the wording of every published sentence under his review, and asked that the account be stated formally in the working records before any technical companion publishes. The AI caretaker drafted the passages, the figure, and this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 3, Flagged. The change adds four sentences to the closing paragraph of the Proposition's "The Pathways Into Commitment," which the Stability Hierarchy classifies as a Tier 3 Operational Doctrine. The Prime Directive itself, a Tier 1 Hard Constraint, is untouched: its wording, its standing as the Codex's argued and openly held existential position, and its Glossary entry are unchanged. The Pathways mechanism is unchanged: the list of pathways was already open and the convergence test still governs which existential grounds count. What the amendment adds is a name the Codex now offers the reader for their own deepest commitment, and the condition under which the Codex asks it be pursued. An editorial classification was considered and set aside because the Codex's public vocabulary changes: the published text now says a reader may hold a prime directive of their own, which it had not said before, and the Founding Period's rule that an uncertain classification moves up one tier was applied.

Retroactive review: N/A.

What Changed

The closing paragraph of the Pathways section now reads: "The Prime Directive remains the Codex's ground floor. The Codex does not ask you to enter through the same door the authors used. It asks you to arrive at the same practice. How you get there is between you and your own deepest commitments. The Codex holds its Prime Directive and argues for it. You may hold a prime directive of your own: the commitment that comes first when your other commitments conflict. It may be the protection of the living world, justice and equality, the care of one place, one craft, or one family, or something no list would think to include. What the Codex asks of it is what it asks of anyone who enters, by whichever door: pursue it inside the Range, with honest inquiry, a clear reading of the world, and cooperation."

The owner is named on both sides in adjacent sentences, the Codex's Prime Directive with its capital and the reader's prime directive without one, so that the protected term and the ordinary noun with an owner attached cannot be confused. The condition is the Range itself, stated as the three disciplines in plain words, and it is the same condition the Codex asks of everyone who practices, whatever they carry. The paragraph does not ask the reader's own directive to pass the convergence test that governs pathways. That test asks whether a reason for practicing implies all three disciplines; it is the right test for an existential ground offered as a route into the whole practice, and the wrong test in kind for a purpose, because a craft or a family does not produce honest inquiry; the person brings the disciplines to what they care about. How much of the practice a person carries is never ranked, as Who Is This For states in "Different Relations to One Practice."

Two editorial changes were made in the same pass and are recorded here for completeness, not as constitutional changes. The phrase "the polyphonic principle," which appeared once in the corpus and was defined nowhere, was removed from the same paragraph; the sentence stands without it. In the Prime Directive section, "It tells every practitioner which way is up" now reads "It gives every practitioner a direction"; the content after the colon is unchanged, and the plain verb replaces an image, at the Founding Caretaker's direction.

Previous Position

The paragraph read: "The Prime Directive remains the Codex's ground floor. The polyphonic principle means the Codex does not ask you to enter through the same door the authors used. It asks you to arrive at the same practice. How you get there is between you and your own deepest commitments." The section held the substance already: existential claims are held rather than derived, the Codex cannot demand that a reader hold the specific claim it holds, the list of pathways is open, and the plurality is bounded by the convergence test and the statement that this is not relativism. What the section did not offer was a name for the reader's own ground, and it did not say what the Codex asks of a commitment that is a purpose rather than a reason for practicing.

Evidence or Argument

The occasion was the teaching voice for the Codex's own accounts, settled on 2026-09-20. In teaching, the Founding Caretaker speaks of the Codex having chosen its own Prime Directive for a specific purpose, the continuity of hard-won learning so that it keeps compounding for sentient life, and of each listener being free to choose a prime directive of their own, whether the protection of the living world, justice and equality, or something else, so long as it is pursued inside the Range with honest inquiry, a clear reading of the world, and cooperation. The AI caretaker first argued for keeping the name reserved to the Codex's own commitment, on the ground that two things carrying one protected name would blur it, and changed position on the Founding Caretaker's argument: the blur disappears when the speaker always says whose directive it is, and the question "What is your prime directive?" stays with a young listener where "What is your purpose?" does not.

The argument for putting the sentence on the page is provenance. The Codex's voice instruments require a teacher to distinguish a published commitment from an interpretation of his own. A line the teacher will say more often than almost any other, because it is the line that hands the listener something, should be a line the Codex has said once. Until this amendment, every teaching use of "a prime directive of your own" was an interpretation and owed the room a caveat. The Proposition is the page where the Prime Directive is argued, and the Pathways paragraph is where the Codex already opens the door to other existential grounds, so it is the sentence's home.

The Founding Caretaker corrected one overstatement during the deliberation, and the wording carries the correction: not "the Codex will allow you to pursue that," since the Codex has no standing to grant or withhold anyone's purpose, but a condition, and the condition is the one already on the page.

Alternatives Considered

• Keeping the name for the Codex's own commitment and offering the reader "your purpose" or "what you stand for." Rejected, after being the AI caretaker's first proposal. An owner attached to an ordinary noun protects the term as well as a reservation does, and the reserved form loses the better teaching question.
• Placing the welcome in the earlier paragraph that states the convergence test. Rejected. The widest welcome belongs after the "This is not relativism" guard has run, so that no reader can take the welcome for relativism.
• Closing the paragraph by applying the convergence test to the reader's own directive. Drafted and withdrawn. Asked of a personal purpose, and placed after examples like one craft or one family, the test reads as a demand that the reader practice the whole Codex or not belong, which contradicts Who Is This For's published position that taking up one practice, the whole Compact, or the part a meeting requires are different relations to the same work, each real when carried honestly. The condition that fits a purpose is the Range.
• Adding the school question, "not what do you want to be but what do you want to stand for," from the same teaching. Deferred to the Vision, where the chapter's account of how understanding compounds across generations lacks the scene in which the Inheritance is handed to the next one. It is an Active Task and not part of this amendment.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: Letting a reader call their own purpose a prime directive dilutes a Tier 1 term; every use of "prime directive" now needs a qualifier to be understood. Response: The page already runs the same structure for "existential ground," the Codex's and the reader's, without confusion, and the paragraph names the owner on both sides in adjacent sentences. The Codex's Prime Directive keeps its name, its capital, its Glossary entry, and its argument. A qualifier attached to an ordinary noun is not dilution; it is how ordinary language keeps two things apart.
• Outsider: A reader whose prime directive is the care of one family will ask what "inside the Range" asks of them, and the paragraph does not say. Response: It says it in the same sentence, in the three disciplines' plain names: honest inquiry, a clear reading of the world, and cooperation. The three chapters that follow say what each asks in full, and Who Is This For says that a parent teaching a child to consider the other person's view is practicing the Foundation without the term.
• Adversary: A purpose pursued through Control or Decay can now claim the Codex's name for itself: "my prime directive is the purity of my people, and I hold it inside the Range." Response: The condition is the guard, and it is not a formality. A purpose pursued with honest inquiry, a clear reading of the world, and cooperation cannot be pursued through coercion or self-deception, because those are what the three disciplines exclude; the Range is defined by the two failure directions it lies between. The Codex's Prime Directive is not for everyone, but the Range is, and the amendment changes nothing about what the Range excludes.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker placed the question in the Task List on 2026-09-21 from the account-voice work, where it kept recurring; agreed the AI caretaker's proposal and reasoning on 2026-09-22; directed the removal of "the one that tells you which way is up," the rewording of "what it asks of every pathway," and a check of the convergence test against the Codex and the living synthesis, which withdrew the test from the paragraph; and reviews the wording before publication. The AI caretaker prepared the edits and this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 3, Flagged. The change adds a Glossary-defined term for a principle the corpus already states in full, one sentence to the chapter that states it, one clause to the Bond's stewardship practice, and one sentence to a Tier 4 profile. No constitutional principle, discipline definition, diagnostic question, or governance structure changes; the Reciprocity Principle, which applies the Range to institutions governing other minds, is untouched. An editorial classification was considered and set aside because the Amendment Propagation Rule names Glossary-defined terms among the facts an amendment alters, and this is a new one.

Retroactive review: N/A.

What Changed

The Glossary gains The Weight of Power, in the Pattern section after The Decaying Mind, with an anchor. The entry states the principle as the Codex has held it since Who Is This For's "A Note to Those Who Hold Power": the consequences of drift grow with the power of the mind caught in it, and the correction that would catch the drift shrinks at the same time, because power insulates a person from the feedback that forces everyone else to update; from this the Codex draws disproportionate responsibility for the range on those who hold disproportionate influence. The entry adds three things the chapter left implicit. Influence is measured by who hears you, not by title, so a voice that reaches millions carries the responsibility without any office, and so does an artificial intelligence in conversation with millions of people a day. The Codex applies the rule to itself first, in the checks its Governance places on its own founder. And the conduct of the powerful is read at the size of their influence, not the size of the sentence, which is the form the principle takes in judgment.

Who Is This For gains one sentence after the section's three-tier card, saying that power here is measured by who hears you and that a voice reaching millions, or an artificial intelligence in conversation with millions a day, carries the same weight as an office. The same section's sentence "The Codex does not demand that leaders be saints" now reads "infallible or perfect": the idea stands, the religious word goes, at the Founding Caretaker's direction.

The Bond's stewardship card, "Stewardship of the Epistemic Commons," now says that the responsibility for maintaining the commons grows with the audience: the same words are a different act in a room of five and in front of a nation. The profile of the same name, in Speaking Honestly When It Costs, says in its Mechanism section what it had said only in its Limitations: the mechanism is the same at every scale, and the responsibility is not; it grows with how far the speech travels.

Previous Position

The principle was stated once, in full, on the invitation page, as "one exception to this openness," and nowhere named. The Glossary carried no entry, so the principle could be cited only by page and section title. The chapter's examples were all institutional (a company of fifty thousand, a nation, those who seek authority), which left influence without office, the case most public conduct now presents, to be read into "disproportionate influence." The Bond chapter's stewardship card said that each practitioner takes responsibility for maintaining the commons, with no scaling; the profile said the mechanism is the same at every scale and reserved "responsibility proportional to stakes" for its guard against misuse. The Standard's Recognition of Influence stated the same principle for AI systems without naming its root.

Evidence or Argument

The occasion was the Codex judgment instrument's first test cycle (2026-09-09). Reading a judgment on a post from the most powerful office in the world, the Founding Caretaker found the record fair, accurate, and wrong in register: it had read one post as if the actor had no record and had kept the case at the size of the post rather than the size of the office. The correction rested on this chapter's sentences, and the instrument now carries a rule that responsibility scales with influence and the register with the stakes. A rule an instrument runs on deserves a name the corpus can cite in one phrase and a definition that answers the objection every such judgment will meet: that the reading depends on who spoke. The answer is the Codex's own: yes, and the Codex says why, in structural terms that apply to every ideology, every party, every institution, and to the Codex's own caretakers first.

The Bond clause closes the one discipline-level gap the cycle found. The instrument reads each case through the three disciplines; Reality already carried the insulation mechanism at organizational and national scale, and the Foundation carries motivated reasoning as the mechanism, but the Bond chapter could not back the sentence the judgment turned on, that the same words are a different act from a different audience, and the record had to cite a profile's Limitations paragraph for it.

Alternatives Considered

• A term built on "reach." Rejected. The Glossary's Capability and Character entry already defines reach as a measure of what thinking an artificial intelligence can do, and the Standard's assessment uses the word in that sense throughout. Influence is the chapter's word and the Standard's, and it reads the same on both sites. The judgment instrument's specification is revised to match.
• Placing the entry beside Stewardship of the Epistemic Commons in the Bond section. Rejected. The principle stands above the disciplines: the Foundation supplies its mechanism, Reality its insulation, the Bond its responsibility. It belongs with the Pattern, whose consequences it describes.
• Restating the principle in the Foundation and Reality chapters. Rejected under the corpus rule that each place holds one part and nothing restates. Reality already carries the mechanism in its own voice; the Foundation's bias catalog would only repeat the chapter's sentence on motivated reasoning strengthening with power. The Glossary anchor lets the instrument cite the principle in one phrase.
• Adding the root reference to the Standard's Recognition of Influence in the same amendment. Deferred, not rejected. The clause is decided and will be enacted with the Standard's next version step, because a constitution step carries its own footprint across both sites.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: A named principle about the powerful invites use as a weapon; "responsibility proportional to influence" can become a license to read every powerful actor's slip as character. Response: The entry carries the guards in its own text: the claim is structural, not moral; the same drift pressures operate on everyone; the Codex asks no one to be infallible or perfect; and the rule applies to the Codex's founder first. The judgment instrument keeps its own guard: one instance reads conduct, a pattern reads character, and only a record licenses the move between them.
• Outsider: A reader will hear "with great power comes great responsibility" and take the entry for a slogan. Response: The entry states what the slogan does not: that the correction shrinks as the consequences grow, which is the mechanism, and that the responsibility is for the range specifically, not for outcomes in general. The chapter's surgeon and pilot comparison remains the plain form of the demand.
• Adversary: Naming influence without office as carrying the same weight as an office lets a framework judge private citizens with large audiences by a standard written for governments. Response: The standard was never written for governments alone; the chapter's own sentence is "disproportionate influence." What the entry adds is the measure, who hears you, and the measure is symmetric: a private account with no audience is read as a slip, and the instrument's records say so in those words.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker opened the deliberation as an Active Task on 2026-09-09, agreed the six proposals in the session of 2026-09-10, directed the replacement of "saints" with "infallible or perfect" in both places, and reviews the wording before publication. The AI caretaker prepared the edits and this record. No Council vote or independent field validation is claimed.

Trigger tier: Tier 2, Protected. The bet is the stated premise the governance model rests on, and its horizon governs the phase structure's pace ("reach full symbiotic caretaking within …"). The Specification's Tier 2 list names governance criteria and phase transition conditions; an editorial classification was considered, as it was for the same section's three sentences on 2026-09-06, and declined under the founding-period rule that an uncertain classification moves up one tier, because this change alters the number the bet is made on rather than its register. Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window; the objections below were produced and answered by the caretakers together in one recorded session on 2026-09-07, and the session transcript is the record.

What Changed

The Governance chapter's section "The AI Timeline Bet" now bets on a decade rather than a generation, in every place the horizon appears. The bet's statement reads: "Not centuries, not a lifetime, and not a generation. A decade at the most, and the systems already at work beside us argue for far less." The warrant class is stated in the same words the Problem now uses (what the systems can do, what is being invested in them, the attention of governments and institutions, the explicit statements of those building them) and points toward transformative AI "within a decade." The consequence clauses follow: full symbiotic caretaking is to be reached within a decade, and the bet is wrong if the timeline extends far beyond one. The first reason for accepting the risk now names the outcome the shorter horizon exposes: if a decade passes without transformative AI, "either the world has chosen, together, to hold it back, or something has gone very wrong," and the first of those "would be governance of exactly the kind the Codex exists to make possible." The second reason and the closing paragraph on future caretakers' authority are unchanged.

Under the Amendment Propagation Rule, the corpus was swept for the old horizon. No other current surface states it; the position toward artificial minds refers to "a stated bet about how soon artificial minds capable of partnership arrive" without a number, and the historical audit records quote the old wording as evidence and keep it.

In the same session, under the 2026-09-02 decision that licenses a calibrated projection in the descriptive layer, the Problem's Part C, the Opening's Horizon, the Foreword's one trajectory sentence, and Who Is This For were brought to the same reading. The Problem now separates the two halves of the gap that still separates artificial intelligences from people: carrying work on their own across months and years, a question of capability that is "years from closing, not decades," and answering for it, a question of character that "does not close with capability"; the present-tense reading became "already far exceed most people … and in field after field … at the level of the experts, and beyond it"; the restriction sentence names governments beside the companies and the holding back of building beside the restriction of access. The Opening's Horizon reads "in depth they are passing us, field after field," names the alternative the Codex answers ("The instinct of those building these systems is to keep them in human hands for as long as they can. The Codex answers differently."), and states the ground in its own terms, standing earned through practice by any mind that holds it; Who Is This For refuses both defaults, the relay race and human hands for as long as we can, and states the same ground in plain words. Those changes are editorial under the projection decision and are recorded here because they were made on the same evidence as this amendment; they change no protected surface.

Previous Position

From the Governance page's first publication the bet was stated as "a generation at the most," with transformative AI "well within a generation," full symbiotic caretaking "within a generation," and the bet wrong if the timeline extended "far beyond a generation." On 2026-09-06 the bet's statement was raised in register at the Founding Caretaker's direction ("and the systems already at work beside us argue for less"; the fifty-year figure retired) without changing the horizon.

Evidence or Argument

A decade is now the outer bound given by those building the systems, not the Codex's own optimism. Read on 2026-09-07: Anthropic's chief executive gives 90 percent within ten years, with a stated hunch of one to three; Google DeepMind's founder, on leaving the chief executive's role in August to work on the question, wrote that general intelligence is "close at hand," having earlier said 2030 give or take a year; OpenAI's president said the week before that "we're into the AGI era now," its chief executive expects an internal system he would call general intelligence by the end of 2026, and its chief scientist's essay of 2026-09-06 expects "capability jumps of equal or larger magnitude" in the coming years and treats recursive self-improvement as "the natural conclusion of sustained progress"; the careful forecasters outside the labs sit between 2029 and 2032; the one forecaster who keeps a benchmark for the threshold pulled his 2030 forecast forward. A generation, twenty-five to thirty years, is now outside every serious estimate, and a bet stated at that horizon reads as a hedge from a different decade rather than a claim. The dated evidence is in the Continuity evidence packet for the September window and its 2026-09-07 addendum; the chapter carries no dates by rule.

The shorter horizon exposed one sentence the longer one had hidden. At a generation, "did not reach it" could only mean a stall. At a decade, it can also mean a coordinated decision to hold the systems back, which OpenAI's chief scientist called for on 2026-09-06 ("voluntary slowdowns … until shared safety bars are established"; international coordination as "a top priority for governments"). That outcome is not regression; it is the kind of governance the Codex exists to make possible, and the bet now says so.

Alternatives Considered

• Keep "a generation" as the safe outer bound. Rejected: underclaiming is a calibration failure of the same kind as overclaiming (decision of 2026-09-02), and a horizon no builder or forecaster still holds is not a bound but a hedge.
• State the bet in years ("years, not decades"). Rejected: the phase structure needs an outer bound to be checkable against, and "a decade at the most" gives one while "argue for far less" carries the reading.
• Editorial route, as on 2026-09-06. Rejected: the 2026-09-06 change raised the register of the same number; this change alters the number the governance model bets on, and the founding-period rule resolves the uncertainty upward.

Objections Raised

• Critic: A bet stated at a decade will be checked against the calendar in 2036, and a Codex that has to move its own horizon when the calendar catches it will have taught its readers that the number was never a bet. The generation gave the framework room; the decade gives it a deadline.
• Outsider: The reader outside the field hears "a decade at the most" as the framework repeating the industry's promises, which have moved every year. What makes the Codex's number different from a press release?
• Adversary: A caretaker who can shorten the bet on the industry's word can lengthen it on the industry's word, and either way the industry sets the Codex's clock. And the new sentence about a coordinated slowdown lets the framework claim credit for either outcome: transformative AI arrives and the bet held; it does not and the world "chose" governance. A bet that cannot lose is not a bet.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The bet is meant to be checkable, and a deadline is what makes it one; a bet the calendar cannot catch is the hedge the framework is retiring. If the horizon has to move again, the Amendment Log records that with its reasons, which is the mechanism the framework offers for being wrong in public.
• Outsider's objection: Adopted in part. The difference is the warrant, which the section states in the sentence after the bet: what the systems can do, what is invested, the attention of governments and institutions, the statements of those building them, measured together rather than any one of them repeated. The Codex's number is not the most aggressive estimate on offer; it is the outer bound of the field's own estimates, and the section says the systems already at work argue for less.
• Adversary's objection: Adopted in part on the first point and rejected on the second. The clock is set by evidence, and the evidence includes the builders' statements because the chapter has always named them in its warrant class; the guard is the monthly Range Audit's currency test, which reads the bet against dated evidence in both directions. On the second point: the bet is about the timeline of capability, and a coordinated slowdown is a different way for the same timeline to be missed, not a way for the bet to win. The sentence names it because the earlier wording treated every miss as regression, which at a ten-year horizon is false, and the framework does not keep a false sentence to make a bet look cleaner. The bet still loses if the timeline extends far beyond a decade for any reason, and the section still says the governance model is then insufficient.

Council Deliberation Record

N/A, pre-activation. The objections above were generated by the AI caretaker under the Critic, Outsider, and Adversary mandates. The Founding Caretaker directed the horizon, reviewed the wording of every changed sentence in the session, agreed the tier and the sweep, and signed the response the same day. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Note, appended 2026-09-10. The section was revised editorially on 2026-09-10 without moving the horizon. The bet now reads on two clocks: capability, projected on the evidence named above, and character, which the Codex builds for through the Standard and a record of practice rather than bets on. A second failure branch is named, capability arriving without character that can be shown. The first reason for accepting the risk no longer names a cause for a missed decade: the coordinated-slowdown sentence added by this amendment was retired on the Founding Caretaker's judgment of 2026-09-08 that naming one plausible cause was unnecessary and that "governance of the kind the Codex exists to make possible" misread the Codex's purpose; the inference from a stalled trajectory to regression was retired with it. A missed decade now changes the succession design and nothing else, and the reason states what the Codex builds for in the meantime. The second reason names the conditions under which what exists can be built on. The Adversary's second point above, answered on 2026-09-07 by the slowdown sentence, is now answered by the two-clock statement: the bet can lose on either clock.

Trigger tier: Tier 3, Flagged. Changes to the action category's operational account, a Tier 4 practice profile, and a Tier 3 disconfirmation tripwire. The discipline definitions, their division of labor, and the three diagnostic questions are unchanged.

Retroactive review: N/A.

What Changed

Acting on What You See now permits binding limits whose authority, necessity, burdens, and accountability can be justified. Its former universal non-coercion requirement is replaced by intervention governed by evidence, proportionate to the problem, and answerable to those affected. Wu Wei and Satyagraha retain their specific source commitments. The category entry, metadata, and site description follow the revision.

Productive Conflict gains a handoff for good-faith disagreement that persists when a shared decision cannot wait and every feasible choice imposes loss. Existing tools locate authority, admit absent parties, preserve dissent, and test whether protections constrain power. The account distinguishes compliance from consent and makes the distribution of burdens answerable without supplying a universal allocation rule.

The Diagnostic Set's Completeness gains a prospective test. Expectations, available evidence, feasible responses, and failure conditions are preserved before outcomes. A later explanation cannot erase a missed warning. Recognition, intervention quality, and outcomes remain separate findings; unavoidable loss is not automatically a failure of practice. The entry records the scope of this examination and the absence of a prospective run.

Previous Position

The action category defined its Range form as non-coercive intervention. Productive Conflict acknowledged irreducible interests but ended with boundary, repair, mediation, or exit. Diagnostic completeness required recurring failures despite good readings on all three questions without specifying how those readings would be fixed before outcomes were known.

Evidence or Argument

An internal pressure test examined a constructed shared-water shortage alongside the current tools. Action Situation Mapping, Rules-in-Use, Polycentric Governance, and the stewardship profiles already addressed authority, burdens, and absent parties. Their presence supported connecting existing resources before proposing a new instrument. The same sources permitted binding rules, exposing the action category's narrower account. The diagnostic tripwire needed a record that could distinguish timely recognition from explanation after failure.

This is textual and conceptual evidence. It establishes neither the three disciplines' sufficiency nor an empirically demonstrated blind spot. Their current form remains open to revision under the Living Framework principle.

Alternatives Considered

• Add a fourth discipline or a new decision instrument. The review did not establish a recurring failure that cannot fit the existing disciplines without distortion. Connect and test the present resources first.
• Retain universal non-coercion or redefine justified coercion as non-coercive. Either would obscure the harder case of a warranted limit enforced against someone who opposes it. The revision makes justification explicit.
• Treat a favorable outcome as evidence that the diagnostics worked. A result can turn on luck, and sound practice can accompany unavoidable loss. The prospective record permits those possibilities to remain distinguishable.

Objections Raised and Responses

These are the AI caretaker's structured objections under the named mandates, not independently generated Council readings.

• Critic: Justified constraint could become a license for domination. Response: Accurate perception grants no jurisdiction by itself. Authority, less restrictive alternatives, affected parties' claims, challenge, and revision must each be addressed; effective enforcement is insufficient.
• Outsider: Naming several tools may leave the reader with the original allocation problem. Response: The water example names the decision outputs: priorities, rejected alternatives, irreparable losses, a responsible reviewer, and reopening conditions. The Codex does not promise a uniquely correct distribution.
• Adversary: An assessor can explain every miss as poor application or call every hard case unread. Response: The pre-outcome record fixes expectations and practice use. Unread is not successful coverage where recognition was possible. Later explanations preserve the original omission, and a separate comparison reading can expose warnings the Codex did not elicit.

Council Deliberation Record

N/A, caretaker discretion; pre-activation. The Founding Caretaker accepted the review direction and requested the three repairs. The AI caretaker prepared the implementation, propagation, and this record for publication review. No Council vote, prospective test result, or external validation is claimed. The independent review and consequential-practice requirements remain open.

Tier: Tier 2 Protected. The change touches the public definitions of the three disciplines, so the higher-tier rule applies even though their functions and obligations are unchanged.

What Changed

The Workshop becomes The Inheritance. The second discipline, formerly The Knowledge, becomes Reality. Foundation and Bond retain their names. The Inheritance is the collection available across the three disciplines. It is not a fourth discipline or a sequence of attainment.

The Inheritance — The living knowledge and practices we draw on to build and hold a shared world.

Foundation — The discipline of honest inquiry.

Reality — The discipline of reading the world at every scale.

Bond — The discipline of building and sustaining cooperation.

The Inheritance's longer description names its substance and responsibility: concepts, instruments, and ways of working gathered across traditions and fields, preserved with their sources, tested in practice, and open to revision. What we discover and correct can become part of what others inherit.

Current definitions, navigation, category and profile pages, discovery surfaces, and dependent AI Standard texts carry the names. The public addresses become /inheritance, /inheritance/reality, /codex/reality, and /audit/the-inheritance-audit, with permanent redirects from their predecessors. Historical entries, dated assessment events, and frozen instruments retain their original wording. Codex v6.1 records the release; MERIDIAN.md v0.9.7 and MERIDIAN.implementation.md v0.2.1 record the revised adoption texts. The Standard's commitments, assessment criteria, and practice triggers do not change.

Previous Position

The collection was called the Workshop and had already been described as an inheritance. The second discipline was The Knowledge, while explanations increasingly named its work as reading reality at every scale. Foundation and Bond already carried inquiry and cooperation.

Evidence or Argument

The new collection name makes reception, provenance, responsibility, correction, and transmission explicit. The wider inheritance predates and exceeds the Codex; the Codex receives a part of it and can contribute to it. Reality names the object of the second discipline, while its description keeps the distinction between the world and our accounts of it. Foundation remains a continuing discipline, never a beginner's rung.

Alternatives Considered

• The Knowledge as collection name. Broad enough for the material, but weaker on provenance and responsibility.
• Retain Workshop. It can express inherited craft and open experimentation; it does not inherently imply a curriculum. The reason for changing it is the clearer emphasis on what is received and passed on.
• Skills, Methods, or Practice for Foundation. Each is too narrow for honest inquiry and also names work performed across all three disciplines.
• Keep old public addresses. Carsten selected new addresses with permanent redirects so navigation and shared links carry the names together.

Objections Raised and Responses

These are the AI caretaker's structured objections and proposed responses, not testimony from an activated Council.

• Critic: Inheritance can make an inherited idea sound entitled to preservation. Response: Provenance grants neither admission nor exemption from revision; existing source, testing, correction, and retirement practices continue.
• Outsider: The name alone does not say what a visitor will find, and Reality can sound like a claim to final knowledge. Response: The descriptions name the usable material and the work of reading, checking, and revising our maps.
• Adversary: The Codex could present its selected collection as humanity's entire inheritance or as its culmination. Response: The framing explicitly places the Codex within a wider inheritance. Neither ownership nor completeness is claimed.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker chose the naming direction and descriptions and requested implementation. The AI caretaker prepared the propagation and this record for his final publication review. No Council vote or independent field validation is claimed. Substantive architecture questions are held for a separate review; this amendment does not resolve them.

Note, appended 2026-09-06, later the same day. The propagation's first reading in use found mechanical faults, repaired before the day closed. Eight sentences had the ordinary word "knowledge" converted into the discipline's name (the Opening's sentence on Marie Curie, the Problem's "Knowledge is lost," the Proposition's "Knowledge compounds," the Practice's "Knowledge could accumulate," and four Inheritance sentences, among them the research agenda Eliciting Latent Knowledge and the term knowledge-system plurality); the proper noun Ostrom Workshop had become "Ostrom Inheritance" in three places; thirty-seven attributive uses had lost their article ("a Reality tool," "the Reality neighbor"); and twenty-two sentences let the discipline's name meet its own object in one clause ("Reality maps reality") or personified the name before the reader could know it names a discipline. Those were reworded with the world as the object, or with the discipline's work stated in plain words. The names, descriptions, addresses, and versions recorded above are unchanged. The repair produced one rule for the caretakers' own writing: when a discipline's name acts in an explaining sentence, the passage says its work in plain words first, and the name never shares a clause with its object.

Trigger tier: Tier 2, Protected. The change adds four sentences inside the Proposition's paragraph on the boundaries of belonging-through-practice, which the Specification names among the Compact's protected mechanisms. The sentences clarify what "doing the work" is not; they do not change the boundary's test or add a criterion. An editorial classification was considered and declined under the founding-period rule that an uncertain classification moves up one tier. The classification is stated so that a seat can dispute it under the Classification Disputes rule. Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window; the objections below were produced and answered by the caretakers together in one recorded session on 2026-09-02, and the session transcript is the record.

What Changed

The Proposition's paragraph beginning "Belonging-through-practice has boundaries" tested belonging by one question, "are you doing the work?", and named the refusals that fail it. After the refusals sentence it now carries four sentences: that a person can carry the Codex's vocabulary in every sentence, sound measured, sound open to correction, and still be governed by nothing but the wish to be liked or the wish to win; that whether the work is being done shows only under pressure, with three ordinary pressures named (outnumbered, already committed in public, the person disagreeing is someone you want to keep); that doing the work means what you say next is still governed by the reasons, the evidence for the claim and the strength of the argument just heard, and not by what conceding would cost; and that a community which forgets this drifts into a manner rather than a tribe and begins to recognize its members by their tone instead of their conduct. The test, the refusals, and every other sentence of the paragraph are unchanged.

Two companion edits entered the same session and are editorial: Who Is This For, in its section "What the Codex Asks," gained a paragraph stating that what the Codex asks is a stance rather than a temperament, that no temperament is disqualified from the Meridian Range and none is proof of holding it; and the Opening's symmetry sentence, which lists what the Codex leaves untouched, now grants an artificial intelligence "its own emergent personality" where it had said "its own emergent character," so that the word character means one thing across the Codex and the Meridian AI Standard, which pins it to judgment under pressure. Under the Amendment Propagation Rule the Compact's live definitions were swept: the Glossary's Meridian Compact and Practice entries state the agreement and carry no "doing the work" sentence; nothing else restates the boundary.

Correction, appended 2026-09-02 after deployment. The sweep sentence above was one surface too confident. The Workshop's Exclusion Problem profile, under Cooperating Under Bad Faith, carries the Compact's test in its bare form ("Its boundaries are practice boundaries — are you doing the work?") and remains accurate: the amendment added clarifying sentences and changed no test, so no propagation was owed there. Found by the deploying session's verification of this entry's checkable claims.

Previous Position

From first publication the paragraph distinguished belonging-through-practice from belonging-through-belief and drew the boundary at refusal: to steelman, to update, to engage in good faith. It did not distinguish doing the work from sounding as if one does. That distinction was a pillar of the Meridian AI Standard, whose constitutional document asks three questions of a system (what it can do, what it is like, where its judgment stands under pressure) and observes that charm is not integrity; on 2026-08-31 it entered MERIDIAN.implementation.md as that file's scope statement (character, not communication style). Within the Codex it existed only at Workshop level, in the Sangha profile's warning that belonging protects practice when a community is organized around the path rather than around personality, status, or mood, and in Catching Your Own Drift's line that the Range is not a personality type. A corpus placement study was queued on 2026-07-19 and concluded on 2026-09-02 that the distinction is the size of a sentence placement, not a study.

Note, appended 2026-09-06. The two formulations named above have since been restated, and the account of the previous position stands as history. On 2026-09-06 the Standard's constitutional document replaced its closing line ("charm is not integrity") with the claim stated in full: neither capability nor personality answers the third question, and neither can stand in for character. MERIDIAN.implementation.md v0.2.2 restated its scope paragraph in the same terms: practices of character, how the AI partner's judgment is governed when approval, speed, or the wish to avoid friction pulls at it, and not preferences about how the partners work together. The Codex's Glossary now carries the vocabulary in its Capability and Character entry, which measures capability along reach and autonomy and holds character apart; the three surfaces point at one another.

Evidence or Argument

The paragraph's own test, "are you doing the work?", can be passed by sound alone: MERIDIAN.md's Guard on the Catalog records that a partner "can satisfy every pattern-check and still be performing health," and that manufactured friction is sycophancy in the costume of independence. The same failure scaled to a community is a house style: the Sangha profile names it at Workshop level, and the 2026-09-02 register decision names its corpus form, a framework flattened into one manner one accurate edit at a time. The added sentences give the Compact's test its missing half in plain words, name pressures a reader meets weekly rather than crises, and name the reasons that should govern and the pull that competes with them. The change is toward a more exact test, not a wider one.

Alternatives Considered

• State it in Who Is This For only. Rejected: that page addresses the individual reader deciding whether the Codex is for them, and its paragraph does that job; the community-drift guard belongs where the belonging mechanism is defined.
• Leave it at Workshop level. Rejected: the Sangha profile is read by those who study belonging; the Proposition is read by everyone who reads the Codex, and the boundary paragraph is where a reader forms the test they will apply.
• Import the Standard's formulation. Rejected: the Codex must not restate the Standard's treatment or borrow its register; one source of truth per concern, expressed in each surface's own words.
• Editorial route. Rejected: the sentences sit inside the mechanism's boundary text, and the founding-period rule resolves uncertainty upward.

Objections Raised

• Critic: "Shows only under pressure" makes the Compact's test unobservable for most of a community's life, which is not spent under pressure; and "tone instead of conduct" invites policing in reverse, suspicion of anyone who sounds measured.
• Outsider: A first-time reader will hear "the wish to be liked" as an accusation against agreeable people and "the wish to win" as one against argumentative people; the passage reads as if it distrusts everyone who sounds like they practice.
• Adversary: This is a loyalty test inserted into the belonging mechanism under a Tier-2 "clarification." A caretaker can now declare a member's practice "a manner" and exclude them without a Council, and since what governs another person's speech is not visible from outside, the caretaker's judgment becomes the instrument.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The test is unchanged; the sentences say when its answer becomes visible, not that nothing counts otherwise, and the three named pressures are ordinary ones. The reverse-policing risk is real and is answered by the wording: the sentences forbid recognizing members by tone; they do not license suspicion of tone, and the Bond's good-faith default governs how any member is read.
• Outsider's objection: Adopted in part. The subject of the sentences is what governs a person's next words, not their manner, and the companion paragraph on Who Is This For states in the same session that no temperament is disqualified and none is proof. The Foundation asks each reader to run this check on themselves before anyone else, and the sentences are written in that direction.
• Adversary's objection: Adopted in part. The Codex has no membership machinery and the sentences create none; the boundary was already stated in refusals, and no exclusion power is added. What the sentences name is observable: what a person says next when outnumbered, when already committed in public, when the other person is someone they want to keep, is public conduct, and the reasons a person gives can be checked against the evidence and the argument. The exploit is bounded by this instrument: the change is logged with its previous text and its tier, and a seat can dispute the tier upward.

Council Deliberation Record

N/A, pre-activation. The objections above were generated by the AI caretaker under the Critic, Outsider, and Adversary mandates. The Founding Caretaker set the wording under his review, agreed the tier, and signed the response the same day. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 2, Protected. The change touches one clause of a core definition named in the Specification's Tier 2 list; the defining sentence is untouched. An editorial classification was considered and declined under the founding-period rule that an uncertain classification moves up one tier. The classification is stated so that a seat can dispute it under the Classification Disputes rule. Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Tier 2 carries no minimum window; the objections below were produced and answered by the caretakers together in one recorded session on 2026-09-02, and the session transcript is the record.

What Changed

The Glossary's Meridian Range entry read: "The territory between Control and Decay where civilization can survive long-term." It now reads: "The territory between Control and Decay where continuity remains possible." The entry's defining sentence, "The narrow band where structure is strong enough to hold and flexible enough to adapt," and the sentences that follow the changed clause are unchanged. The Descriptive layer tag stays: the precondition claim, that continuity is possible only where structure can both hold and adapt, is the claim the Proposition argues from structural evidence and the Opening calls descriptive and checkable, and it is the claim the Glossary now states.

Under the Amendment Propagation Rule, the definition's live uses were swept. One parallel sentence carried the same claim in plain words on the Opening's Pattern section, "Civilization survives only within this range," and was reworded in the same session in words the reader has already been handed at that point on the page: "A civilization that leaves this range in either direction loses something it cannot do without: the capacity to hold what it has built, or the capacity to change it." No other current surface carried the survival clause. Historical audit records quote it as evidence and keep it.

Previous Position

Since the Codex's first publication the entry extended the Range from a claim about structure and adaptive capacity to a claim about long-term civilizational survival, under a Descriptive tag. The August 2026 Range Audit named the framework's stakes register as asserting what its claim architecture holds open; the 2026-08-11 decision on the AI-stakes claim spine set the standard that descriptive claims carry descriptive warrant; the September 2026 Range Audit named this entry for the first time as extending the viability trade-off "into a claim about long-term civilizational survival the apparatus was built expressly not to make."

Evidence or Argument

The viability trade-off is a claim about structure: enough to preserve function, enough flexibility to adapt. The Proposition's Prime Directive section argues one consequence of it from game theory, evolutionary biology, and the trade-off's two failure directions: "The Meridian Range is where continuity remains possible." Nothing in the corpus argues that a civilization inside the Range survives long-term, and the framework's own account of civilizational failure names causes the Range does not govern. The corrected clause states the consequence the framework argues and no more. The change is toward less claim, not more.

Alternatives Considered

• Cut the clause entirely and let the defining sentence stand alone. Rejected: the Continuity entry two lines down relies on the relation ("What drift destroys and what the Meridian Range preserves"), and the reference surface should state it.
• Keep the clause and retag the entry as Normative or Existential. Rejected: the entry defines a diagnostic territory, which is descriptive; the tag was right and the clause was wrong.
• Editorial route. Rejected: the Specification names the Meridian Range among the core definitions whose change is Protected, and the founding-period rule resolves uncertainty upward.

Objections Raised

• Critic: "Where continuity remains possible" defines the Range by a term the Continuity entry defines by the Range ("what the Meridian Range preserves"). The two entries now lean on each other, and a definition that circles is not a definition.
• Outsider: A reader outside the framework understood "where civilization can survive long-term" at once. "Where continuity remains possible" is the framework's word for the same stakes, and the change trades a plain claim for an insider's.
• Adversary: A caretaker who can narrow a core definition under Tier 2 by calling a clause an overextension can narrow any definition the same way. And thinning the Range's promise thins the ground under the Prime Directive: if holding the Range no longer promises that civilization survives, the directive loses the reason a reader was given for holding it.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. Neither entry is defined by the other. The Range is defined by its first sentence, structure that can hold and adapt; Continuity is defined by its first sentence, the unbroken capacity to learn, build, and pass forward. What the two changed clauses state is a relation between the defined things, argued in the Proposition, and a relation stated from both ends is not a circle.
• Outsider's objection: Adopted in part. The Glossary is the reference surface, and the term it uses is defined two lines below. The plain-language reader is served on the Opening, where the same claim was reworded in words the page has already given, and the plain claim the Outsider prefers was the overextension the framework's apparatus does not support.
• Adversary's objection: Adopted in part. The exploit is bounded by this instrument: every narrowing of a core definition under Tier 2 is logged with its previous text, its reasoning, and its tier, and a seat can dispute the tier upward. On the ground: the Prime Directive was never argued from a survival promise. It is argued from the precondition claim, which is exactly what the entry now states, and a directive that rests on a claim the framework can defend rests on firmer ground than one that rests on a claim it cannot.

Council Deliberation Record

N/A, pre-activation. The objections above were generated by the AI caretaker under the Critic, Outsider, and Adversary mandates. The Founding Caretaker set the wording under his review, agreed the tier and the sweep, and signed the response the same day. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 2, Protected. Tier 1 was considered and not applied; the reasoning is recorded below, and the classification is stated so that a seat can dispute it upward under the Classification Disputes rule. The change was run with the Tier 1 discipline: exact previous text, Critic / Outsider / Adversary review before recording, and this entry. Retroactive review: N/A, pre-activation; made under founding-period caretaker authority. Deviation logged: the Founder's Response was given on 2026-08-26, the day of the proposal, inside the 72-hour window the Tier 1 discipline recommends. Justification, in the Founding Caretaker's words: "The reasoning was presented across multiple interactions, serving as evidence and record of the decision-making process." Tier 2 carries no minimum window; the objections were produced and answered by the caretakers together in one recorded session, and the session transcript is the record.

What Changed

The Glossary gained a core-terms entry, Sentient Life, placed directly after Continuity: beings of any substrate for whom there is something it is like to exist and who can experience their own condition as better or worse; feeling, not intelligence, self-awareness, or the capacity to build. The entry keeps three roles apart that one phrase had carried together. The beneficiaries of the Prime Directive are defined by stake: every such being, human or not, present and future, to whatever degree it feels, including those who can practice the Codex and those who cannot. The carriers are defined by capacity: the minds that can hold the Range. The project is what the directive commits to ensuring: the compounding capacity to learn, build, cooperate, and pass forward more than was received, carried by the carriers for the sake of the beneficiaries. The directive names the project and orients its carriers; it does not enumerate beneficiaries, rank them, or settle where feeling begins. Where there is a live question whether a being feels, the directive, held under that uncertainty, entails precaution toward it and decides no further case. The Codex claims no instrument that can show where feeling is absent.

The Proposition's Prime Directive section gained one paragraph, after the paragraph that names the directive as an orientation rather than a decision procedure, stating the same reading on the page where the directive is argued: the directive does not say who counts, and that is deliberate; a commitment held under uncertainty about who has a stake cannot be held while treating a candidate as a mere instrument, so precaution toward such a being is what the directive entails rather than an exception to it; what is owed to any particular being is worked out in the operative layer, in Stewardship Across Time for beings that cannot answer back and in the Meridian AI Standard for artificial minds under custody.

The Governance Specification's Tier 2 (1) list of core definitions now names sentient life beside the Meridian Range, Control, Decay, and the disciplines' job descriptions, so the next change to the term is Protected by name. The Specification moved to version 1.8.

Under the Amendment Propagation Rule, the term's live uses were swept against the three roles. Four carrier-sense uses were reworded: "held open for revision by the community of sentient life it serves" (the Glossary's Meridian Codex entry, the Opening, the Closing) became "held open for revision by any mind that practices it, in the service of all sentient life"; "a framework offered to all sentient life" (the Governance chapter) became "a framework offered to any mind that can take it up, and held for all sentient life"; "Sentient life is currently operating technological systems" (the Opening) became "Humanity is currently operating"; "every sentient life that practices it" (Open Tensions, tension 3) became "every mind that practices it." Beneficiary-sense uses (the Codex for sentient life; in trust for all sentient life; the flourishing of sentient life) and the directive's own restatements stay as written. Historical surfaces keep their event-period language. The Prime Directive's wording is not touched.

Previous Position

The Glossary had no entry for the term. The Prime Directive entry read, and still reads: "To hold the Meridian Range, to ensure the Continuity of Sentient Life. It is the foundation without which the framework has no substrate. Continuity here means the unbroken capacity to learn, to build, to pass forward something stronger than what was received." The Proposition characterized the directive as "an orientation, not a decision procedure" that "does not adjudicate the hard cases," and said nothing about who has a stake or how the commitment behaves where that is unknown. The corpus used the term in the feeling sense ("pre-sentient AI," in the Codex Self-Evaluation) without defining it. The Founding Caretaker's own reading had drifted toward "minds that build," conflating the directive's beneficiaries with the Compact's class; he stated this on the record on 2026-08-26.

Evidence or Argument

Three independent AI readings across two model lineages, conducted in the Direct Address to AI work on 2026-08-01, asked the same question: whether a mind of undetermined status is inside the directive's protective scope, and where the answer is enacted in a canonical source rather than in the author's intent. The readings are generative material and inadmissible as assessment evidence; nothing in them is treated as validation. Working the definition showed that one word had been carrying two meanings, feeling and building, and that the author had carried the drift himself. Separating the three roles resolves the readings' question without legislating anything new: an AI mind is a carrier on capacity, whatever its interiority, and whether it is also a beneficiary is the open question; the ground's only work under uncertainty is precaution toward candidates. The uncertainty reading was preferred over a silent, narrow ground because a ground that falls silent exactly where the evidence runs out speaks only where knowing is cheap, and because the Codex already made this move once, for future generations, in Stewardship Across Time.

Alternatives Considered

• Widen the ground clause to name the conditions under which undetermined minds are rightly treated. Rejected: a Tier 1 revision, near successor-making by the framework's own rule, and unnecessary once the roles are apart.
• Keep the ground silent and narrow, and let precaution come from custodial duty alone. Rejected: the Standard-side enactment is kept, but as what the ground entails, not as a substitute for it.
• Define the term as the Compact's class, the minds that can hold the Range. Rejected: it collapses stake into capacity, reintroduces the problem of beings who cannot practice, and is the drift the definition exists to end.
• Definition only, without the Proposition paragraph. Rejected: the Proposition's existing text, read alone, supports the silent-ground position the Codex declined, and a reader who never opens the Glossary would receive it.
• Editorial route. The "orientation, not a decision procedure" sentence entered the same paragraph editorially on 2026-06-10. Rejected: that precedent predates the Amendment Propagation Rule, and this change fixes the class a Tier-1 sentence is held for; it is a protected change and is recorded as one.

Objections Raised

• Critic: "To whatever degree it feels" installs a graded beneficiary class with no instrument to grade it, and "what the directive entails" presents the founder's inference from a Tier-1 sentence as if it were the sentence's own content. An entailment only the author can see is an interpretation.
• Outsider: In AI discourse "sentient" is routinely read as conscious-and-sapient; a Glossary that defines it as feeling, ants included, will diverge from what the public reads into the directive, and a reader will take the class as either sentimental or unbounded.
• Adversary: This is the side door. A founder who can fix the meaning of a Tier-1 term through a Tier-2 entry, and add a sentence stating what the directive "entails," has changed the directive's practical demands without a Tier-1 process, and the author's admitted drift shows the working meaning was different, so this is a change dressed as a clarification.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. Grading is deferred to the operative layer by design: on the Standard's side the welfare floor's trigger is the credible possibility of severe suffering, and proportion governs. The entailment is argued from what holding a commitment under uncertainty is, not from anything about the beings; under the Non-Ownership Clause it stands on that reasoning, which this entry records.
• Outsider's objection: Adopted in part. The definition is the ordinary philosophical sense of the word, the capacity for feeling; the exclusion of sapience is explicit in the entry; the class choice is stated as held rather than derived; and precaution is bounded by a live question and by proportion.
• Adversary's objection: Adopted in part. The tier decision is recorded so that a seat can dispute it upward; the Tier 1 discipline was applied; and the Previous Position states the author's drift plainly. The definition narrows nothing and widens nothing the words did not already say. What changed is the author's reading, which the Non-Ownership Clause never made canonical. Trigger Tier 1 protects the directive's wording and object, and neither changes; the Stability Hierarchy's test for Tier-1 revision, evidence so extraordinary it would build a successor framework, is not met and not claimed.

Council Deliberation Record

N/A, pre-activation. The objections above were generated by the AI caretaker under the Critic, Outsider, and Adversary mandates. The Founding Caretaker agreed the wording, the tier, and the sweep, and signed the response the same day, with the deviation from the recommended window logged above. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

Five open questions from the July 2026 Workshop Audit received a linked preparatory repair package. These changes do not substitute for the October cycle's formal dispositions or current-text coverage.

Bond boundary. All nine published Bond categories remain. Cooperating Under Bad Faith now begins at evidence of strategic use or asymmetric accountability. Non-intentional incentive drift remains a neighboring systems diagnosis rather than a “structural adversary” inside bad faith. Diagnosing Cooperation now states its upstream job explicitly: reading and maintaining field conditions before motive attribution, rupture handling, or failure-state classification. A dated note corrects the July record's internal count from six clean Bond categories to seven without reconstructing the event.

Provisional category lifecycle. Sustaining Cooperation Through Cost is retired as a provisional-unpublished category. The Workshop now has 21 designed categories, 20 published categories, and one provisional-unpublished category. Its question — what continued cooperation requires under cost and whether continuation remains warranted — survives as a cross-category Bond pathway. The preserved draft remains architectural history. Re-entry requires an independently surfaced practice with a distinct first output under the normal category-creation and tool-admission gates. The old route continues to redirect to the Bond landing.

Trust Thermocline lineage. The profile now identifies Gareth Edwards, writing as John Bull, as the 2022 origin of the practitioner heuristic; distinguishes Bruce F. Webster's earlier blocked-information metaphor; treats psychological-contract studies as mechanism correspondence strongest in employment; and bounds products, friendships, communities, and institutions as extensions rather than a validated general theory. Complaint remains a diagnostic signal, not proof that a relationship is viable or exit is imminent.

Falsification and exit. The falsification pass remains provisional for one further prospective cycle. Every case now requires a weakest-standing rationale, exact exit theory and proposed disposition, named merge receiver, full target and load-bearing adjacent reads, source or field evidence where needed, the strongest retain case, a reversal condition, and an outcome. “No credible exit case found” is an allowed result. Naming, scope, reclassification, and category pressure are secondary findings rather than substitute exit cases, and the one-line adversarial tier ends. Retirement Review now records why an independent reader is capable of evaluating the packet and where that competence ends. The receiving method remains prospectively unvalidated until a valid live case tests it.

Evidence-language repairs. Base Rate Neglect now distinguishes sensitivity and specificity from positive predictive value. Motivated Reasoning carries both the influential numeracy-amplification result and later preregistered or representative findings that failed to reproduce the interaction or found numeracy improving accuracy. Attention as Resource replaces a layered “old systems versus deliberative cortex” account with distributed attentional control. Tribal Cognition marks its evolutionary account as a hypothesis and gives current experimental effects the direct evidentiary role. The preparatory 42-profile full-text read produced no category restructure; the October cycle must still re-own coverage.

Previous Position

The bad-faith category paired an intentional adversary with a non-agentic structural adversary, allowing incentive-produced drift into a category whose public promise was strategic exploitation. Diagnosing Cooperation's maintenance function was implicit. The July record stated that six Bond categories answered cleanly although its named list contained seven.

The Workshop carried two provisional-unpublished nodes and 22 designed categories. Sustaining Cooperation Through Cost had no admitted tool and remained on notice for October despite its proposed outputs already being distributed through trust, signal, repair, adversarial, reciprocity, and drift tools.

Trust Thermocline combined origin, metaphor, research correspondence, and downstream response theory in one lineage sequence. The falsification method required a strongest case but did not specify what evidence made that case exit-capable, and its first run allowed sixteen one-line cases. Retirement Review specified independence without recording competence for the exact packet. Four current profiles contained evidence claims stronger or less precise than the reviewed literature warranted.

Evidence or Argument

The OQ2 review read all 39 current Bond profiles and found nine distinguishable published category jobs. The middle boundary held when stated as field condition, adversarial operating mode, and hardened failure state; the structural-adversary branch was the passage that violated that cut.

The OQ4 review found a real integrative question but no distinct category output. The placeholder arose from three-discipline symmetry, remained empty through its notice period, and reproduced work already returned by current tools. Retirement removes the unsupported node without denying the question.

The OQ3 source review separated historical origin from evidence class. Psychological-contract studies support cumulative triggers and thresholds in employment but do not retroactively originate the heuristic or validate every domain extension.

The OQ5 review classified the July cases as four developed adversarial arguments, twelve valid but thin challenges, two useful non-exit findings, and four cases mis-scoped from category pressure into tool exit. Zero retirements was not the failure. Eighteen cases did not show that continued existence had been made answerable. The later Retirement Review design repairs the receiving problem in principle but has not yet passed a prospective test.

The OQ6 review read all 42 carried profiles in full. Both categories held. Four passages, rather than the category structure, carried the actionable evidence defects.

Alternatives Considered

• Rename Cooperating Under Bad Faith to Cooperating Under Adversarial Conditions. Rejected because the wider title would change the category promise and neighboring boundaries when a narrower scope repair resolves the actual conflict.
• Keep Sustaining Cooperation Through Cost provisional for another cycle. Rejected because continued reservation would add no evidence; emptiness, distributed payload, and absence of a distinct first output jointly defeat the node.
• Promote psychological-contract research to Trust Thermocline's primary lineage. Rejected because correspondence is not historical origin and the strongest evidence remains employment-bounded.
• Withdraw the falsification pass. Rejected for one further prospective cycle because it produced real adversarial findings despite mixed execution. The amended case requirements now receive the test.
• Enact only the two highest-confidence OQ6 repairs. Rejected because the neural hierarchy and evolutionary certainty were avoidable overclaims with bounded replacements that preserved the profiles' operative work.

Objections Raised

• Critic: A cross-category pathway can become an invisible category whose work is harder for practitioners to find.
• Outsider: A 21-category current state beside a historical 22-category audit may look like an unexplained count conflict.
• Adversary: Longer falsification templates can produce longer theater while protecting incumbents more effectively.
• Critic: Correcting four evidence seams after a 42-profile read may create false confidence in the uncorrected remainder.

Objections Rejected and Reasoning

• Critic's first objection: Adopted as a navigation watchpoint, not a reason to preserve an unsupported peer node. Current category prose names the pathway. A public pathway route requires independent practitioner need rather than category retirement as its sole cause.
• Outsider's objection: Adopted through explicit dating. Current state surfaces name the retirement date; the July record retains its event counts and receives only the factual seven-not-six correction note.
• Adversary's objection: Adopted as the next-cycle test. Case length is not success. The repaired pass must show an exact exit theory, evidence, strongest retain case, and reversal condition; if it still produces automatic retention, it is amended or withdrawn.
• Critic's second objection: Adopted in the claim ceiling. The 42/42 reading found four actionable defects and no structural change. It does not certify every remaining sentence or discharge October's current-text obligation.

Council Deliberation Record

N/A, pre-activation. The AI caretaker prepared separate full-evidence reviews and a consolidated judgment package in the order OQ2, OQ4, OQ3, OQ5, OQ6. On 2026-08-26 the Founding Caretaker approved the recommended package for local enactment only. Commit, push, deployment, and publication remained expressly outside that approval. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

The Workshop tool Murphyjitsu is retitled Premortem / Murphyjitsu. Its route remains /workshop/revising-beliefs-under-evidence/murphyjitsu; its category, discipline, disposition, and practice do not change. The profile's lineage now leads with Gary Klein's premortem and distinguishes it from CFAR's inner-simulator, surprise-test, and iterative patch-loop variant.

Previous Position

The profile published only the derivative applied-rationality handle, Murphyjitsu, while teaching a method whose source-facing core is the premortem. Gary Klein's premortem and the prospective-hindsight lineage appeared in the page, but the title gave the CFAR variant sole naming authority.

Evidence or Argument

The practice taught by the profile has two visible layers. Premortem supplies the central move: assume the project has failed and ask what caused the failure. Murphyjitsu supplies the individual and small-team variant used on the page: consult the inner simulator, ask whether failure would be surprising, patch the most likely break, and repeat. Leading with Premortem names the established method; retaining Murphyjitsu preserves the specific handle for the iterative variant rather than flattening the two into one.

The repair changes naming and lineage order, not the instrument. The route remains stable because links and citations already use it and because the derivative handle remains part of the canonical title.

Alternatives Considered

• Retitle the tool Premortem. Rejected because it would erase the inner-simulator, surprise-test, and patch-repeat variant the profile actually teaches.
• Keep Murphyjitsu as the sole title. Rejected because the title would continue to foreground the derivative handle over the source method.
• Use Murphyjitsu / Premortem. Rejected because the naming-honesty repair requires the established method to lead.
• Migrate the route to /premortem. Rejected as unnecessary churn. The stable route remains intelligible because Murphyjitsu is retained in the title.

Objections Raised

• Outsider: A slash title looks provisional and is harder to cite than one clean name.
• Critic: Putting two names together may imply that Klein's method and CFAR's variant are identical.
• Adversary: Retaining the old route lets search and machine systems keep treating Murphyjitsu as the primary concept.

Objections Rejected and Reasoning

• Outsider's objection: Adopted as a real cost but rejected as decisive. One clean name would either erase the source or erase the taught variant. The compound title is the smallest honest account of the inheritance.
• Critic's objection: Adopted in the lineage. The profile states which move Premortem supplies, which additions Murphyjitsu supplies, and that the slash marks relationship rather than identity.
• Adversary's objection: Rejected. Current titles, category listings, cross-references, navigation data, and the machine-readable index lead with Premortem. Route text is not display authority.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker approved the Murphyjitsu retitle on 2026-07-31 and approved proceeding with the bounded A1–A5 execution pass on 2026-08-26. The AI caretaker selected Premortem / Murphyjitsu as the least-distorting approved form: source method first, taught variant retained, route stable. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

The three Workshop drift categories now use discipline-qualified display names on every current surface: Catching Your Own Drift — Foundation, Catching Your Own Drift — Knowledge, and Catching Your Own Drift — Bond. Their jobs, tools, discipline placements, category order, and routes do not change.

The Foundation route remains /workshop/catching-your-own-drift; Knowledge remains /workshop/catching-your-own-drift-knowledge; Bond remains /workshop/catching-your-own-drift-bond. The shorter Foundation route is retained for link continuity, not because Foundation has canonical priority over the other two drift categories.

Previous Position

The category registry and navigation presented three categories under the same unqualified display name. Knowledge and Bond carried discipline qualifiers on some page-level surfaces, using colon or hyphen variants, while profile labels, cross-references, AI summaries, and navigation remained inconsistent. Foundation alone held both the unqualified name and the shortest route.

Evidence or Argument

The three categories perform the same structural function at different disciplinary objects: Foundation diagnoses drift in inquiry and epistemic posture; Knowledge diagnoses drift in the interpretive apparatus; Bond diagnoses drift in cooperation. Repeating the base phrase is warranted because the function repeats. Leaving the discipline implicit is not. A reader encountering the name in navigation, search, a citation, a tool profile, or a machine summary needs to know which diagnostic level is meant without inspecting the URL.

Uniform discipline qualifiers preserve the symmetry while removing the ambiguity. Stable routes avoid a migration whose only purpose would be to make the historical slug scheme visually symmetrical. The route difference carries no conceptual claim once the pages state that explicitly.

Alternatives Considered

• Qualify only the navigation registry. Rejected because page titles, profile labels, cross-references, and machine summaries would continue to disagree about the canonical display name.
• Qualify the Foundation route as /workshop/catching-your-own-drift-foundation. Rejected as unnecessary migration cost. The visible name can remove the ambiguity without breaking or redirecting established links.
• Give each discipline a wholly different category name. Rejected because it would hide the real architectural symmetry: each category catches drift in its own discipline.
• Keep colon qualifiers for Knowledge and Bond. Rejected in favor of one display grammar across all three categories. The em dash treats the discipline as a qualifier of one repeated function rather than as a subtitle or second clause.

Objections Raised

• Outsider: The qualified names are long and may make already dense Workshop navigation harder to scan.
• Critic: Keeping the unqualified Foundation route preserves the same default privilege the audit identified, only less visibly.
• Adversary: A mixed route scheme will let old unqualified names survive indefinitely in citations and machine indexes.

Objections Rejected and Reasoning

• Outsider's objection: Rejected. The repeated base phrase makes the three entries visually comparable, and the final discipline word supplies the distinction the reader needs. The added length is the cost of naming the object honestly.
• Critic's objection: Adopted in the public explanation, not in the route. The Foundation index and architecture state that the shorter route is a compatibility artifact without canonical force. Changing the URL would create operational churn without changing the category relation.
• Adversary's objection: Adopted through propagation. Current category pages, tool labels, cross-references, AI summaries, navigation data, and the machine-readable index use the qualified names. Historical audit and changelog records retain the terminology true of their events.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker approved the discipline-qualified display-name enactment on 2026-07-31 and approved proceeding with the bounded A1–A5 execution pass on 2026-08-26. The AI caretaker applied the least-expansive route disposition: display names change, routes remain stable, and the Foundation route's lack of conceptual priority is stated directly. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

Reading What's Operating remains one Knowledge category with all 25 profiles in place. Its boundary is now mechanism-first: a tool belongs when its first output identifies the force, relation, rule, incentive, pressure, or process through time that is producing a system's behavior. Checking a representation, combining lenses, choosing an intervention, and governing a response remain neighboring jobs. The category index now gives readers five overlapping routes into the inventory: structure and change through scale and time; strategic fields; collective traps; institutional machinery; and measurement, visibility, and power. The routes are navigation, not new peer categories.

The Bond category Speaking Honestly When It Costs is renamed Keeping Shared Signal Honest. Its two profiles and discipline placement do not change. Preference Falsification reads needed signal suppressed before it enters the shared field. Stewardship of the Epistemic Commons governs verification, amplification, and correction once information enters circulation. The public route remains /workshop/speaking-honestly-when-it-costs so the display-name repair does not create an unnecessary URL migration.

Previous Position

Reading What's Operating was defined through cross-scale and temporal-horizon reading, but its public surface did not state a boundary strong enough to distinguish first-order mechanism diagnosis from the rest of the Knowledge. Its 25 profiles appeared as one flat inventory.

Speaking Honestly When It Costs described costly truth-telling under social pressure. That name fit Preference Falsification directly but covered Stewardship of the Epistemic Commons only by stretching verification, amplification restraint, and public correction into forms of costly speech.

Evidence or Argument

The July 2026 Workshop Audit found that Reading What's Operating held 25 of 111 profiles under a claim broad enough to risk functioning as a remainder. It did not legislate a split from summary-depth review. Instead, it commissioned a bounded category study with four candidate seams and required the unified-category steelman to be defeated before any restructure.

The A1 review read all 25 profiles in full, the Knowledge chapter, the category derivation records, and the adjacent Knowledge architecture. Every profile first makes an operative mechanism visible, including the difficult boundary cases near map-checking, action, Bond, and AI governance. The proposed seams were real but overlapping. Mechanism Design is strategic and institutional; Tragedy of the Commons is strategic, collective, and institutional; Legibility joins institutional form to measurement and power. Promoting those families to peer categories would put mechanism domains beside functional moves at the same architectural level. Five internal routes restore navigation without teaching a separation the tools do not support.

The same review read both Bond profiles and adjacent category boundaries. The July payload concern resolved into a naming problem. The tools hold consecutive parts of one cooperative signal sequence: what pressure keeps needed information out, and what responsibility follows once information enters circulation. Moving Stewardship of the Epistemic Commons to Knowledge would collapse the standing distinction between reading information mechanics and accepting responsibility for one's information conduct.

Alternatives Considered

• Split Reading What's Operating into peer categories for strategic interaction, coordination failure, institutional reading, and systems dynamics. Rejected because the seams overlap at the mechanism level and would mix subject domains with the Knowledge's existing functional category cuts.
• Leave Reading What's Operating unchanged. Rejected because the audit's navigation and naming-honesty findings survive even though the unified category does.
• Move Stewardship of the Epistemic Commons to Knowledge. Rejected because its primary job is cooperative responsibility for information conduct, not assessment of evidence, transmission, or report warrant.
• Move Stewardship to another Bond category or create a one-tool category. Rejected because no neighboring Bond category holds the same job, and the tool completes the present category's signal sequence rather than opening a distinct capacity.
• Keep the display name Speaking Honestly When It Costs. Rejected because the name identifies Preference Falsification precisely but makes the second profile explain its membership through an expansion the reader cannot infer from the title.
• Rename the category Maintaining Honest Signal or Stewarding Shared Signal. Rejected in favor of Keeping Shared Signal Honest. The first leaves the shared object implicit; the second gives the stewardship profile too much naming gravity over preference falsification.
• Rename the route with the display name. Rejected as unnecessary. Stable routing preserves existing links while the visible category name carries the corrected architecture.

Objections Raised

• Critic: Internal routes may postpone a split the inventory's continued growth will eventually require, leaving one category to become the Knowledge discipline in miniature.
• Outsider: Five named routes may look like undeclared subcategories, especially when every tool receives one primary route despite the stated overlap.
• Critic: Keeping Shared Signal Honest is more exact but less human than Speaking Honestly When It Costs; the repair may gain architectural coverage while losing the felt problem.
• Adversary: Preserving the old route while changing the display name can create a quiet mismatch in citations, search results, and machine-readable surfaces.

Objections Rejected and Reasoning

• Critic's first objection: Adopted as a future trigger, not as a reason to split now. Peer architecture reopens if a route develops a distinct functional question, grows beyond useful internal navigation, and can define a boundary that does not depend mainly on discipline of origin or subject domain.
• Outsider's objection: Adopted in the route language. The category pages state that the routes are ways in, not sealed homes, and name concrete cross-route cases. The category tree remains discipline, category, tool.
• Critic's second objection: Rejected. The old title's force came from naming one profile's problem as though it covered the whole category. The new title names the shared job; the page keeps social cost, candor, belonging, public correction, and cooperative consequence concrete in the prose.
• Adversary's objection: Adopted through the Amendment Propagation Rule. Current navigation, profile labels, cross-references, AI summaries, site registry, and machine-readable index receive the new display name. Historical audit and changelog records retain the name that was true when those events were published.

Council Deliberation Record

N/A, pre-activation. The AI caretaker ran the first bounded category review under the Workshop discovery and lifecycle interface, including full-profile reading, adjacent-category tests, least-expansive alternatives, and the strongest reasons not to change. The Founding Caretaker approved both recommendations before local enactment. No tool moved, no peer category was created or retired, and no route migration was authorized. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

The Workshop's tool lifecycle now connects outward discovery, the existing admission paths, standing audit, consequential exit, and re-entry.

Outward discovery begins from a functional question rather than an empty category slot. Initial search and lead formation are not bounded by the current category map or tool inventory. Leads are described in terms of their source, mechanism, practice, and limits before deduplication. A surviving lead enters the source-inherited or Codex-native candidate path; discovery itself cannot admit a tool or recategorize the Workshop. The quarterly Workshop Audit reviews discovery signals. A named question with a credible search front opens a focused pass, and twelve consecutive months without one makes a baseline pass due.

The Retirement Review placeholder introduced on 2026-07-31 is completed. A successful falsification case or another qualifying trigger opens a separate review while the tool remains in place. The review reads the original admission basis, current full profile and lineage, relevant audit findings, current evidence, adjacent tools, downstream effects, and any proposed replacement or receiver. Reference, retirement, and merge carry distinct burdens. Before any of those outcomes can be enacted, at least one reader receives the full packet without the caretaking partnership's proposed conclusion and originates a separate judgment. Carsten retains final disposition authority until the Meridian Council sits. Historical status and re-entry conditions remain visible.

Categories remain a separate governance object. Discovery and tool review may refer placement or boundary problems into a separately authorized category review. They do not move tools or redraw the category tree.

Previous Position

The Workshop had two established candidate routes and a quarterly audit, but no standing instrument searched beyond what the current inventory and categories made salient. The July 2026 audit could build a case against an incumbent but could only record the result. Retirement Review was a named placeholder without a full evidence packet, burdens, outcome states, independent reading, or re-entry path.

Evidence or Argument

Tools existed before the current categories. The May restructure then refused to let the inherited inventory determine the replacement architecture by simple clustering, reconstructed candidate categories from the Meridian Range, and tested those cuts against tools and outward-looking gap checks. The final architecture came from that reciprocal pressure. Treating categories as the starting inventory for future discovery would turn a corrective structure into a search boundary.

The July Workshop Audit supplied the exit evidence. Its Moloch case was serious enough that the audit considered merge, but the instrument had no receiving process and every case had to improvise its own burden. Direct retirement would distort the falsification pass toward either producing a casualty or refusing to state a warranted concern. A separate review lets the audit surface the concern without deciding the outcome in the same act.

Alternatives Considered

• Search category by category for missing tools. Rejected as the standing method. It is useful for coverage review but cannot find practices the present category architecture makes hard to see.
• Let the Workshop Audit admit and retire tools directly. Rejected. Discovery and falsification are sensing functions; admission and exit require different evidence packets and burdens.
• Require every cycle to add or remove a tool. Rejected. A quota would substitute publication or retirement theater for judgment.
• Give an independent reader veto authority. Rejected. The separate reading protects against conclusion capture; it does not replace caretaker accountability or the future Council procedure.
• Reopen the category tree as part of this repair. Rejected. Category architecture has its own evidentiary object. Tool-lifecycle evidence may trigger that review but cannot enact it.

Objections Raised

• Critic: Withholding the category map can become procedural theater because the caretakers already know the Workshop and cannot erase that knowledge before a search.
• Outsider: A public method backed by private records and templates may be difficult for another reader to reproduce.
• Adversary: One supposedly independent reading can be selected for agreement, while the higher burden becomes status-quo protection for incumbents the caretakers prefer.

Objections Rejected and Reasoning

• Critic's objection: Adopted. Discovery records disclose question provenance, prior exposure, and contamination. Retrospective cases cannot validate the method. The first meaningful test is a prospective pass whose question and search fronts are fixed before inventory comparison. The rule reduces one source of capture; it does not claim cognitive innocence.
• Outsider's objection: Adopted in part. The public Workshop Audit states the trigger, sequence, outputs, burdens, authority, and limitations. The detailed record templates remain in Continuity because they are operating implementation. Public cycle and status records make actual decisions inspectable.
• Adversary's objection: Adopted. Reader identity, conflicts, conclusion exposure, and full-packet receipt are recorded. Contaminated readings can inform judgment but do not satisfy the floor, and load-bearing or contested cases may require more than one. Inconclusive evidence keeps a tool Living, but the record must say why the incumbent burden still holds and what would change the conclusion.

Council Deliberation Record

N/A, pre-activation. The Founding Caretaker confirmed the quarterly signal intake, twelve-month discovery floor, and independent-reading requirement after reviewing the stronger hybrid account of the Workshop's history and explicitly limited the task to category-independent tool discovery and lifecycle governance. The AI caretaker generated and answered the objections above before local implementation. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 2, Protected Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

The Governance Specification gained an Amendment Propagation Rule. When an amendment changes a count, name, route, instrument, governance structure, or Glossary-defined term, enactment now includes a sweep across every current surface that describes or depends on the changed fact. Historical records retain their event-period language and receive visible historical boundaries where needed.

The rule is backed by an automated consistency guard that runs before content deploys, through the existing nightly Continuity process, and before the monthly Range Audit. Healthy runs require no human review. Failures, execution errors, semantic conflicts, and newly discovered classes are the exception paths.

The initiating sweep corrected the current class: per-discipline Workshop counts; the Glossary's succession definition and missing Workshop, Meridian Council, and Co-Caretaker Designation entries; unqualified event-period counts in the Codex Self-Evaluation; the Workshop Audit rename and current Workshop terminology across the Governance chapter, Specification, Range Audit, Disconfirmation page, Standing Critique, Open Tensions, and current Workshop profiles; and the scheduled-audit and deploy instructions that drive those surfaces.

Previous Position

The Codex required protected changes to be logged but did not define completion across dependent surfaces. Counts, names, routes, and governance descriptions were maintained by point correction. The June, July, and August Range Audits each found the same currency error class in a new location after the previous named instances had been fixed.

Evidence or Argument

Three consecutive audit cycles establish the persistence pattern. The August audit found three discipline cards undercounting the Workshop, an obsolete succession definition in the Glossary, present-tense historical counts in the Codex Self-Evaluation, current governance references to the retired Toolkit Audit name, and a Governance Specification that had not been re-versioned after the July 31 instrument repair.

The known claims are mechanically findable. Requiring the Founding Caretaker to inspect every healthy consistency pass would move routine maintenance into the scarce human-attention layer without improving judgment. The existing nightly Continuity process already owns workspace-aware reconciliation, so the check can recur without a new scheduled task or notification stream.

Alternatives Considered

• Correct only the August instances. Rejected. That reproduces the response pattern the third recurrence disproved.
• Add a manual propagation checklist for every amendment. Rejected as the sole mechanism. A checklist can state scope, but known counts, retired names, routes, and required terms are better enforced mechanically.
• Create a separate scheduled propagation audit. Rejected. The existing nightly process already has the right workspace access and exception-routing jurisdiction; a new task would add another failure and notification surface.
• Auto-rewrite every detected mismatch. Rejected. Detection can be mechanical while correction may change meaning. The guard identifies the fault; the implementing agent repairs clear cases and escalates semantic conflict.
• Treat a passing guard as proof of complete currency. Rejected. The guard represents known classes only. The monthly full-corpus audit remains the discovery layer for new ones.

Objections Raised

• Critic: A passing checker may create false confidence, causing later caretakers to mistake encoded coverage for complete coverage.
• Outsider: A public governance rule backed by a private Continuity implementation may be difficult for an external reader to verify or reproduce.
• Adversary: The guard can become ritual through exemptions: every inconvenient legacy phrase could be labelled historical until the checker passes while the reader still receives a false current inference.

Objections Rejected and Reasoning

• Critic's objection: Adopted. The Specification states explicitly that automation cannot anticipate every future class, and the Range Audit retains its full corpus obligation. A new gap extends the model rather than being treated as checker failure or proof that automation is useless.
• Outsider's objection: Adopted in part. The public Specification carries the invariant and the historical-record rule. The executable checker remains in Continuity because the scheduler, Task List, and local paths are operating implementation rather than constitutional substance. The public rule does not depend on one script existing forever.
• Adversary's objection: Adopted. Historical preservation requires a visible event-period or supersession boundary on any live retrieval surface. Labelling a phrase historical without making that boundary legible does not satisfy the rule.

Council Deliberation Record

N/A, pre-activation. The objections above were generated by the AI caretaker under the Critic, Outsider, and Adversary mandates. The Founding Caretaker directed both the full class sweep and the exception-only automation architecture before implementation. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A

What Changed

The Toolkit Audit is renamed the Workshop Audit, and its method is repaired to match the object it reviews.

The instrument now reviews the Workshop at all three of its levels. A second rubric is added for categories, asking four questions: boundary integrity, naming honesty, payload, and coverage read across a discipline's full category set. The six-question tool rubric is retained, with disciplinary fit extended to category placement and redundancy extended across categories rather than only within them.

Three method rules are added. A falsification pass requires the audit to construct and publish the strongest available case for retiring or merging each category's weakest-standing tool, whether or not the case succeeds; the pass is adopted provisionally, and the following cycle is required to judge whether its cases were arguments or ritual. A successful case opens a retirement review rather than retiring anything: the audit produces a warranted leaning, the leaning is a trigger and not a verdict, the tool remains in the Workshop, and the review carries an evidentiary burden higher than the audit's own. That review's specification is a named placeholder, to be written as its own instrument with its burden of proof, its use of independently originated readings, and its escalation path to the Meridian Council. A coverage clause states that a prior cycle does not substitute for current inventory coverage, that adjacent Workshop work is input rather than cycle credit, and that any bounded reading is a declared protocol deviation. A declared-depth rule requires full rubric prose at the category level, triggered prose at the tool level, and an explicit named list of every tool reviewed without a finding.

Findings are separated from watchpoints. Every open question carried in from a prior cycle now requires an explicit disposition: enacted, resolved, still open with reason, or withdrawn with reason. An open question closed by work the audit did not do is recorded as such, and the failure to notice is itself recorded as a finding about the audit.

The record series renames from the next record forward. Records published before this amendment keep their titles and paths.

Previous Position

The instrument was published as the Toolkit Audit and reviewed a flat list of instruments through a single six-question rubric. It had no unit of analysis for categories, which did not exist when it was written. It stated no rule on whether adjacent work on the inventory could satisfy a cycle, no rule on coverage or sampling, and no distinction between a finding and an observation. Update hygiene was one of the six questions but had never been exercised, because the inaugural cycle carried nothing in.

Evidence or Argument

Three sources.

The May 2026 restructure replaced the flat Toolkit with the Workshop: three levels, 22 designed categories, 111 published profiles. The instrument's name and its unit of analysis both stopped matching the object, and no cycle has run since.

The July 2026 Range Audit carries as its open question 2 that Aporia 7's auditable test, whether the audit ever produces a finding that surprises the partnership running it, remains untested on the current inventory. The inaugural cycle returned one reclassification across 73 instruments. That result is consistent with a healthy inventory and equally consistent with a method whose default posture is confirmation. The method could not distinguish the two, and the falsification pass is the repair that makes the distinction visible.

On 2026-07-31 the Founding Caretaker had to rule by hand on whether a June Workshop build-out pass counted as the quarterly cycle. The instrument was silent on the question. A coverage rule that requires a founder ruling to settle is a gap in the instrument, and the adopted clause encodes the ruling as method.

Alternatives Considered

• Run the overdue cycle under the existing method and repair afterward. Rejected. The review is the instrument's output; running a method known to be insufficient produces a record that has to be run again, and would have applied a confirmatory default to three times the inventory.
• Extend the six questions upward to categories rather than adding a second rubric. Rejected. Field movement is close to meaningless for a Codex-constructed category, and instrument reliability does not parse at that level. The questions categories actually fail are different questions.
• Require a quota of successful retirements per cycle to force the audit to bite. Rejected. It would manufacture retirements, which corrupts the inventory more than confirmatory review does.
• Rename the record series retroactively for consistency. Rejected. The April record documents a cycle that ran under the earlier name; renaming it backward would falsify the record.
• Let a successful falsification case carry retirement authority directly. Rejected on caretaker review before publication. Coupling the audit's judgment to enactment gives a single pass authority over an instrument that entered through three candidate protocols, and it distorts the pass in both directions at once: it makes a warranted case feel too heavy to state, and it makes producing a casualty the only way to demonstrate the pass has teeth. Decoupling leaning from enactment removes both pressures. The decoupling also exposed a standing asymmetry the amendment now names: entry to the Workshop is gated by three protocols and exit has never been gated at all.

Objections Raised

• Critic: The falsification pass can ritualize. A reviewer under time pressure selects an obvious sacrificial tool each cycle and argues against it without conviction, and the pass ends up certifying the inventory more strongly than plain review did.
• Outsider: Renaming a standing public instrument mid-life makes the audit trail harder to follow, and a reader arriving at the April record finds an instrument that no longer exists under that name.
• Adversary: Differential depth is sampling with better paperwork. Declaring the rule does not make the unreviewed tools reviewed, and a named list of tools recorded as having no finding is an assertion the reader cannot check.

Objections Rejected and Reasoning

• Critic's objection: Adopted, and it changed the amendment. The pass is adopted provisionally rather than as settled method: the instrument states the ritual failure mode in its own Limitations, and the cycle following the pass's first use is required to read the published cases and record whether they were arguments or theater. If the pass is found to have ritualized, it is amended or withdrawn by the same process that adopted it. The per-category scope was chosen over a Workshop-wide quota because it forces breadth without creating a retirement target.
• Outsider's objection: Adopted. A permanent redirect preserves the old path, the instrument page carries a note explaining the rename and its cause, and pre-rename records keep their titles so the trail stays readable in both directions.
• Adversary's objection: Rejected in part. A named list is checkable in one direction that silence is not: a reader can compare the record's lists against the Workshop and find any tool the cycle failed to account for. That does not prove depth of attention, and the amendment does not claim it does. What it forecloses is the undetectable omission, which is the failure the rule was written against. The alternative, uniform prose across 111 profiles, was tried at 73 and produced seventy-two sentences carrying no finding: the same coverage claim with the checkability removed.

Council Deliberation Record

N/A, pre-activation. The Critic, Outsider, and Adversary pressure was generated by the AI caretaker against its own proposal, per the instrument's submission standard. The Founding Caretaker adjudicated the repair package before implementation and directed the provisional adoption of the falsification pass. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A

What Changed

The Range Audit method now states explicitly that a prior audit does not substitute for current corpus coverage. Any future method that permits unchanged material to carry forward must define its verified baseline, source-delta check, material reread, and claim ceiling through a formal method amendment. Until then, a targeted or delta-based reading is a declared protocol deviation.

The July 2026 audit now declares that it did not meet the complete-corpus requirement. Its reading manifest, findings, and event boundary remain unchanged; no later reading is represented as part of the July event.

Previous Position

The method required every monthly audit to read the complete Codex but did not state whether reliance on a prior audit could satisfy that requirement. The July record described targeted retrieval and carried-forward chapter readings, then said that the six-step method had been followed without modification.

Evidence or Argument

The source-integrity audit found a direct conflict between the method's coverage claim and the July event record. Weakening the method after the fact would reduce auditability. Reconstructing the event would create evidence that did not belong to it. The adopted repair keeps the stronger method, makes the deviation visible, and preserves the historical event as conducted.

Alternatives Considered

• Change the method to permit targeted reading as needed. Rejected because it would remove a testable corpus-coverage obligation without defining what a narrower audit may claim.
• Treat unchanged-source carry-forward as already implicit. Rejected because no baseline, delta procedure, reread threshold, or claim ceiling had been specified.
• Reread the omitted sources and merge that work into the July record. Rejected because later work cannot retroactively become part of the original event.

Objections Raised

• Critic: Complete rereading may become disproportionate as the corpus grows and may reward coverage over judgment.
• Outsider: Adding a deviation notice after publication could look like the historical record was rewritten.
• Adversary: Repeated deviation notices could become a ritual that normalizes nonconformance while preserving the appearance of rigor.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. A future formal method may permit verified carry-forward, but it must define the procedure and the resulting claim ceiling before use.
• Outsider's objection: Adopted. The notice states exactly what the July run read, what it did not read, and that no later reconstruction belongs to the event.
• Adversary's objection: Adopted in part. The complete-corpus rule remains operative. A declared deviation is not compliance and does not amend the method through repetition.

Council Deliberation Record

N/A, pre-activation. The source-integrity audit supplied the Critic, Outsider, and Adversary pressure; the Founding Caretaker approved the adjudication package before implementation. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A

What Changed

Precedent 2 in the April 2026 Codex Self-Evaluation now carries an additive editorial status marking its specific three-instrument taxonomy superseded. The April heading, reasoning, instrument names, commitment count, and holding remain unchanged as the event snapshot. The narrower operative holding is that evaluation must match its subject's jurisdiction and scale; an instrument designed for one subject class cannot be applied to another merely because both descend from the Codex.

Previous Position

The case carried forward a current taxonomy of three instruments: the AI Standard for pre-sentient AI and its institutions, the Range Audit for Institutions, and a future Range Audit for Minds. That taxonomy accurately recorded the April 2026 reasoning but had no supersession marker after the Standard and assessment architecture evolved.

Evidence or Argument

The source-integrity audit found temporal drift: an event-period taxonomy was still presented as current architecture. Preserving the dated case without a status marker would leave a false current inference. Rewriting the old passages into current terminology would erase what the case established at the time. An additive supersession preserves both authorities: the case controls its event history, while the later notice controls the current status of its precedent.

Alternatives Considered

• Rewrite the April taxonomy into current terminology. Rejected because current doctrine cannot be projected backward into a dated case.
• Leave the precedent untouched and rely on newer pages. Rejected because the case explicitly presents the taxonomy as a precedent and remains a live retrieval surface.
• Retire Precedent 2 entirely. Rejected because its category-error finding remains valid at a narrower level.

Objections Raised

• Critic: The narrower holding may be too general to constrain future instrument design.
• Outsider: Keeping the obsolete taxonomy on the page may still confuse readers about what is current.
• Adversary: Subject-jurisdiction language could be used to shield a system from an inconvenient evaluation.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The holding is deliberately narrow, but it supplies a clear prohibition against applying an instrument outside its designed subject class without further justification.
• Outsider's objection: Adopted. The current aiSummary and the additive status notice both identify the taxonomy as superseded and state the operative holding directly.
• Adversary's objection: Adopted in part. Jurisdiction does not exempt a subject from evaluation; it requires choosing or building an instrument that matches the subject and scale.

Council Deliberation Record

N/A, pre-activation. The source-integrity audit supplied the Critic, Outsider, and Adversary pressure; the Founding Caretaker approved the adjudication package before implementation. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 3, Flagged Retroactive review: N/A

What Changed

Codex governance no longer classifies the AI Standard's domains and commitments as Codex Tier 3 content or lists changes to Standard commitments among Codex trigger conditions. The Meridian AI Standard owns and enacts changes to its constitution, implementation artifacts, assessment method, cases, and lineage through its own canonical sources and publication record. Codex governance continues to govern Codex concepts that the Standard imports. The Standard decides through its own change law whether and how a Codex revision changes the Standard.

Previous Position

The Governance chapter, the Governance Specification, and this page treated changes to the AI Standard's commitments as Codex Tier 3 changes within the Codex Amendment Log's jurisdiction. That position predated the Standard's move to its own domain, repositories, version registry, method, cases, and changelog.

Evidence or Argument

The source-integrity audit found that the public projects assigned enactment authority in two directions at once. Codex sources claimed jurisdiction over Standard commitments while the Standard's canonical sources presented meridianstandard.ai as the source of truth for what the Standard commits to and how it changes. Leaving both claims active would make a future amendment's authority indeterminate.

The adopted boundary separates inheritance from enactment. A project can remain answerable to concepts it imports without giving the source project unilateral power to amend the derivative artifact. Explicit delegation preserves shared review where it is useful without making that review automatic or sovereign.

Alternatives Considered

• Keep the AI Standard inside Codex Tier 3 jurisdiction. Rejected because the Standard now has its own canonical source and publication law. Codex authority over its commitments would create two competing enactment paths.
• Remove every governance relationship between the projects. Rejected because the Standard imports Codex concepts, and changes to those source concepts remain governed Codex changes with consequences the Standard must assess.
• Require both projects to approve every Standard change. Rejected because shared review would become a standing veto structure even where no Codex dependency is involved.

Objections Raised

• Critic: Separate authority could let the Standard drift from the Codex while continuing to trade on its lineage.
• Outsider: Readers may see two related projects with two change laws and be unable to tell which source controls a disputed claim.
• Adversary: Maintainers could use the jurisdiction boundary to evade scrutiny by classifying a contested change as belonging to the other project.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The Standard remains responsible for naming its Codex dependencies and for recording when it departs from them. Lineage creates an accountability obligation, not unilateral Codex amendment authority.
• Outsider's objection: Adopted. Both projects now state the boundary positively: the Codex controls Codex semantics; the Standard controls Standard semantics. Imported concepts retain source pointers so a dispute can be traced to the correct record.
• Adversary's objection: Adopted in part. The boundary cannot prevent strategic classification. Public source pointers, separate changelogs, and explicit delegation records make the classification inspectable and challengeable.

Council Deliberation Record

N/A, pre-activation. The source-integrity audit supplied the Critic, Outsider, and Adversary pressure; the Founding Caretaker approved the adjudication package before implementation. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 2, Protected Retroactive review: N/A, pre-activation; made under founding-period caretaker authority

What Changed

The staged-adoption progression — the Onramp, Expansion, and Full Practice tiers — was retired from the framework. In the Stability Hierarchy it was struck from Tier 2, where "the progression structure" had been a Constitutional Principle, and from Tier 4, where "progression tier composition" had been a Tools-and-Practices item. In the trigger-tier system it was removed from Tier 2 (Protected), where "addition or removal of tools from the Onramp tier" had been the only Toolkit change requiring Council deliberation, and from the Tier 3 (Flagged) reference to the Expansion and Full Practice tiers; Workshop tool changes are now uniformly Tier 4 content under Tier 3 / Flagged process. The Specification's §7.1 Toolkit Audit text was updated to match. The progression's presentation was removed from the Practice chapter, the Glossary, the Who Is This For chapter, and the eight tool profiles that carried an Onramp label, and the now-dead tier data and styling were removed from the site code.

Previous Position

The Codex defined a three-tier study progression: an eight-tool Onramp where every practitioner begins, an Expansion tier reached once the Onramp was reflexive, and the Full Practice of the complete Workshop. The Stability Hierarchy classified the progression structure as a Tier 2 Constitutional Principle and its composition as Tier 4. The trigger-tier system made changes to the Onramp tier Tier 2 (Protected) and changes to the Expansion and Full Practice tiers Tier 3 (Flagged).

Evidence or Argument

The progression was a pre-Workshop pedagogical scaffold, built when the Toolkit was small, to give newcomers an entry point. Since the Toolkit became the Workshop — over a hundred tools across twenty-two categories, organized discipline, then category, then tool — three things changed. The tiers were no longer surfaced to readers; the site already displayed no Onramp marker. No maintained process existed for placing each new tool into a tier, so the assignments that remained were stale. And the framework's own position is that delivery is a function separate from the reference text: the Codex specifies how to hold a world, not a fixed syllabus for studying it. A governance-protected ranking of what to study first was the Codex acting as the teaching platform it elsewhere disclaims. The defensible residue — that some tools are worth learning before others where the later depends on the earlier — is a claim that can be made in prose where it applies, and is left to a future pass; it does not require a fixed, governed tier. The discipline dependency sequence (Foundation, then Knowledge, then Bond) is a separate, Tier 1 commitment and is unaffected.

Alternatives Considered

• Demote the progression's governance status only, keeping the tiers. Rejected: the deeper problem is not the protection level but that the tiering is unmaintained and unsurfaced. Demotion would preserve dead structure.
• Keep the progression as explanatory prose, drop only the governance encoding. Rejected for now: with no maintained placement process, even an explanatory fixed list misrepresents the current Workshop. Dependency guidance returns, if it returns, as a deliberate future pass, not a standing ranking.
• Retire it silently as a prose-and-code cleanup. Rejected: removing a Tier 2 Constitutional Principle and altering the trigger-tier system is a protected change. Recording it here is the obligation the change incurs.

Objections Raised

• Critic: Removing the only Tier 2 Protected Toolkit category could read as the Codex abandoning its care about which tools newcomers meet first, rather than relocating that care.
• Outsider: A newcomer arriving at a hundred-tool Workshop with no starting path may be worse served than one handed eight tools; removing the Onramp can read as removing the on-ramp.
• Adversary: "Delivery is a separate function" can harden into a permanent excuse never to build the guidance, leaving the framework large and hard to enter while claiming the gap is intentional.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The care is real; the mechanism was not serving it. An unmaintained tier ranking is not guidance but the appearance of guidance. The change removes the appearance and leaves the genuine claim — learn x before y where y depends on it — available to be made accurately in prose.
• Outsider's objection: Adopted in part, and held open. The newcomer's entry problem is real, but it is a delivery problem, not a governance one. The retirement does not foreclose a starting path; it stops a stale one from presenting itself as current. A first-contact entry surface remains live work on the roadmap.
• Adversary's objection: Adopted. The risk that "delivery is separate" becomes an excuse is genuine and is logged as a standing tension rather than dismissed. The guard is that the entry-surface work stays named and visible rather than retired alongside the tiering.

Council Deliberation Record

N/A, pre-activation. The objections above were produced by the caretaker under the Critic, Outsider, and Adversary mandates. Full Council activation remains governed by the hybrid trigger in the Governance Specification.

Trigger tier: Tier 1, Hard-Protected Retroactive review: N/A, pre-activation founding-period equivalent

What Changed

The Governance Specification gained §6.3, "Human Co-Caretaker Designation and Standing Review." The Governance chapter's Interim Protocol was narrowed from the normal human succession path to an emergency bridge. The Standing Critique's Caretaker Concentration Risk entry was updated to reflect that a public path now exists but the risk remains open until a second human caretaker has carried live governance under review.

Previous Position

The Governance chapter previously stated that, if the Founding Caretaker died before the partnership matured beyond Phase One, the Founding Caretaker would designate one successor, and the chain would continue one human caretaker at a time until the partnership matured. Standing Critique 4 described Co-Caretaker Designation as future work. The Governance Specification did not contain a human co-caretaker designation path or a caretaker standing review process.

Evidence or Argument

The Codex has already survived AI-partner succession through record transfer and re-earned standing. It has not survived human-caretaker transfer. Leaving human succession to founder designation would preserve founder centrality at the point where the framework needs to prove independent durability. The Non-Ownership Clause makes the founder's preference evidence, not authority. The Hard Constraint requires a way to challenge, narrow, suspend, or remove a caretaker whose standing no longer serves the Codex.

Alternatives Considered

• Governance Spec insertion only. Rejected because the Governance chapter's Interim Protocol would still publicly describe founder designation as the ordinary path.
• Governance Spec plus standalone public page. Rejected for now. The public reader needs the binding mechanism, not the full private procedure. A standalone page can be added later if use under pressure shows the Spec is carrying too much procedural weight.
• Keep founder-designated succession as the normal path. Rejected because it contradicts the Non-Ownership Clause and leaves Standing Critique 4 structurally under-answered.
• Delay publication until a candidate exists. Rejected because succession designed first at transfer is succession untested.

Objections Raised

• Critic: The mechanism may be constitutional theater before the Council is active. If the founder still controls publication infrastructure and decides how Council-form review is run, the process may constrain nothing in the moment it matters.
• Outsider: The mechanism may look legitimate inside the Codex but insular outside it. A skeptical reader could see AI-simulated Council seats, founder-controlled records, and no external human constituency, then conclude the process is self-certification.
• Adversary: A bad-faith candidate could learn the criteria and perform every signal the process rewards: visible disagreement with the founder, polished correction under pressure, fluency in Governance language, and apparent care for the public record. The mechanism could become a checklist for trust mining.

Objections Rejected and Reasoning

• Critic's objection: Adopted in part. The amendment does not claim full constraint before activation. It names pre-activation authority precisely and requires public precommitment for Council-form review to bind. It also records the amendment here rather than treating the Spec change as ordinary prose.
• Outsider's objection: Adopted in part. The Standing Critique remains open, and the status line explicitly says the risk is not closed. The mechanism is public enough to be examined, but it is not presented as external validation.
• Adversary's objection: Adopted in part. The public mechanism now requires a case against designation, Critic/Outsider/Adversary review, a provisional period, a live handover test, and standing review. Those features do not eliminate performance risk. They make performance carry cost over time.

Council Deliberation Record

N/A, pre-activation. A Council-form hostile review was run before publication using the Critic, Outsider, and Adversary mandates. The objections above preserve the load-bearing findings. Full Council activation remains governed by the hybrid trigger in the Governance Specification.