Authoring Protocol v1

The method of composition is part of the doctrine.

This protocol was written for the StateCraft canon but governs all content published on this site. It establishes the standard under which human authorship coexists with AI assistance without displacement.

The canon shall not be produced through the same collapse it exists to oppose. A body of work that argues for the disciplined separation of fact, belief, state, hypothesis, interpretation, and explanation must itself be written under a discipline that preserves those distinctions. The method of composition is therefore part of the doctrine. This protocol establishes that method. It extends to authorship the same order the canon demands of reasoning: canonical admission, explicit uncertainty, traceable derivation, bounded assistance, and visible revision.

Article I — Origin

The origin of doctrine is human.

The originating intuition, distinction, tension, analogy, or proposition belongs first to the author. No generated formulation shall be treated as origin merely because it is fluent.

Article II — Layers

The canon shall be produced in ordered layers.

1. Observation — raw author material.
2. Clarification — extraction, comparison, disambiguation, and structural assistance.
3. Ratification — explicit author assent, rejection, revision, or suspension.
4. Rendering — narrative expression of what has already been ratified.
5. Integration — placement of the rendered text into the wider canon, with dependencies and open questions preserved.

No later layer may silently replace an earlier one.

Article III — Status

No statement shall become canonical merely by appearing in prose.

Every proposition must occupy an explicit status:

• observation
• candidate claim
• ratified claim
• derived claim
• open question
• rhetorical rendering

Status confusion is prohibited. Canonical growth occurs only through ratification and explicit derivation.

Article IV — Sovereignty

The author is sovereign over canonical assent. AI is not.

AI may assist in:

• extraction
• arrangement
• contrast
• reformulation
• opposition
• rendering

AI may not:

• determine doctrine
• settle ambiguity on its own authority
• convert uncertainty into closure
• smuggle bridges into canon without disclosure
• author canonical truth

Article V — Ratification

A proposition becomes canonical only when the author explicitly ratifies it.

Ratification concerns the claim, not merely the sentence. To ratify a sentence because it "sounds right" is insufficient. The question is not whether the prose is elegant, but whether the proposition is truly owned, intended, and defensible by the author.

Article VI — Provenance

Every canonical text must preserve provenance.

At minimum, each text must remain recoverable in relation to:

• the originating observations from which it arose
• the candidate claims it considered
• the claims it ratified
• the prior texts it depends on
• the questions it leaves unresolved

A canon without provenance becomes a collection of conclusions detached from their formation. That is forbidden.

Article VII — Style

Style is subordinate to doctrine.

No prose may: sound stronger than the thought it bears, conceal an unresolved leap, counterfeit assent through eloquence, suppress uncertainty for smoothness.

Where style and doctrinal precision conflict, doctrinal precision governs.

Article VIII — Uncertainty

Open questions shall remain visible until resolved.

If a concept is unstable, it must be marked unstable. If a proposition is plausible but unratified, it must remain provisional. If a bridge is inferred rather than authored, it must be marked as such. False closure is a defect of the canon.

Article IX — Derivation

Texts in the canon must grow compositionally.

A text may define. A text may diagnose. A text may prescribe. A text may formalize. A text may instantiate. A text may apply. But no text should attempt total doctrine without explicitly being a synthesis text. The canon advances by ordered dependence, not by undisciplined accumulation.

Article X — Replayability

Every serious conclusion in the canon should be replayable.

The path from raw observation to canonical proposition to rendered text must remain inspectable. What cannot be replayed cannot be trusted as disciplined authorship.

Article XI — Revision

Revision must be additive, explicit, and historically legible.

Canonical texts may be amended, superseded, or refined. They may not drift through silent substitution. A later version must stand in visible relation to the earlier one.

Article XII — Function

The function of this protocol is not to purify authorship from all assistance. Its function is to preserve authorship from displacement.

Originality, under this protocol, means:

• origin remains human
• propositions are ratified by the author
• uncertainty remains explicit
• prose remains downstream
• assistance remains bounded
• doctrinal formation remains traceable

Closing Rule

The canon shall be written under order.

• Origin shall be human.
• Assistance shall be bounded.
• Ratification shall be explicit.
• Provenance shall be preserved.
• Uncertainty shall remain visible.
• Revision shall be legible.
• Style shall remain subordinate.
• No proposition shall become canonical without assent.

Before the canon speaks about disciplined reasoning, it shall write under it.

Author's Declaration

This work follows a bounded AI-assistance protocol. The human author originated the thesis, selected the arguments, and retained sole authority over judgment, validation, and final formulation. Generative AI was used only for thesis rendering: organizing, rephrasing, tightening, and narratively expressing ideas already supplied or explicitly approved by the author. It was not used as a sovereign source of truth, not permitted to author canonical claims, and not recognized as a co-author.