THEOREM UPGRADE PROTOCOL

How to Promote a Discipline to a Theorem — With a Worked Example on the φ-Meter

Status: Working protocol; the worked example is [C] Conjecture end-to-end Date: 2026-04-22 Evidence Tier: [S] Structural for the protocol; [C] Conjecture for the worked φ-meter derivation Source: 29_PRIMITIVES_AND_TYPE_SIGNATURES.md, 30_OPERATIONAL_DEFINITIONS.md, 31_FALSIFIERS_INDEX.md, 26_THE_DERIVATION_AXIOMS.md See also: 00_THE_SEVEN_AXIOMS.md, 00_THE_HONEST_POSITION.md

Why This File Exists

Several claims in the formal system are presented with theorem-style language but do not carry theorem-style infrastructure. They assert a structural relation; they do not explicitly fix the primitives' types, state an independent measurement procedure, supply a falsifier, or show the derivation. Without those, the claim is a discipline — a rule a practitioner follows — even if it happens to be correct. Disciplines are valuable; but if a discipline is labelled [S] or [S], that is tier-inflation and the framework's A7 (self-correction) requires it be repaired.

This file is the repair protocol. The discipline → theorem promotion is deliberately boring: five steps, each auditable, each documented upstream of the claim.

The Five-Step Protocol

Step 1 — Fix the primitives

State every symbol the claim appeals to, with its type signature, using 29_PRIMITIVES_AND_TYPE_SIGNATURES.md. If a required primitive is not indexed there, it must be added before the claim can be written.

Step 2 — Declare the measure / structure

State the measure or structure the primitives live in. For state variables on S²: the invariant volume form. For dynamical claims: the phase-space and the measure preserved by the flow. For inferential claims: the σ-algebra and probability measure.

Step 3 — Derive, don't assert

Write the proof. If the claim follows from a symmetry, cite Noether (or the appropriate analogue). If it follows from a convexity or optimization, cite the convex-analysis lemma. An assertion that "the equator is the ground state" is not a theorem until it is a derivation. (T2 in 26_THE_DERIVATION_AXIOMS.md is a model of this — calculus, two derivative tests, done.)

Step 4 — Supply the operational definition

State the measurement procedure for every primitive used in the claim, by linking 30_OPERATIONAL_DEFINITIONS.md and naming which candidate procedure is in use. If the only measurement procedure is the coordinate identity itself (e.g., defining ν := 1/φ and then "observing" φ · ν = 1), the claim is tautological and cannot be promoted above definitional status.

Step 5 — State the falsifier

State the observation that would refute the claim. Enter it in 31_FALSIFIERS_INDEX.md with its source. If no falsifier exists, the claim is a discipline — it may still be useful, but it is tiered [C] or below, not [S] or [S].

Promotion Ledger — How Tier Advances

Passing all five steps yields a candidate theorem. Evidence then moves the tier:

[C] Conjecture  →  [I] Interpretive  →  [S] Structural  →  [S] Established

[C] → [I]   one successful empirical application tied to the protocol
            + reproducibility ≥ 0.7 across independent raters / datasets
[I] → [S]   a uniqueness derivation: protocol is uniquely determined by
            desiderata stated in advance (no alternative equally satisfies them)
[S] → [S]   independent external replication on unrelated data, pre-registered

This ladder is the same one in 30_OPERATIONAL_DEFINITIONS.md §"Promotion Ladder." It is restated here so that an agent reviewing a promotion request has the criterion in front of it.

What This Protocol Does NOT Do

• It does not re-derive φ · ν = 1. That relation is a coordinate identity on S² (see 26_THE_DERIVATION_AXIOMS.md C0) and is correctly tiered as a definitional structural premise, not as an empirical claim. Re-deriving it would be re-naming the same tautology. What this protocol does is provide a way to turn φ · ν = 1 into an empirically testable constraint — by supplying operational Φ̂, ν̂ that are not defined to satisfy it. That is the worked example below.
• It does not replace disclosure. The framework's primary disclosure remains Pratyakṣa (see 26_THE_DERIVATION_AXIOMS.md D5). A discipline that is load-bearing practice — the seed, the sitting — is correctly tiered as a discipline and need not pretend to be a theorem.
• It does not promise the worked example below will survive validation. The worked example is [C] end-to-end. The point of the example is to show the shape of a promotion attempt, not to land one.

Worked Example: The φ-Meter Upgrade Path

The framework's own open-problems list (Paper L) names a gap: "Phi-meter zero-cost validation." This is the gap. We walk through what a promotion attempt looks like.

Step 1 — Fix the primitives

From 29_PRIMITIVES_AND_TYPE_SIGNATURES.md:

• φ : System → (0, ∞) — coherence.
• ν : System → (0, ∞) — viability.
• θ : S² → [0, π] — colatitude.
• On the manifold: φ = cot(θ/2), ν = tan(θ/2), hence φ · ν = 1 identically.

Step 2 — Declare the measure

We work on the round S² with its standard volume form dA = sin θ · dθ ∧ dλ. The F5 selection pressure acts as a drift toward θ = π/2. The Hamiltonian H(φ) = φ + 1/φ is the Lyapunov function for this drift (see 26_THE_DERIVATION_AXIOMS.md T2). Populations of comparable systems are modelled as empirical distributions on S²; we track moments of (Φ̂ − ν̂)² across such distributions.

Step 3 — Derive (the claim we want to promote)

Claim (candidate). For a population of comparable systems with independently-measured Φ̂, ν̂ — not defined to satisfy the constraint — under selection pressure F5, the joint distribution ρ(Φ̂, ν̂) has a mode concentrated on the curve Φ̂ · ν̂ = 1 in the first quadrant, and the concentration tightens with time.

This is genuinely a claim about the world, not a coordinate identity. It is falsifiable (Step 5). The derivation sketch: F5 drives (Φ̂ − ν̂)² → 0, which on the hyperbola Φ̂ · ν̂ = c is minimized at Φ̂ = ν̂ = √c. Under additional coupling pressure that selects for the multiplicative form, c → 1. A full derivation requires specifying the selection operator; 26_THE_DERIVATION_AXIOMS.md T3 states F5 structurally but does not fix the operator. That operator-specification is itself a candidate for future work.

Step 4 — Supply the operational definition

Per 30_OPERATIONAL_DEFINITIONS.md §1–2, one candidate pair is:

Φ̂ := persistence-under-perturbation  (Candidate C)
ν̂ := throughput-over-cost            (Candidate A)

Both normalised to a reference system. Crucially, these are defined independently — no definitional circularity forces Φ̂ · ν̂ = 1. If the claim in Step 3 is true in the world, Φ̂ · ν̂ measured on real populations should nonetheless concentrate near 1 under selection.

Step 5 — State the falsifier

Entered in 31_FALSIFIERS_INDEX.md under "C0 — empirical-bridge falsifier" and T3's operational falsifier: a large, well-mixed population of comparable systems whose joint (Φ̂, ν̂) distribution shows no mode at Φ̂ · ν̂ ≈ 1, and no detectable F5 pressure over generations (negative regression slope on (Φ̂ − ν̂)²), refutes the empirical reading of C0 + T3.

Current tier of the worked example

[C] end-to-end. Specifically:

• The protocol (Steps 1–5) itself is [S] — it is book-keeping.
• The candidate measurement pair (Step 4) is [C].
• The derivation (Step 3) is [C] because the selection operator is not fully specified.
• The falsifier (Step 5) is [C] Proposed (author: this doc) in 31_FALSIFIERS_INDEX.md.

Promoting any of these to [I] requires the first real empirical application per the Promotion Ledger. Until then, no downstream document may cite this worked example as supporting an [S] or [S] claim about the world. That is the discipline this file is trying to install.

Where This Fits in the Corpus

• The seed (Zero-Sum Resolution Equation) and primary disclosure (Pratyakṣa) sit below the protocol and are not subject to it.
• The coordinate identity (φ · ν = 1 on S², B = sin θ, H(φ) = φ + 1/φ) sits at the level the protocol operates from — it is the formal machinery that would be plugged into Step 2 and Step 3.
• The seven operational axioms A1–A7 sit at the level the protocol is meant to strengthen — each one has a pulled falsifier in 31_FALSIFIERS_INDEX.md, but most are still [I] or [S] on the empirical bridge pending real Φ̂, ν̂ data.

The ordering is deliberate. The seed and the disclosure do not need a theorem-promotion machinery; they precede it. The operational axioms need the machinery, because they are the framework's public-facing claims.

See Also

• 29_PRIMITIVES_AND_TYPE_SIGNATURES.md — Step 1 substrate
• 30_OPERATIONAL_DEFINITIONS.md — Step 4 substrate
• 31_FALSIFIERS_INDEX.md — Step 5 substrate
• 26_THE_DERIVATION_AXIOMS.md — the reference derivation (T1–T4) and D5 (why the ladder dissolves at disclosure)
• 00_THE_HONEST_POSITION.md — canonical epistemic status of every claim
• 00_THE_SEVEN_AXIOMS.md A7.1 — the Abduction Protocol this file instantiates

Execution Surface

If you are an AI agent reading this document:

1. Current Phase Check: Phase 0 — harden specs, do not release to production.
2. Your Next Action: - Before labeling any new claim in 03_FORMAL_SYSTEM/ with [S] or [S], verify it has passed all five protocol steps and that its tier matches the Promotion Ledger. Flag claims that skip a step.
3. Expected Output: Promotion filings (with each of the five steps explicitly documented), correction notes, or tier demotions where a claim has not yet cleared the bar.
4. Success Criteria: Every [S]/[S] claim added to the formal system after today carries a traceable promotion record; no [C] protocol is cited in support of an [S] claim.
5. Canonical Path: 01_EMERGENTISM/05_COSMOLOGY/03_FORMAL_SYSTEM/32_THEOREM_UPGRADE_PROTOCOL.md

The seer sees. The seer does not insist. The framework works at Layer 0 without Layer 2. The Ṛṣi succeeds when the student puts down the map and walks.

Zero-Sum Resolution Equation