soul-frame/history/260928-scope-intent-gap-assessment.md

# Scope vs. Intent — Gap Assessment (2026-09-28)

**Context:** SCOPE.md updated the same day to reflect actual repo state
(Phases I–IX complete, SOUL-WP-0001…0010 finished). This note assesses that
state against `INTENT.md`'s standing purpose, six determinations, and
success criterion, and ranks the gaps.

---

## 1. Assessment against INTENT's six determinations

| # | INTENT: "determine…" | State | Verdict |
|---|---|---|---|
| 1 | what is already well established | 13 workstream literature maps, SF-R### ledger, five phase syntheses | **Met** for the traversed fields; fiction set and some WS depth (WS11, WS13) remain thin |
| 2 | what Soul Frame recombines usefully | `research/concept-crosswalk.md`, related-work §, phase syntheses | **Met** |
| 3 | what concepts require refinement | H1–H10 all phase-reviewed; RQ1–RQ8 carry working answers with explicit residuals | **Met as process**; residuals live in OpenQuestions |
| 4 | where terminology conflicts with usage | Terminology standard; RQ4 Soul retention decided with disclaimer policy; Deppert false-friend flagged | **Met** |
| 5 | what genuinely new propositions emerge | B1–B8 / P1–P12 profiles, three-axis independence, Sp./Gh. as provenance descriptors, recognition test, continuity channel set | **Met**, but propositions are profile-shaped with **no aggregation/threshold principle** (→ G1) |
| 6a | can these support a rigorous theory | Paper draft exists, adversarially reviewed | **Partially met** — rigor currently rests on per-dimension analysis; sufficiency questions ("how much personhood is *enough* for Beinghood v0.2") are open everywhere they matter (→ G1) |
| 6b | …and eventually a practical architecture | Deliberately downstream (INTENT itself: "downstream of conceptual clarity") | **Out of repo scope** — not a gap to close here |

## 2. Assessment against the success criterion

INTENT's success criterion: one coherent vocabulary describing **nine
reference entity types** (human, animal, short-lived AI agent, persistent
autonomous Agent-being, Einstein `Sp.`, authorized `Gh.`, original
artificial companion, distributed AI, radically non-human system) *while
still making meaningful distinctions among them*.

**Finding: this test has never been run as such.** The closest artifacts
cover fragments — `personhood-dimensions-map.md` §3 profiles five entities
on three coarse axes; thought experiments each stress one entity class.
No document applies the full vocabulary (B1–B8, P1–P12, continuity
channels, addressability/recognition, Agent/Sp./Gh. descriptors) across
all nine and checks that the distinctions survive. The program's own
success criterion is currently **asserted, not exercised** (→ G2).

## 3. Gap register

| ID | Gap | Source | Addressable now? |
|---|---|---|---|
| **G1** | No principled weighting/sufficiency scheme for the B-, P-, and continuity profiles; "sufficient personhood," "enough closure," and continuity thresholds are undefined wherever they are load-bearing | RQ1, RQ2, RQ5 (OQ-043); paper §17 item 2 | **Yes** — conceptual work, no dependency |
| **G2** | Success-criterion coverage matrix (nine entity types × full vocabulary) never produced | INTENT success criterion | **Yes** — synthesis of existing material |
| **G3** | Soul packaging unresolved: relations vs consequences vs both; Cluster D formal depth | RQ4 remainder, flagged **paper-critical** | **Yes** — bounded concept work |
| G4 | Metrics instruments scoped but not built (B3, psychological continuity) | WP-0010 residual | **Gated** — open decision `0401cf28`; do not start without resolution |
| G5 | Companion line items 2–4 (moral status survey, non-derivation statement, jurisdictional tracking) | Charter | **Gated** — charter's own next-step rule; slice-by-slice |
| G6 | Paper not yet circulated; internal-review stage depends on coulomb-research being operational | WP-0009-T06 decision | **Blocked externally** — facility still being set up |
| G7 | Thin coverage: WS11 (death/persistence depth), WS13 fiction set breadth, hybrid display conventions, eval harnesses | §17 item 6; RQ7/RQ8 residuals | Yes, but low urgency — none is load-bearing for the paper |

## 4. Ranking — most relevant gaps

**G1, G2, G3**, in that order, are the gaps worth addressing now:

- **G1** is the deepest structural debt: every profile the framework
  introduces ends in "weighting and thresholds open." Determination 6a
  cannot be claimed while the framework has no stated position on
  aggregation — even a principled *refusal* to weight (context-of-claim
  relativity) would be a position; silence is not.
- **G2** is the cheapest high-value item: it directly exercises the
  success criterion, likely surfaces vocabulary strain exactly where
  INTENT predicts it (distributed AI, radically non-human), and produces
  the paper's strongest possible exhibit.
- **G3** is the only remaining item OpenQuestions marks paper-critical.
- G4–G6 are gated or blocked by design; G7 is real but not urgent.

**Disposition:** G1–G3 are picked up by **SOUL-WP-0011** (created with this
assessment). G4–G6 stay with their existing gates. G7 remains in
OpenQuestions/§17 as agenda, not debt.