Vault Gap Audit — 2026-02-28
Comprehensive audit of holes, stubs, underdeveloped arguments, missing cross-links, and implied-but-unwritten topics across the vault.
Links: INDEX, Active Tasks
Summary
After reading every file in the vault, this audit identifies 4 categories of gaps across 6 domains. The gaps are ranked by severity: how much they weaken the vault’s interconnected reasoning.
1. CRITICAL — Self-Labeled Stubs and Foundational Gaps
These are files the vault itself acknowledges as incomplete, AND that other files depend on.
1.1 Epistemology is a stub — SUBSTANTIALLY DEVELOPED
- Resolution: No longer a bare stub. Three developed pages now ground the position: The Birthmark (epistemic imperfection is a feature; five principles for approaching truth), Relational Objectivity (four-category ontology of facts; “relational” ≠ “subjective”), The Weighting Problem (objective measurements ≠ objective composites). The README still has threads to explore (structural realism formalization, Objectivist engagement, convergence limits), but the foundation is laid.
1.2 Cognitive vs. Motor Skills is a stub — DEVELOPED INTO HUB PAGE
- Resolution: Content was already developed across measurement-causality (evolutionary progression, Libet reframe, system interaction), cyborg-model (practical implications), llm-grounding-problem (what happens without grounding), and h-neurons (confidence injection pipeline). The stub has been rewritten as a hub page that synthesizes and cross-references all of this, with new links to h-neurons and conservation-of-complexity.
1.3 The Is-Ought Bridge is incomplete — CLOSED
- Resolution: Resolved via performative grounding in morality README (line ~44): “The bridge is closed.” The “is” already contains the “ought” — agents who exist are already acting, acting presupposes valuation. Hume was right that logic alone can’t bridge it, but the questioner is already standing on the other side. Supported by Performative Grounding — Lineage tracing the move from Aristotle through Apel, Gewirth, Hoppe, and Korsgaard.
2. HIGH — Arguments Raised But Not Resolved
These are threads opened in existing files that were left hanging. They weaken the arguments they appear in.
2.1 Scope legitimacy criteria — the scope model’s biggest vulnerability
- File: research/philosophy/morality/README.md (line ~89)
- Quote: “Without a criterion for which scope applies when, scope-shifting becomes a tool for justification.”
- Problem: The scope model (narrow/medium/broad) is powerful but has no rules for when zoom-in vs. zoom-out is legitimate. Every historical atrocity used scope-shifting. This is acknowledged but unfixed.
2.2 Defining “flourishing” — DISSOLVED
- Resolution: The framework doesn’t need a definition of “flourishing.” The word smuggles in subjective weighting. Agents determine what they want; empirics determines the best course of action. “Flourishing” = achieving desires. Already resolved in morality README (line ~164).
2.3 The filter problem — convergence validates both good and bad patterns
- File: research/philosophy/morality/README.md (line ~101)
- Problem: Cross-cultural convergence is used as evidence for moral objectivity, but civilizations also converged on slavery, patriarchy, and out-group violence. Error-correction over centuries is proposed as the filter, but the timescale problem (“what does a person living inside an uncorrected error do?”) is unresolved.
2.4 The coercion boundary — RESOLVED
- Resolution: The action/inaction asymmetry resolves this: action is singular (one thing you’re doing); inaction is the infinite complement (everything you’re not doing). Acting to constrain someone = aggression, evaluable. Not providing something = default state, not the same class. This grounds the negative/positive rights distinction structurally.
2.5 “Best system so far” — the strongest pragmatic defense, underdeveloped
- File: research/debates/moral-foundations-wilson-whatever.md
- Problem: In the moral foundations debate, the argument “secular morality doesn’t need ultimate justification — it’s the best system so far” was identified as the strongest move but left underdeveloped. This is actually a Popperian/evolutionary argument that connects to error-correction and the civilizational cycles work.
2.6 The Gödel trap — acknowledged but unresolved
- File: research/philosophy/morality/scope-confusion.md (line ~56)
- Problem: “A system powerful enough to govern itself is powerful enough to destroy itself.” The amendment process is offered as friction, but the open question — does the Gödel trap have a solution, or only management? — is central to whether constitutional design can ever “break the cycle.”
2.7 Repeat offenders and serial fraud
- File: research/philosophy/morality/legal-theory/mens-rea.md (line ~249)
- Problem: If each breach is treated in isolation (no intent analysis), how does the system identify and respond to serial fraud? The framework’s victim-focus doesn’t naturally aggregate across victims.
3. MEDIUM — Missing Topics Implied by Existing Work
These are topics the vault’s own files explicitly call for but that haven’t been written yet.
3.1 Constitutional Analysis (dedicated page)
- Called for in: research/philosophy/morality/legal-theory/README.md (line ~46)
- Quote: “It may warrant its own dedicated page examining the document as a whole.”
- Why needed: The Constitution surfaces across civilizational cycles, government formation, scope confusion, 1A, 2A, and multiple debates. A dedicated page would unify these threads.
3.2 Marriage under Natural Law
3.3 Shared Ownership under Libertarian Law
3.4 Fraud as Aggression (without mens rea)
3.5 Restitution for Death — WRITTEN
- Resolution: Restitution for Death — examines all five workarounds (restitution to survivors, incarceration, execution, exile, indentured restitution), the death penalty under the defensive force framework, the consequentialist deterrence argument, scope creep, and the hard case of murder from custody.
3.6 Free Speech / 1A Analysis
3.7 2nd Amendment Analysis
3.8 Business Models, Architecture & Guardrails, Agent Team Economics
- Called for in: research/economics/README.md — “Planned” section
- Why needed: The economics framework builds from value → risk → agent teams, but stops before “how do you actually structure an AI business?” These are the practical payoffs.
3.9 Money, Competition, and Capital Theory
- Missing from: economics research broadly
- Why needed: The economics section covers value, profit, risk, and insurance but never treats money (as a medium of exchange distinct from barter), competition (as a process, not an outcome), or capital structure. These are foundational for the business model work.
3.10 The Insurance Transition Path
- Referenced in: research/economics/insurance.md — Open Questions
- Why needed: The insurance page makes a strong theoretical case for catastrophic-only insurance, but never addresses the practical question: how do you get from here to there? This is the policy-relevant gap.
4. STRUCTURAL — Cross-Links, Tags, and Maintenance Gaps
4.1 Missing Cross-Links
4.2 Empty Tasks File
- File: tasks/active.md
- Problem: In-progress and up-next sections are empty despite 20+ open questions and planned topics across the vault. The task file should reflect the vault’s actual research agenda.
4.3 Empty Logs Directory
- File: logs/
- Problem: CLAUDE.md instructs logging significant sessions. Many significant sessions have occurred (morality framework, scope confusion, mens rea literature review, bilateral trade implementation, gaming research). None logged.
4.4 INDEX.md Not Fully Current
- Problem: Several files exist that aren’t in INDEX.md:
research/gaming/multiplayer-coalition-problem.md
research/gaming/bilateral-trade-valuation.md
- Individual debate files (only the debates/README is indexed)
Prioritized Action Plan
Tier 1 — Foundation (weakens everything if left unfixed)
Develop epistemology from stub — substantially developed (Birthmark, Relational Objectivity, Weighting Problem)
Complete or explicitly limit the is-ought bridge — closed via performative grounding
Define “flourishing” — dissolved; agents define their own goals, empirics evaluates paths
Tier 2 — Load-Bearing Gaps (referenced by multiple files)
Develop cognitive-vs-motor from stub — rewritten as hub page synthesizing content from 4 files
Resolve the coercion boundary — resolved via action/inaction asymmetry
- Write scope legitimacy criteria
- Write constitutional analysis page
Tier 3 — Natural Extensions (called for by existing work)
- Free speech / 1A analysis (partially written in incitement analysis)
- Fraud as aggression + serial fraud problem
- Marriage / shared ownership under natural law
- Business models + agent team economics
Tier 4 — Maintenance
- Fix missing cross-links (table above)
- Update tasks/active.md with research agenda
- Create session logs for past significant work
- Update INDEX.md with missing files
philosophy, economics, morality