← all plans

Codex — Philosophy Plan

Drafted 2026-05-17. Final per-domain plan in the five-domain expansion. Sibling of PHYSICS_PLAN.md / CHEMISTRY_PLAN.md / BIOLOGY_PLAN.md. Subordinate to BIBLE_EXPANSION_PLAN.md. Authored last per the umbrella §6 sequence — philosophy is mostly a receiver of cross-domain hooks, so it benefits from seeing what shape the other four want to point at it.

Status: Draft. Locks the hybrid mode decision (philosophy gets both the existing synthesis essays AND a new tiered-units track for analytic philosophy of science), the subsection allocation under §20, anchor literature, mastery endpoints, the inbound-hook taxonomy that determines where new units are seeded, and the essay-ID retrofit.

Read before this plan: OVERVIEW.md; BIBLE_EXPANSION_PLAN.md (esp. §2.5 phil mode + §3 contract); the 6 existing essays at site/src/content/philosophy/01-reflexivity06-realization; PHYSICS_PLAN.md, CHEMISTRY_PLAN.md, BIOLOGY_PLAN.md outbound-hook taxonomies (they tell us what phil needs to land).


§1 What this plan does

  1. Resolves the mode question (open per umbrella): philosophy adopts hybrid mode, with two distinct content registers running parallel: synthesis essays (existing) + tiered analytic-philosophy units (new). Both forms have legitimate roles; neither subsumes the other.
  2. Retrofits the 6 existing essays with IDs 20.essays.0120.essays.06 so cross-refs from other domains resolve. This was a hard requirement from BIBLE_EXPANSION_PLAN.md §10 step 4.
  3. Allocates subsections under §20 for analytic philosophy of science — one subsection per scientific domain we mirror, plus logic, phil-of-science-general, and phil-of-mind.
  4. Names Wave 1 seeds: complete the essay retrofit + produce 2 analytic units chosen because they are the highest-leverage inbound-hook targets across all four other domains.
  5. Specifies the cross-domain return surface — how phil emits "see also" hooks back into science units (the reverse direction of the dominant hook flow).

It does not:

  • Lock the long-term shape of 20.essays (more essays will be added; this plan locks the format, not the count).
  • Decide whether phil eventually splits into multiple top-level sections (it stays at §20 for v1).
  • Override the existing 6 essays' content — those are kept verbatim; only frontmatter is touched.
  • Pretend Tyler has analytic-phil expertise; reviewer recruitment is even more important here than in chem or bio.

§2 What philosophy is for in this project

Two complementary roles, both load-bearing:

2.1 Role A — synthesis and orientation (the existing essays do this)

The 6 existing essays (reflexivity, geometry-of-disclosure, perfected-figure, what-passes-between, practice, realization) are foundational orienting documents. Their function is to provide a stance from which the rest of the curriculum can be read. They are not philosophy-of-science in the analytic sense; they are wisdom-tradition synthesis with structural awareness — Vedanta-meets-Sufi-meets-mathematics-meets-contemplative-practice.

They serve a learner approaching the project who needs to see what shape of knowing the project assumes before stepping into math/physics/chem/bio. They will continue to exist, be added to over time, and be cross-linkable from any unit in the project.

2.2 Role B — analytic philosophy of science (new in this plan)

When a physics unit on the measurement problem hooks out to phil, what does it hook to? Not a synthesis essay — the physics unit wants a specific analytic treatment of "what the measurement problem is, what positions exist, what the canonical arguments are." That requires tiered analytic-philosophy units alongside the essays.

Role B fills this gap. Each tiered unit covers a specific analytic-phil topic at three depths, anchored to the standard analytic-phil literature, with operationalized mastery endpoints. These are the natural landing places for hooks_out from physics §10/12/13, chem §14/15/16, bio §17–19, and math §00–08.

2.3 Why two modes, not one

Could the synthesis-mode subsume the analytic mode (everything as essays)? No — analytic phil-of-science arguments have testable mastery endpoints (reconstruct the argument, identify the hidden premise, produce a counterexample) that the contemplative/synthesis mode does not. Forcing them into the same form would either reduce essays to argument-reconstruction practice (losing what makes them essays) or reduce units to free-form synthesis (losing the testability that justifies units).

Could the analytic mode subsume the synthesis mode (everything as tiered units)? No — the existing 6 essays would not survive being broken into prereq-graph atoms; their power comes from the integrative arc.

Two modes is the right answer; the umbrella's hybrid decision is right; this plan operationalizes it.


§3 §20 subsection allocation

Philosophy occupies a single top-level section (§20) per the umbrella prefix table. Within §20:

ID prefix Subsection Role
20.essays.NN Synthesis essays Role A; existing 6 + future additions
20.01.NN Logic and formal methods Tiered units; covers propositional / first-order / modal logic, formal semantics, computability foundations
20.02.NN Philosophy of mathematics Tiered units; Platonism, structuralism, formalism, intuitionism, indispensability, set-theoretic foundations
20.03.NN Philosophy of physics Tiered units; interpretation of QM, classical-quantum boundary, spacetime ontology, the arrow of time
20.04.NN Philosophy of chemistry Tiered units; the bond ontology, reduction, structural realism in molecular chem
20.05.NN Philosophy of biology Tiered units; species, units of selection, fitness, teleology, life, agency, the gene concept
20.06.NN Philosophy of mind Tiered units; consciousness (hard problem), intentionality, computationalism, embodiment
20.07.NN General philosophy of science Tiered units; explanation, theory choice, evidence, scientific realism vs anti-realism, methodology

Locked. This is the §20 layout.

The 20.essays.NN subsection is a single flat namespace; essays don't subdivide by topic because their integrative arc spans multiple. The 20.01–07 subsections subdivide because analytic phil-of-science has clearly distinct sub-disciplines that map onto the other domains' content.

Addressing-semantics note. §20 uses the same dotted-triple section.chapter.unit form as math/physics/chem/bio, but with a semantic asymmetry: in math/physics/chem/bio, level-2 is a chapter inside one section. In §20, level-2 is a sub-discipline (logic / phil-of-math / phil-of-physics / etc.) — there is no "chapter of philosophy" the way there is "chapter of QM". The validator regex (^\d{2}\.\d{2}\.\d{2}$) treats both forms identically; the asymmetry is in human-readable interpretation only. Cross-domain hooks_out targets like 20.03.01 parse as section=20, sub-discipline=03 (phil-of-physics), unit=01 — not as section=20, chapter=03 in the math sense. Document this so authors don't conflate the two readings.

3.1 Why these 7 subsections, not more or fewer

Not more: ethics, aesthetics, political phil, history of phil are out of v1 scope for Codex (math-and-sciences focus per OVERVIEW.md). Phil of language, epistemology, metaphysics-as-general appear only via the phil-of-X subsections that they bear on (e.g., the epistemology of testimony in science appears in 20.07; metaphysical commitments of QM interpretation in 20.03).

Not fewer: collapsing phil-of-physics and phil-of-chemistry into "phil-of-natural-science" would obscure the very different methodological commitments each touches. Keeping them parallel to the science sections makes the cross-domain hook contract clean (a physics unit hooks to 20.03; a bio unit hooks to 20.05; etc.).


§4 The hybrid mode operationally

BIBLE_EXPANSION_PLAN.md §2.6 locks the general unit-vs-essay criteria. Specific to phil:

4.1 When to write a synthesis essay (the 20.essays.NN slot)

  • The content is integrative across multiple sub-disciplines of philosophy and/or across multiple traditions (analytic / continental / wisdom-tradition / contemplative). Cannot be anchored to a single prereq tree.
  • The content is interpretive or orienting — its job is to make a stance visible, not to teach an argument.
  • The author wants to draw a structural analogy across traditions (e.g., the 01-reflexivity essay drawing the same structure from Vedanta, Sufism, and mathematics).
  • There is no operationalized mastery endpoint — the essay is intended to change how the reader sees, not to be tested.

4.2 When to write a tiered analytic unit (under 20.0120.07)

  • The content is a position, an argument, or a concept with a definable scope.
  • It has standard literature anchors at Intermediate / Master (textbook canonical articles + primary papers).
  • It has a testable mastery endpoint — reconstruct, identify, counterexample.
  • It is the natural landing target for a hooks_out from a science unit.

4.3 Default in ambiguous cases

Default to tiered unit when ambiguous — same default as BIBLE_EXPANSION_PLAN.md §2.6. Essays accumulate worse than units; phil already has 6 essays and adding 20 more without discipline would shift the section's character.

4.4 Essay budget for v1

The 6 existing essays form an initial cluster. New essays are added one-at-a-time with explicit justification ("couldn't be a unit because X"). Soft cap: 12 essays total by end of v1, allowing for ~1 essay every couple of waves. Hard cap: 20.


§5 Tier anchors

5.1 Essays (20.essays.NN)

No tier system. Essays are written at a single register — accessible-but-substantial prose, comparable to The Hedgehog Review or a long-form Aeon essay. Reading-time scope: 20–60 minutes per essay.

5.2 Tiered units (20.01–07)

Anchored to analytic-philosophy literature at three tiers, per subsection.

20.01 Logic and formal methods

Tier Anchor
Beginner Tomassi Logic; Velleman How to Prove It; Logicomix (popular but rigorous structural intro)
Intermediate Boolos-Burgess-Jeffrey Computability and Logic; Mendelson Introduction to Mathematical Logic
Master Shoenfield Mathematical Logic; Marker Model Theory; Kanamori The Higher Infinite (set theory at master); Lean / Mathlib formalization of foundational results (cross-cites math §01)

Lean status note. 20.01 is the one phil subsection where Lean has direct purchase — propositional and first-order logic are exactly what formal verification handles. Master-tier units in 20.01 should aim for lean_status: full where Mathlib's logic libraries cover the result.

20.02 Philosophy of mathematics

Tier Anchor
Beginner Russell Introduction to Mathematical Philosophy; Hofstadter Gödel, Escher, Bach (popular but structurally serious)
Intermediate Shapiro Thinking About Mathematics: An Introduction to the Philosophy of Mathematics; Brown Philosophy of Mathematics: A Contemporary Introduction to the World of Proofs and Pictures
Master Shapiro Philosophy of Mathematics: Structure and Ontology; Maddy Realism in Mathematics + Naturalism in Mathematics; Benacerraf-Putnam Philosophy of Mathematics anthology (Benacerraf's "What Numbers Could Not Be" and "Mathematical Truth" are canonical); Field Science Without Numbers; Linnebo Philosophy of Mathematics

20.03 Philosophy of physics

Tier Anchor
Beginner Maudlin Philosophy of Physics: Space and Time; Albert Quantum Mechanics and Experience (early chapters)
Intermediate Albert Quantum Mechanics and Experience; Sklar Philosophy of Physics; Maudlin Philosophy of Physics: Quantum Theory
Master Wallace The Emergent Multiverse (Everettian QM); Bell Speakable and Unspeakable in Quantum Mechanics; Earman Bangs, Crunches, Whimpers, and Shrieks (singularity / GR phil); primary lit (Studies in Hist. & Phil. of Modern Physics)

20.04 Philosophy of chemistry

The smallest analytic-phil sub-discipline because the field is relatively young.

Tier Anchor
Beginner Bensaude-Vincent & Stengers A History of Chemistry (philosophical history)
Intermediate Hendry, Needham, Woody (eds.) Philosophy of Chemistry (Handbook of the Philosophy of Science, vol. 6) — the canonical reference
Master Scerri The Periodic Table: Its Story and Its Significance; primary lit (Foundations of Chemistry)

20.05 Philosophy of biology

Tier Anchor
Beginner Sober Philosophy of Biology (intro); Godfrey-Smith Philosophy of Biology (Princeton intro)
Intermediate Sober Philosophy of Biology (full); Godfrey-Smith Darwinian Populations and Natural Selection
Master Sober Evidence and Evolution; Okasha Evolution and the Levels of Selection; Wilson & Sober Unto Others; primary lit (Biology & Philosophy, Studies in HPS-C)

20.06 Philosophy of mind

Tier Anchor
Beginner Searle Mind: A Brief Introduction; Chalmers The Conscious Mind (intro chapters)
Intermediate Chalmers The Conscious Mind (full); Dennett Consciousness Explained; Block Imagery / Consciousness, Function, and Representation (collected papers)
Master Chalmers (collected papers); primary lit (Journal of Consciousness Studies, Mind & Language); contemporary debates (Tegmark / Tononi IIT vs higher-order theories vs global-workspace)

20.07 General philosophy of science

Tier Anchor
Beginner Okasha Philosophy of Science: A Very Short Introduction; Chalmers What Is This Thing Called Science?
Intermediate Curd-Cover-Pincock Philosophy of Science: The Central Issues (anthology); Kuhn Structure of Scientific Revolutions; Hempel selected papers
Master Salmon Scientific Explanation and the Causal Structure of the World; Cartwright How the Laws of Physics Lie + The Dappled World; van Fraassen The Scientific Image; Stanford Exceeding Our Grasp; primary lit (Philosophy of Science, Synthese, British Journal for the Philosophy of Science)

§6 Mastery endpoints per tier

6.1 Essays

No mastery endpoint. The reader emerges with a stance — testable only over years, not over a chapter.

6.2 Tiered units (across 20.01–07)

Tier Endpoint
Beginner Read and accurately summarize the argument in an intro-philosophy paragraph; identify the position being argued for; recognize obvious objections
Intermediate Reconstruct a formal argument from a canonical paper (e.g., Putnam's "What Theories Are Not", Lewis's "How to Define Theoretical Terms", Bell's Speakable); identify hidden premises; produce a counterexample to a stated position; write a paragraph-length defense or critique
Master Read a contemporary primary phil paper (post-2010), assess its argument, identify its place in the existing literature, and write a paragraph-length original response that engages with both the argument and at least one alternative position; do this consistently across the subsection (20.0N)'s scope

These mirror the math/physics/chem/bio mastery endpoints in structure (testable, evidence-able) while honoring philosophy's specific cognitive moves.

6.3 Machine-verifiability

Per BIBLE_EXPANSION_PLAN.md §5: phil at I/M tier is very low machine-verifiable (lowest of the five domains). Argument reconstruction at intermediate tier can be partially LLM-checked ("does the reconstruction preserve premise-conclusion structure"); at master tier, only human review counts.

Exception: 20.01 Logic and formal methods at Master tier where Lean covers the formal results — that subset is machine-verifiable like math. This makes 20.01 the easiest analytic-phil subsection to ship at scale.


§7 Cross-domain link contract — phil-specific specifics

7.1 Inbound hooks (the dominant direction)

Phil receives hooks_out from all four other domains. The expected target distribution (rough, to be updated after Wave 2):

Source domain Likely target subsections
Math 20.01 logic, 20.02 phil-of-math
Physics 20.03 phil-of-physics (heaviest landing site, esp. from §12 QM and §13 GR); also 20.07 for explanation / theory questions
Chemistry 20.04 phil-of-chem (sparse); 20.07 for reduction questions
Biology 20.05 phil-of-biology (heavy, esp. from §19 eco/evo); 20.06 phil-of-mind (from §17.09 + §18.05)

7.2 Outbound hooks (the rare direction)

Phil can emit hooks_out back into science units. Examples:

  • A unit on 20.03.01 The measurement problem in QM may hook back to physics 12.01.02 Stern-Gerlach ("see also: the experimental setup this debate centers on").
  • A unit on 20.05.02 Units of selection may hook back to bio 19.03.NN natural selection units.
  • An essay may hook back to multiple units across multiple domains as "where this view shows up empirically."

Validator treats phil's outbound hooks identically — proposed/confirmed, why ≥ 30 chars, etc.

7.3 The "see also" footer convention (UI-level, not contract-level)

Per BIBLE_EXPANSION_PLAN.md §2.5: every science unit that hooks into a phil target should, when rendered, display a "Philosophical questions this raises" footer linking to the phil targets. This is a rendering convention — site-side — not a frontmatter contract; the data is already in hooks_out.

7.4 Pending-hook absorption for phil

Per BIBLE_EXPANSION_PLAN.md §3.6 step 1: this plan's Wave 1 backlog is informed by harvesting manifests/connections.json for kind: proposed edges with target in §20. At this writing, that backlog is empty (no physics/chem/bio units have shipped yet under the new contract). It will populate as those Wave 1 units ship. The phil plan revisits its seed-unit choices once that backlog is non-empty.


§8 First wave — essay retrofit + 2 tiered seed units

Three deliverables, executed in this order:

8.1 Essay ID retrofit (mandatory)

Required by BIBLE_EXPANSION_PLAN.md §10 step 4. Adds id: 20.essays.NN frontmatter to each existing essay:

File Assigned ID
01-reflexivity.md 20.essays.01
02-geometry-of-disclosure.md 20.essays.02
03-perfected-figure.md 20.essays.03
04-what-passes-between.md 20.essays.04
05-practice.md 20.essays.05
06-realization.md 20.essays.06

Each essay also gets an applies_to: frontmatter field listing the section numbers it can be cross-referenced from (likely [20] for all 6 since they're integrative; deferred per-essay to retrofit-time).

8.2 Two tiered seed units

Chosen because they are the highest-leverage inbound-hook targets — the topics that the other four domains' Wave-1 units will most likely want to hook into:

Seed ID Subsection Why this one
The measurement problem in QM 20.03.01 Phil-of-physics Physics §12 Wave 1 ships Stern-Gerlach (12.01.02). Every QM interpretation question hooks here. Most-densely-anticipated phil target. Anchored in Albert + Maudlin + Wallace.
The unit of selection 20.05.02 Phil-of-biology Bio §19 Wave 1 ships Hardy-Weinberg (19.02.01). Every "what is selection acting on" question hooks here. Anchored in Sober + Okasha.

Two units, not more, because: (a) Wave 1's purpose is to validate the contract on canonical analytic-phil topics; (b) reviewer recruitment for phil is itself a bottleneck (§9); (c) the backlog from other domains is still empty at draft time, so seeding more than two without those signals is premature.

8.3 Sequence

20.essays retrofit (all 6, batch) → 20.03.01 measurement problem → 20.05.02 unit of selection.

Rationale:

  • Retrofit first because it's blocking other plans' validators and is mechanical (no content change).
  • Measurement problem second because phil-of-physics has the densest hook surface and is the cleanest analytic-phil topic to start with.
  • Unit of selection third because it tests hooks_out to bio at the same time as hooks_in from bio, exercising bidirectional cross-domain edges.

8.4 Success criteria (Wave 1 phil)

  • All 6 essays carry id: 20.essays.NN and applies_to: frontmatter; validator resolves cross-refs into them.
  • 20.03.01 and 20.05.02 exist in content/20-philosophy/, structured as tiered units with B/I/M sections, frontmatter hooks_out, anchored references.
  • 20.03.01 declares hooks_out → 12.01.02 Stern-Gerlach (kind: proposed regardless of target ship state — promotion to confirmed requires a physics-side reviewer's semantic-correctness attestation per umbrella §3.2, which is not part of Wave 1).
  • 20.05.02 declares hooks_out → 19.02.01 Hardy-Weinberg (kind: proposed, same rule).
  • Validator resolves all phil cross-refs in the site build.
  • At least one science unit from physics/bio Wave 1 references 20.03.01 or 20.05.02 directly (closes the inbound-hook loop on those targets).
  • Retro in docs/pilot-lessons.md — what worked, what didn't, what the analytic-phil rubric needs.

The wave fails if:

  • The essay retrofit breaks the existing site's rendering of philosophy essays. (Validators must be additive, not breaking.)
  • The bidirectional hook on 20.05.02 ↔ 19.02.01 doesn't actually resolve in connections.json.

(The Lean-anchor expectation applies to 20.01 Logic and formal methods only, not to 20.03 phil-of-physics — see §6.3 machine-verifiability note.)

8.5 Production strategy

Item Strategy Notes
Essay retrofit Automation + spot-check Mechanical frontmatter addition; one script + visual check; not authoring
20.03.01 measurement problem Manual First analytic-phil unit; needs human author judgment on philosophical-position rendering
20.05.02 unit of selection Manual Same; ships before agent-drafted phil is attempted

Agent-drafted phil units are deferred to Wave 2 at earliest. Manual gives us the rubric calibration we need — and analytic phil has a higher prose-craft demand than mechanism-explanation does.


§9 Reviewer roster

The hardest reviewer-recruitment problem in the project. Phil reviewers:

  • Cannot be the same people as the science reviewers (different training).
  • Are not typically integrated into the cross-disciplinary review processes the science domains use.
  • Cover sub-disciplines that are themselves siloed (phil-of-physics ≠ phil-of-biology in practitioner pool).
Subsection Tyler-solo? Outside reviewer needed
20.essays Yellow (synthesis writing is a Tyler-strength based on the existing 6 essays) Optional external read for new essays
20.01 logic Yellow with Lean grounding Logician for Master-tier (where Mathlib doesn't cover)
20.02 phil-of-math Yellow with math grounding Phil-of-math specialist for Master
20.03 phil-of-physics Yellow / Red Phil-of-physics specialist (with physics-side credentialing) for I/M
20.04 phil-of-chem Red Phil-of-chem specialist (small field; may need to commission)
20.05 phil-of-bio Yellow / Red Phil-of-bio specialist for I/M
20.06 phil-of-mind Red Phil-of-mind specialist for I/M (this is one of the most contested subfields in philosophy)
20.07 general phil-of-science Yellow / Red Phil-of-science generalist for I/M

Recruitment priority (in order):

  1. Phil-of-physics + phil-of-biology specialist OR generalist phil-of-science specialist with strong cross-domain reading — covers the densest hook landing sites (20.03 + 20.05). Single highest-leverage hire.
  2. Phil-of-math + logic specialist — covers 20.01 + 20.02 + the formal side of 20.07.
  3. Phil-of-mind specialist — covers 20.06; can't easily be substituted.

20.04 (phil-of-chem) is small enough that commissioning a one-off review for each Master-tier unit is acceptable in v1; no dedicated reviewer required.

Tyler-as-essay-reviewer. The synthesis-essay role is the one place where Tyler is plausibly the right reviewer — the existing 6 essays demonstrate the project's voice. New essays go through Tyler review by default. Tiered units do not, except in 20.essays.NN.

LLM-augmented review is least reliable here. LLMs can catch obvious argument errors and bad citations; they cannot adjudicate live philosophical disputes. Master-tier phil units should be held in draft until a human reviewer is named.


§10 Open questions (deferred)

Question Deferred to Why now is wrong
Should ethics / political phil / aesthetics get a §20.08+ subsection in v2? v2 scope discussion v1 is explicitly math-sciences-focused; expanding to normative phil is a different product question
Does 20.essays need internal sub-organization (e.g., a 20.essays.philosophical/, 20.essays.contemplative/) once it grows past ~10? After essay count exceeds 10 Premature with 6 essays
Should 20.06 phil-of-mind absorb a "cognitive science" interface as bio fills in §17.09 + §18.05? After Wave 2 bio Premature; bio sets the agenda for what the phil side wants to absorb
Whether to introduce a "history of science" subsection as v1 grows Probably v2; lean against Not a sub-discipline of phil-of-science proper; would distort §20 scope
How essays handle citation — same [ref: source] system as units, or freer narrative referencing? Per-essay; default to the same system Forcing strict citation may not survive the essay's voice (the existing 6 use a different reference style)
Does the synthesis-mode register need its own style guide separate from style/editorial-voice.md? After 3 new essays accumulate Premature; revisit when the pattern is observable

§11 Risks

Risk Mitigation
Reviewer recruitment for phil stalls completely; master-tier phil units accumulate in draft Restrict phil to Beginner+Intermediate until ≥ 1 of each priority-reviewer-role is named. Synthesis essays can ship under Tyler review without external phil credentials.
The two registers (essays + analytic units) feel like different products on the same site Site rendering treats them differently but visibly related — /philosophy route shows both, with the essay arc as featured content and the tiered units as a search-and-browse catalog. UI work for BIBLE_EXPANSION_PLAN.md site phase.
Analytic phil-of-physics or phil-of-bio units default to one position (e.g., Everettian QM, or gene-selectionism) and present it as canonical Mastery rubric requires that each unit covers the position space, not advocates for one; reviewer enforces. This is a known failure mode in popular analytic-phil teaching.
Essay accumulation pace becomes unsustainable as more essays seem warranted Soft cap 12, hard cap 20 (§4.4) enforces discipline. Each new essay requires explicit "couldn't be a unit because X" justification at review.
Cross-domain hook flow becomes uneven — physics dumps 30 hooks into 20.03 while chem barely touches 20.04 Acceptable; this is real. 20.04 phil-of-chem will be smaller than 20.03 phil-of-physics because the underlying analytic-phil subfield is. No artificial balancing.
The 6 existing essays' style conflicts with the analytic-unit style and the project looks tonally inconsistent Style guide for 20.01–07 follows style/editorial-voice.md; 20.essays.NN continues its existing voice. The two registers are visibly different — by design — and the site treats them as such.
Phil-of-mind (20.06) becomes a magnet for fringe content (panpsychism, simulation theory, etc.) at master tier Master-tier sources are the gatekeeping — Chalmers, Block, Dennett are the standard; popular and fringe positions appear only insofar as they are responded to in the canonical literature.

§12 Decision log entries (to be added to OVERVIEW.md §12 on the same date)

Date Decision Rationale
2026-05-17 Philosophy adopts hybrid mode: existing synthesis essays continue at 20.essays.NN; new tiered analytic-philosophy units at 20.01–07.NN Essays cannot subsume analytic argument-reconstruction (no testable endpoint); tiered units cannot subsume the integrative arc of essays. Two registers, both load-bearing.
2026-05-17 §20 subsection allocation locked: 20.essays + 20.01 logic + 20.02 phil-of-math + 20.03 phil-of-physics + 20.04 phil-of-chem + 20.05 phil-of-bio + 20.06 phil-of-mind + 20.07 general phil-of-science Mirrors the science domains plus logic; ethics / political / aesthetics out of v1; subsection-per-domain makes cross-domain hook contract clean
2026-05-17 Essay retrofit: 6 existing essays get IDs 20.essays.0120.essays.06 Required by umbrella §10 step 4; mechanical change; unblocks cross-refs from other domains
2026-05-17 Initial essay budget: soft cap 12, hard cap 20 by end of v1 Essays accumulate without validator discipline; explicit caps prevent register drift
2026-05-17 20.01 is the one phil subsection that targets lean_status: full at master tier where Mathlib covers Propositional + first-order logic + parts of computability are exactly what Mathlib covers; this is the bridge between phil and the math/Lean pipeline
2026-05-17 Wave 1 phil: essay retrofit + 2 tiered seed units (20.03.01 measurement problem, 20.05.02 unit of selection) Highest-leverage hook landing sites for physics and bio Wave 1; calibrates analytic-phil rubric on canonical topics; reviewer recruitment gates more
2026-05-17 Tyler is the default reviewer for 20.essays.NN; tiered units require outside reviewers per §9 priority list Synthesis writing is a demonstrated Tyler strength (the 6 existing essays); analytic phil is not — be honest about this
2026-05-17 Master-tier phil units held in draft if no named human reviewer is available; LLM-augmented review explicitly not sufficient for master Phil mastery endpoints (assess primary lit, write original response) are the least machine-verifiable in the whole project

§13 Next immediate actions

Gated on BIBLE_EXPANSION_PLAN.md §10 patches (UNIT_SPEC + validator + OVERVIEW patches) AND chem/bio Wave 1 retros so the phil-side seed slate can be adjusted if the inbound-hook taxonomy from those waves differs from this plan's prediction:

  1. Essay retrofit pass — write scripts/retrofit_essay_ids.py (mechanical: parse the 6 essay files, prepend id: and applies_to: to frontmatter, idempotent). Run; spot-check rendering of /philosophy/01-reflexivity etc. for regressions; commit.
  2. Reviewer recruitment — start outreach for the three priority reviewer roles (§9). This is the longest-pole item for phil, so start immediately.
  3. Sourcing passreference/_meta/SOURCES.md extensions for phil. Phil sourcing is harder than science sourcing because much canonical phil is in copyrighted anthologies; identify which works have open / fair-use accessible versions.
  4. Coverage manifestmanifests/production/phil-coverage.json mapping the §20 subsection anchor literature to (eventually) Codex unit IDs.
  5. Harvest pending-hook backlog — once Physics + Chemistry + Biology Wave 1 units have shipped, query manifests/connections.json for kind: proposed edges targeting §20; absorb each into the phil-side backlog per BIBLE_EXPANSION_PLAN.md §3.6.
  6. Produce 20.03.01 measurement problem manually — first analytic-phil unit; establishes the analytic-phil rubric.
  7. Retro on 20.03.01 — what's missing in the analytic-phil pedagogy spec; update the rubric.
  8. Produce 20.05.02 unit of selection manually.
  9. Cross-domain audit on Wave 1 phil: did the bidirectional hooks (20.03.01 ↔ 12.01.02 and 20.05.02 ↔ 19.02.01) resolve? Are there obvious missing hooks from the other domains?
  10. Update the existing philosophy.tsx route and the _layout.tsx philosophy nav-link to expose both registers — essays as the featured arc, tiered units as a browse catalog. (Site work — coordinate with whoever is doing site updates after BIBLE_EXPANSION_PLAN.md §10 step 4.)

This plan is the canonical reference for the philosophy axis and concludes the per-domain plan sequence. When uncertain about phil mode, subsection allocation, or essay-vs-unit decisions, check here. When updating, propagate to OVERVIEW.md §12 and BIBLE_EXPANSION_PLAN.md §6 the same day.


With this plan drafted, the five-domain expansion has a complete planning surface: one umbrella + four per-domain plans. The next phase is implementation — UNIT_SPEC and validator patches, OVERVIEW decision-log appends, Physics Wave 1 production, then chem ∥ bio production informed by physics retro, then phil informed by everything else. See BIBLE_EXPANSION_PLAN.md §6 for the operational sequence.