The Methodology Reference

Framework Synthesis

Purpose: Pull together what the methodology computation framework IS and DOES as of. Not the methodology document itself (which describes L1-L4 layers); this is the picture of the framework AS DEVELOPED, integrating the architecture + computational stack + manifestation corpus + the 7 structural findings produced by this and prior sessions.

This is the "architecture pulled together" doc the user has been asking for. Sits above the per-session notes and below the canonical methodology + canonical-architecture-strategy.

Status: Prototype. Operational. Nothing here is closed; everything can develop further. The point of pulling together is to make what's been built coherent and visible so further development can be informed.

1. What the framework IS

1.1 The methodology as a four-layer analytical instrument

The methodology produces structural understanding of domains. Per methodology.md it operates at four layers:

L1-L3 are structural (build abstract knowledge). L4 is operational (apply that knowledge). The framework has internal feedback: applied findings reveal gaps in the structural layers and trigger refinement.

1.2 The scope ladder

Per canonical-architecture-strategy.md §2.2, every analytical claim has a scope Sc:

ScNameWhat entersValidity
0UniversalPure category structurePatterns across all instances
1ArrangementLattice topologyCorridors, walks, bottlenecks (topological only)
2WeightedRates + populationsProbability distributions, expected times
3EmpiricalSpecific manifestationsPosterior distributions, real-system positions
4EventSingle cross-arrangement eventPoint-collapsed manifestations across coupled arrangements

Time enters at Sc=2; probability enters at Sc=2; specific systems are Sc=3 evidence; cross-arrangement coupling is Sc=4 (single event) or Sc=3 sustained (paired-manifestation form).

1.3 The five arrangements

Each arrangement is a chain of domains connected by bridges plus a context root. The framework has 5 currently operational arrangements:

At Sc=4 these arrangements collapse onto shared physical substrate (the developer-keypress event is one demonstrated example).

1.4 The compute layer

1.5 The data corpus

2. The 24 structural findings

[Updated (Viterbi pass): Findings 23 + 24 added from the Viterbi coherent-path implementation that closed the long-standing backwards-inference gap. Finding 23: bridge bottlenecks render as contiguous flat shelves in Viterbi paths — abiogenesis-r0-to-r2 spends ranks 7→17 in the chemistry-to-biology-bridge with substrate primitives frozen; chemistry-to-biology-crossing shows a clean three-phase decomposition (chemistry → bridge → substrate as three contiguous Viterbi blocks). Modal-per-rank scatters these phases. Finding 24: total Viterbi path probability is a corridor-branching diagnostic — single-chain-level min-to-max walks at small rank land at P=0.5–1.0 (effectively single path); multi-chain-level or large-rank walks land at P=1e-04 to 1e-06 (best-of-many roughly-equiprobable paths). Order-of-magnitude span across walks. Compute: compute/lib/rate_corridor.py viterbi_path + viterbi_vs_modal_summary. Renderer: compute/scripts/plot_backwards_inference.py extended to render Viterbi as dashed lines alongside modal-per-rank. Result files: output/results/<walk>-viterbi.v1.json (5 walks).]

[Updated Findings 21 + 22 added from the V5 paired-coupling pass — restricting Finding 18's cognitive-load × reach analysis to the 34 entity-paired cultural-artifact Mns sharpens the layering-trap correlation to r=−0.408 (vs −0.331 on the full N=47 cultural-artifact corpus); cog cultural-artifact Mn-kind has structurally-fixed 2-level shape across all 34 paired Mns. Compute: compute/scripts/plot_paired_coupling.py, compute/scripts/analyze_paired_coupling.py. Data: data/topologies/paired-coupling.v1.json. See topology-views-disentangle.md V5 row.]

[Updated Finding 20 added — Finding 7's "gap at P" is NOT universal; different ecosystem aggregates leave different irreducible gaps. git-platform → P (Δ=3); database → P (Δ=4) + I (Δ=2); decentralized → M (Δ=3); messaging → X+P (Δ=3 each). General principle: ecosystem aggregates leave gaps at primitives their constituents weren't designed to provide. Reframes entity-system's unique contribution as "coherent six-primitive substrate" rather than "P specifically". Compute: compute/scripts/analyze_ecosystem_aggregates.py. See structural-surfacings-2.md Finding 20.]

[Updated Finding 19 added — bridge-level structural fingerprint reveals three regimes by chain position (substrate-to-substrate / substrate-to-surface / surface-to-ecosystem); cross-arrangement L3 isomorphism is measurable at the bridge-fingerprint level (biology-to-organism ↔ neural-to-cognitive d=0.237). Multivariate filter prediction is overfit at N=11 (LOO R² = −0.243); the MDS clustering is the solid result. Compute: compute/scripts/analyze_bridge_fingerprint.py. See structural-surfacings-2.md Finding 19.]

[Updated (corrections + methodology): Finding 12 walked back per user feedback — the cross-arrangement quadrant ordering used non-comparable numbers (different chain levels measured for different groups, zero-vector cosine fallback artifacts). Re-stated as hypothesis pending proper cross-arrangement normalization. Finding 13 added from methodology fingerprint pass with a per-chain-level same-arrangement spread comparison: biology 0.008 < cultural-artifacts 0.023 < cognition-organism 0.16 < entity 0.32 < methodology 0.52, directionally tracking mechanism transferability. Finding 11 reframed from design-vs-evolution to substrate-transferability per user correction.]

[Updated (later): Findings 10 and 11 added from the structural-fingerprint comparison pass. Finding 10: three structural classes (tools/apps/runtimes) in entity-arrangement that substrate-rank cannot distinguish. Finding 11: biology fingerprint spread is 30× tighter than entity. See structural-fingerprint-comparison.md. Compute: plot_structural_fingerprint_landscape.py + new plot_mn_comprehensive.py per-Mn card.]

[Updated (earlier): Findings 8 and 9 added from a within-arrangement-cascade + bridge-corpus surfacing pass. Finding 6's filter row corrected to 19.5% (was 6.1% by error). See structural-surfacings-2.md.]

These are the load-bearing analytical observations the framework has produced. Each is reproducible from existing data — none required new manifestation authoring beyond what was already in the corpus by.

Finding 1 — Encoder × Evaluator pair-load tracks evaluator determinism

Cross-arrangement: (entity E×X) heavy, (biology G×R) heavy, (cognition Sy×Ev) medium. The cognition exception is the predicted outcome: per the split-evaluator note in methodology.md, cognition has a SPLIT evaluator (Sy formal Kd4, Ev linguistic Kd1-2). The medium classification at (Sy, Ev) is "encoder × soft-evaluator" — the heavy signature appears only when the evaluator is separable + Kd4-deterministic.

Document: cross-arrangement-pair-structure.md Finding 1.

Finding 2 — Substrate heavy-pair density ranks by design-vs-search axis

Entity substrate 73% heavy / biology substrate 47% / cognition substrate 40%. Descending order matches the calibration-derived regime axis: designed (entity) → Darwinian-searched (biology) → emergent / soft-evaluated (cognition).

Document: cross-arrangement-pair-structure.md Finding 2.

Finding 3 — Calibration's scale non-invariance + within-trajectory regime decomposition

step_duration (wall-time per weighted corridor step) spans ~8 orders of magnitude across 8 calibrated trajectories (~47 days/step chimp ontogenesis → ~30 Myr/step phylogenesis-stem). Within-trajectory residual signs partition into 3 regimes:

Document: calibration-final-summary.md.

Finding 4 — Per-transition population-vs-rate-corridor decomposition gives 3 distinct trajectory shapes

Population-context wired into compute. For each transition snapshot[i] → snapshot[i+1]: effective Nλt and regime classification. Three trajectory shapes emerged:

Each pattern aligns with the trajectory's regime classification. Independent confirmation of Finding 3's regime axis.

Document: population-context-wiring.md.

Finding 5 — Ecosystem domains have a cross-arrangement L3 invariant at ~7%

Coarse filter stringency, partitioned by chain-level role across 24 domains + 11 bridges:

Rolenavg filterrange
ecosystem36.8%6.4 – 7.2%
substrate921.9%9.4 – 37.5%
bridge1122.7%4.1 – 42.2%
surface316.5%4.9 – 23.6%
context525.9%14.1 – 34.4%

The ecosystem-role L3 invariant is the cleanest cross-arrangement filter pattern. Three ecosystem domains across 3 distinct arrangements (biological, cultural, digital) converge to ~7% — ecosystems are tightly co-dependent population structures regardless of substrate type.

Document: structural-surfacings.md Finding 5.

Finding 6 — The 12-mechanism bridge L3 analogy holds at primitive-count, breaks at dependency-density

Biology-to-organism-bridge (12 developmental mechanisms) vs entity-to-app-bridge (12 system extensions):

Propertybiology→organismentity→app
Primitive count1212
Internal dependencies283
Filter stringency19.5%42.2%

[Filter row corrected was 6.1% / 42.2%, now 19.5% / 42.2% per the JSON. 2.2× ratio not 7×.]

Same primitive count, opposite internal dependency density. Biological development is cascade-integrated (CDif requires CDiv, etc.); entity extensions are additive-modular (most extensions don't presuppose each other). Two different design strategies producing the same surface form (12 mechanisms each).

Document: structural-surfacings.md Finding 6.

Finding 7 — Git ecosystem aggregates within Δ=4 of entity-system; the irreducible gap is at Peer

Per-primitive max across {git, github, docker, kubernetes, nix, http-rest}:

Primitiveentity-system Mn (max)Git+ecosystem jointGap
E (Entity)44✅ COVERED
I (Identity)33✅ COVERED
T (Tree)44✅ COVERED
M (Emit)44✅ COVERED
X (eXecute)54Δ=1
P (Peer)52Δ=3

5 of 6 substrate primitives are reached by ecosystem accretion. The irreducible structural gap is at P — no ecosystem member provides peer-symmetric capability-based authority (GitHub is client-server P=2, not peer-equal P=5). At the application-architecture level, the joint covers all 12 surface primitives.

Document: structural-surfacings.md Finding 7.

Finding 8 — Within-arrangement chain-level filter cascade has a three-act shape

Reading filter% along each arrangement's chain (substrate-ground → … → substrate-top → surface → ecosystem):

Arrangementsubstrate-topsurfaceΔ ratioecosystemsurface→eco squeeze
Biology12.5%15.0%1.20×7.23%2.07×
Cognition26.6%37.7%1.42×8.8%4.28×
Entity14.06%21.1%1.50×7.23%2.92×

Substrate ladder tightens monotonically; surface relaxes ~1.4× looser than substrate-top (narrow proportional band); ecosystem re-tightens to the Finding-5 invariant. Methodology (self-referential, no ecosystem) does NOT follow this shape — positive control.

The three-act cascade is a within-arrangement structural prediction: any new arrangement with a realization-chain structure should produce the same pattern. Substrate encodes constraint, surface inherits ~1.4× degrees of freedom, ecosystem re-constrains by population co-functionality.

Document: structural-surfacings-2.md Finding 8.

Finding 9 — Bridge dep-density × filter at corpus scale (N=11)

Finding 6 compared two bridges. Extending to all 11: Pearson r(dep_density, filter%) = −0.438. The dep-density-filter mechanism holds at corpus scale (sign as predicted, moderate magnitude). Two structural outliers explain most of the residual:

Bridge filter ≈ dep_density × substrate_inherited_tightness × ecosystem-style-filter_if_terminal. Findings 2, 5, 6 unified.

Document: structural-surfacings-2.md Finding 9.

Finding 10 — Structural fingerprint reveals three classes in entity-arrangement that substrate-rank cannot distinguish

Per-Mn fingerprint vector: raw partial-level positions + pair-realizations (intensity-weighted: heavy=3 / medium=2 / light=1) + composition-realizations (min member level × member count). Cosine distance, classical MDS, distance heatmap. Three structural classes emerge in entity-arrangement N=35:

Wall-zone (substrate ≥14) splits across all three classes. The entity-system max-substrate Mn is structurally nearest to smtp-email (rank 7) at d=0.015 — shape similarity outweighs the rank gap. Substrate-rank × adoption-rank landscape cannot distinguish these classes.

Compute: compute/scripts/plot_structural_fingerprint_landscape.py. Per-Mn comprehensive card view: compute/scripts/plot_mn_comprehensive.py. Documents: structural-fingerprint-comparison.md.

Finding 11 — Biology fingerprint spread is 30× tighter than entity (design-vs-evolution signature)

Same fingerprint construction on biology arrangement (N=20):

entity (N=35)biology (N=20)
Max pairwise distance~0.80~0.025
Cluster count at d=0.153 classes + singletons1 cluster (everyone within 0.025)

Biology Mns occupy a structurally tight region; entity Mns spread widely. Inside biology's tight region, sub-blocks mirror PHYLOGENY without any phylogenetic input in the data: mammals (chimpanzee, mus, human) → vertebrates (zebrafish, python, gallus) → bilaterians+plants+fungi → single-celled eukaryotes → prokaryotes (ecoli, methanococcus, halobacterium farthest from mammals).

Interpretation. Evolution explores the full substrate uniformly (every organism needs G + T + R + P + Reg + Mem all functioning — no "shape choice"); design picks substrate emphasis (git can omit X and P entirely). Fingerprint-distance dynamic range is a structural prediction of the design-vs-evolution axis from Finding 2, surfaced from a different angle.

Document: structural-fingerprint-comparison.md Finding 11.

[Reframed per user correction: the axis is substrate transferability, not design-vs-evolution. Biological substrate is non-transferable — each organism must be individually coherent across the full substrate. Digital substrate is exchangeable — mechanisms transfer between Mns via libraries / protocols / dependencies. Design is downstream of transferability, not upstream. Genetic engineering is the boundary case where biology becomes design-accessible. Beaver dams / termite mounds are the rare biological scaffolding counterexamples. Finding 7's aggregate-as-degenerate is the same axis from a different angle — only exchangeable substrates support aggregate-as-degenerate.**]

Finding 12 — Cognition fingerprint (walked back; the two-mechanism reading remains as hypothesis)

[Walked back.] The initial cognition pass mixed organism-cognition Mns with civilizational-cognition Mns into one "individual cognition" group, and computed cosine distance between zero-vectors (cultural artifacts on cognitive-substrate level) which the script's fallback returned as 1.0. The "biology < cultural < cog < digital" four-quadrant ordering was constructed from numbers measured at DIFFERENT chain levels for different arrangements — not a legitimate cross-comparison.

What is empirically defensible (per chain-level, within-arrangement):

Conceptual hypothesis (NOT a finding yet): two distinct mechanisms can tighten a fingerprint distribution — (a) individual-coherence requirement (biology) and (b) ecosystem-transmissibility selection (cultural artifacts). Both produce tight clustering at the constraint's chain level. Needs proper cross-arrangement-normalized test to elevate to finding.

Document: structural-fingerprint-comparison.md Finding 12.

Finding 13 — Within-arrangement fingerprint spread orders by mechanism transferability

Five arrangements compared, each at its substrate chain level, computed identically:

ArrangementNsubstrate chain levelmean pairwise distance
Biology20biology-substrate0.008
Cultural artifacts (at cultural-ecosystem)47cultural-ecosystem0.023
Cognition organism-cog (at cognitive-substrate)7cognitive-substrate0.16
Entity35entity-system0.32
Methodology13methodology-layer10.52

Caveat: arrangements have different primitive-space dimensions; absolute distances aren't strictly comparable. The ORDERING is meaningful, the ratios are not literal.

Directional reading. Mechanism transferability runs lowest (biology — organisms can't share ribosomes) → highest (methodology — SWOT can't import primitives from OKRs). Spread tracks transferability inversely: low transferability forces individual coherence which forces convergence; high transferability allows specialization which allows divergence. Cultural artifacts are the exception that proves the mechanism — they're highly transferable as patterns, but their shape is selected by cultural-ecosystem transmission requirements, which acts like a different kind of coherence constraint and produces similar tight clustering.

Methodology spread (0.52) being WIDEST is the new evidence. Methodologies have zero transferable mechanisms between each other — each is a self-contained analytical design — and the spread is maximal. This is consistent with the transferability axis.

Document: structural-fingerprint-comparison.md Finding 13.

Finding 23 — Viterbi reveals bridge bottlenecks as contiguous flat shelves

The Viterbi most-likely coherent path (rate-weighted, single-step coherent moves, product of edge probabilities maximized) closes the long-standing gap left by modal-per-rank — which returns the highest-joint-mass position at each rank INDEPENDENTLY and does not produce a connected trajectory. The two views answer different questions: modal-per-rank is the per-rank posterior peak; Viterbi is the single most plausible coherent trajectory from origin to destination.

Five walks Viterbi'd this pass:

WalkRate fileMax rankViterbi total PAgreement with modal-per-rankBridge shelf?
entity-system-min-to-git-substrateentity-system-placeholder95.17e-0240%n/a (single chain level)
entity-system-min-to-git-substrategit-evolution-calibrated99.98e-0270%n/a
abiogenesis-r0-to-r2abiogenesis-calibrated291.16e-0663%YES (ranks 7→17 all in bridge)
biology-substrate-min-to-maxphylogenesis-stem-calibrated65.00e-01100%n/a (single chain level, +1-each destination)
chemistry-to-biology-crossingphylogenesis-stem-calibrated184.02e-0632%YES (chemistry → bridge → substrate as three contiguous phases)
application-architecture-min-to-maxpostgres-evolution-calibrated125.79e-0569%n/a (single chain level)

Both multi-chain-level walks show bridge-traversal as a contiguous block of steps: Viterbi walks the entire bridge chain level in one stretch before resuming substrate-primitive advancement. In chemistry-to-biology-crossing this is canonical — ranks 0–5 advance chemistry primitives (El, Bd, St, Rx, Eq, Kn), ranks 6–11 advance bridge primitives (Cd, Cat, Fb, Fx, Cmp, Gr), ranks 12–17 advance biology-substrate (G, T, R, P, Reg, Mem). Three phases, no interleaving.

In abiogenesis-r0-to-r2, the bridge shelf occupies ranks 7→17 of 29 (~37% of the trajectory). On the abiogenesis-substrate primitive plot this renders as a flat horizontal segment — substrate primitives don't change during bridge traversal, then resume advancement at rank 18.

Modal-per-rank does NOT show this structure. It scatters bridge and substrate primitive advancement across ranks because it averages over many paths reaching each rank. The bottleneck shape that Findings 7/8 described abstractly is now visible as a concrete duration in the rate-weighted single-path view.

This is the visual mechanism behind the bridge-bottleneck claim — and it's specific to Viterbi, not derivable from modal-per-rank.

Compute: compute/lib/rate_corridor.py viterbi_path + viterbi_vs_modal_summary. Results: output/results/<walk>-viterbi.v1.json. Figures: output/figures/backwards-inference-*.png (renderer extended with dashed Viterbi lines).

Finding 24 — Total Viterbi path probability is a corridor-branching diagnostic

The total probability of the Viterbi path (product of edge probabilities along the most-likely coherent trajectory) varies by order of magnitude across walks and meaningfully diagnoses how branching the rate-weighted reachability structure is:

This is independent of corridor width (the count of structurally reachable positions at each rank) — Viterbi P measures the rate-weighted concentration. A wide corridor with one rate-favored path still gives high Viterbi P; a narrow corridor with uniformly-weighted moves gives low Viterbi P.

Also notable: calibrated rates concentrate Viterbi probability vs placeholders. Same walk (entity-system-min-to-git-substrate), same destination, but switching from entity-system-placeholder to git-evolution-calibrated doubles Viterbi P (5.2% → 10%) and raises modal-vs-Viterbi agreement from 40% to 70%. Calibrated rate models reduce path-multiplicity in the rate-weighted view, even though the structural corridor is unchanged.

Diverging-ranks pattern. Where Viterbi disagrees with modal-per-rank tells us where path-multiplicity is highest. Git divergent at ranks 2–7 (the middle of the walk); abiogenesis divergent in the early-mid and late phases; chemistry-to-biology-crossing divergent across most of the walk (32% agreement). The walk's "middle" is consistently where the two views diverge — single-path Viterbi vs mass-averaging modal-per-rank tell different stories about middle-trajectory structure.

Compute: same as Finding 23.

Document: this section + compute/lib/rate_corridor.py docstring.

3. How the findings reinforce each other

3.1 The regime axis is now triple-confirmed

Findings 2, 3, 4 are independent measurements of the same structural axis:

Three independent inputs (pair classifications authored by L1 analysis; rate calibration anchored against empirical milestones; population-context heuristics from literature) all point to the same regime structure. This is the kind of convergent evidence that distinguishes a real structural axis from a data artifact.

3.2 The L3 abstraction surface partitions cleanly

Findings 5 and 6 together describe what kind of cross-arrangement claims hold:

The methodology calls L3 work "abstract shared structure across arrangements." The framework now produces BOTH validations (ecosystem invariant) AND surfaces where claims need nuance (bridge dependency density differs by design strategy). Both forms of result are paper-relevant.

3.3 The design strategy axis bifurcates

Combining Findings 2 and 6:

These are TWO opposite engineering strategies producing surface analogies (both have 6-primitive substrates, both have 12-mechanism bridges, both produce ~9-primitive surfaces and ~9-primitive ecosystems). The L3 analogy holds at the surface count level; below that, the two strategies are nearly opposite. This is itself a major framework-derived observation: the L3 patterns describe convergent surface organization, not convergent internal dependency.

3.4 The unique-contribution claim has a structural signature

Finding 7's "Git ecosystem aggregates everything except P" makes the entity-system's unique contribution concrete: it's not "we have E + I + T + M" (Git ecosystem already provides all of these); it's "we provide them from a single coherent substrate WITH peer-symmetric authority." The Peer primitive is the irreducible gap that requires coherent-substrate design rather than ecosystem accretion.

This connects to Finding 1: cognition's collapsed En/Vr (cognition has Sy carrying both encoder and formal-evaluator roles) explains why cognition is the only substrate where the framework predicts a single primitive doing two SSA roles. Entity-system's unique-contribution signature is the opposite — separating roles that the existing ecosystem can't aggregate together.

4. What the framework can now state about systems generally

Combining the 7 findings gives the framework a set of structural predictions:

About substrate types:

About trajectories:

About ecosystems:

About cross-arrangement coupling:

About what the framework can't yet predict:

5. Architecture of the framework AS DEVELOPED

Per user observation: "we have phases within phases, and also the architecture. We probably need to pull that together."

The framework now has a coherent structural shape that wasn't fully visible piece-by-piece:

                    ┌─────────────────────────────────────────┐
                    │         METHODOLOGY (L1-L4)             │
                    │  defines: primitives, partial levels,   │
                    │  dependencies, pairs, compositions,     │
                    │  arrangements, scope ladder, SSA        │
                    └─────────────────────────────────────────┘
                                       │
                              ┌────────┴────────┐
                              │                 │
                  ┌───────────▼─────┐ ┌─────────▼─────────┐
                  │  ANALYTICAL     │ │  COMPUTATIONAL    │
                  │  CONTROL (L4)   │ │  STACK            │
                  │  Fw/Sc/Mn/Ls/   │ │  12 schemas +     │
                  │  Cpl/Tj         │ │  14 lib modules + │
                  │  + 7 discipline │ │  65 scripts +     │
                  │  rules          │ │  validators       │
                  └───────────┬─────┘ └─────────┬─────────┘
                              │                 │
                              └────────┬────────┘
                                       │
                    ┌──────────────────▼──────────────────┐
                    │            DATA CORPUS              │
                    │  5 arrangements × 24 domains × 11   │
                    │  bridges × 133 manifestations × 14  │
                    │  trajectories × 8 calibrated rates  │
                    │  × 12 population contexts           │
                    └──────────────────┬──────────────────┘
                                       │
                    ┌──────────────────▼──────────────────┐
                    │       STRUCTURAL ANALYSES           │
                    │                                     │
                    │  Sc=1: lattice + corridor + walks   │
                    │  Sc=2: rates + calibration +        │
                    │        population-context wiring    │
                    │  Sc=3: manifestation positioning +  │
                    │        trajectory regimes +         │
                    │        landscape views +            │
                    │        Sc=3 sustained coupling      │
                    │  Sc=4: single-event coupling        │
                    │  L3:   abstractions across          │
                    │        arrangements                 │
                    └──────────────────┬──────────────────┘
                                       │
                    ┌──────────────────▼──────────────────┐
                    │         FINDINGS (7 so far)         │
                    │                                     │
                    │  1. En×Vr pair-load by Kd           │
                    │  2. Substrate pair-density by       │
                    │     design-vs-search                │
                    │  3. Calibration regime decomposition│
                    │  4. Per-transition Nλt regime        │
                    │  5. Ecosystem L3 invariant (~7%)    │
                    │  6. 12-mechanism bridge L3 nuance   │
                    │  7. Aggregate-as-degenerate (P gap) │
                    └──────────────────┬──────────────────┘
                                       │
                    ┌──────────────────▼──────────────────┐
                    │            PAPERS                   │
                    │                                     │
                    │  Tier 1: 0, 1, 2, 10               │
                    │  Tier 2: 5, 6, 7, 9, 11            │
                    │  Tier 3: 3, 4, 8                   │
                    │  Tier 4: 12, 13                    │
                    └─────────────────────────────────────┘

Key relationships not visible from any single piece

What the framework recursively does

Per methodology-advanced-topics.md §4.4: the methodology may instantiate its own SSA topology. The framework IS an information substrate that processes structural understanding:

This recursive structure is one of the framework's signature properties. The methodology landscape (N=13 strategic-analysis methodologies + SPA self-reference) is the empirical manifestation of this recursion.

6. Mapping to papers

Each finding plus the earlier architectural work feeds specific papers. This is the integrated paper-mapping (extending paper-integration.md):

Paper 0 — The Entity System: Six Primitives

Paper 1 — The Entity Core Protocol

Paper 2 — The Entity Church Architecture

Paper 6 — Convergent Evolution of Information Systems ⭐ MOST IMPACTED

Paper 7 — DEOS: Distributed Entity Operating System

Paper 10 — Entity System Security Architecture (still outline-only)

Paper 11 — Structural Methodology ⭐ DIRECT INSTANCE

Paper 12 — Abiogenesis Theory

Paper 13 — Physics as Information Substrate

7. Gaps and where the framework can develop further

The framework has lots of structural pieces in development. Pulling together what's loose:

Foundations that need extending

Synthesis that's been sketched but not pulled through

Computational extensions queued

Domain-specific authoring on HOLD (per user direction — not dropped)

Chain-level coverage asymmetries surfaced (Mn-kind audit)

Per Rule 8 in canonical-architecture-strategy. Each arrangement audited for chain-level coverage consistency:

Paper-shipping work (the largest gap)

8. The composite state

What's been pulled together this session arc:

From the cleanup pass (early):

From the structural surfacings pass (later):

The synthesis (this doc):

What this means for the framework: the model is producing structural observations that match domain intuition (biological cascade vs software modularity; the Peer gap in client-server software; the late-hard bottleneck of abiogenesis; ecosystem co-dependency at ~7%) from independent inputs (lattice structure, pair classification, calibration, population context). The convergent evidence is itself the validation — the framework isn't just describing one structural fact; it's surfacing the same structural axes from multiple analytical angles.

The framework hasn't been finished. It's been developed to a state where it's producing more than the work directly asked of it. The findings keep accumulating from "ask the data the next structural question." There's a lot more to extract.

Document map

MEMORY.md carries the index pointing at this synthesis as the top-level read for the next session.