Architecture

Parent-Layer Arithmetic

The fundamental unit is the parent layer (N): the number of generation intervals separating a past event from the present. Given a date T years before present and a mean generation interval g:

N = T / g

The population-weighted mean generation interval from Wang et al. (2023) is g = 26.9 years (95% range: ~25–30 years). Key sex-specific values: paternal g ≈ 30.7 years; maternal g ≈ 23.2 years. For patrilineal or matrilineal transmission chains (certain songlines, craft specialisations), this divergence is material. The protocol treats g as a population-level average; future lineage-specific scoring may require optional g_maternal / g_paternal parameters.

Three-Level Transcendence Hierarchy

Creative traditions are assessed across three levels of increasing abstraction. Each level can be scored independently; higher levels require evidence of lower levels (the cascade rule).

Level 1 — Material Transcendence

Physical artefacts or marks survive across N > 40 parent layers. This is the easiest to assess: dated objects exist; stratigraphy anchors them. Examples: Bizmoune shell beads, Blombos engraved ochre, Sulawesi hand stencils, Sahul ground-edge axes.

Level 2 — Practice Transcendence

Not just the objects but the techniques for making them persist across N > 40. Evidence: recurring production methods visible in the archaeological record (e.g. consistent knapping sequences, pigment preparation techniques, instrument construction methods) or confirmed by living practitioners.

Level 3 — Semantic Transcendence

The meaning system associated with the tradition persists across N > 40. This is the hardest to assess and the most consequential. Evidence: recurring narrative motifs across temporally separated panels, custodial oral traditions linking current practice to ancient sites, ritual frameworks whose structure maps onto archaeological sequences.

Dual Evidence Channels

Stratigraphic Channel (Strat)

Archaeological dating, stratigraphy, radiometric analysis, art-science techniques (LA-ICP-MS U-series, OSL, radiocarbon). Anchored in physical measurement. Available for all sites but limited to what survives in the material record.

Custodial Channel (Cust)

Oral traditions, living practice, ceremonial continuity, community knowledge maintained by Traditional Owners or cultural custodians. Has internal falsification mechanisms: cross-clan corroboration, ritual sanctions for false claims, multi-elder verification. Available primarily for Australian sites and some other living traditions.

Custodial Null (defined in v0.3)

If no living community claims custodianship and no oral tradition references a site, the custodial channel is scored “—” (absent). This is distinct from RUPTURED. If contemporary Traditional Owners explicitly disavow connection to a site (not merely unknown, but actively severed), this constitutes Custodial RUPTURE despite material continuity.

Scoring Protocol

Continuity Threshold

The discriminant threshold is N > 40 (~1,000 years at g = 26.9). This is the boundary between fashion (transmittable within living memory networks) and tradition (requiring institutional, ritual, or material infrastructure for persistence).

Validation: all securely scored traditions in the current compilation are either far above the threshold (Huenul 1 comb motif ~120 layers; Kimberley ~692 layers; Nawarla Gabarnmang ~1,120 layers for dated art; Sahul axes ~1,416 layers) or far below it. No entry currently sits near the boundary, reducing sensitivity to the exact threshold value.

Scale Separation

• Tradition — identifiable transmission linkage at a specific location or within a specific community. Scored per-site.
• Category — species-wide behavioural capacities (mortuary practice, body ornamentation, music-making). Capacity continuities, not single transmission chains.

Scale separation prevents over-claiming. The fact that humans have always ornamented themselves does not mean any single ornamental tradition is continuous.

Rupture Taxonomy

Dating Revision Protocol

When new dating evidence emerges, the protocol asks three questions:

1. Does the new date cause N to cross the threshold (N = 40) in either direction?
2. Does the new date require species reassignment (e.g. from H. sapiens to Neanderthal)?
3. Does the new date invalidate the evidence chain for any scored level?

If all three answers are “no,” the classification is stable and only the numerical N-value updates. The worked Sulawesi example (v0.3 §2.6) demonstrates this resilience: Leang Karampuang (≥51,200 ya, N ≈ 2,048) and Liang Metanduno (≥67,800 ya, N ≈ 2,712) both remain far above threshold. Classification stable.