The Lattice Thesis

• 1Introduction & method
• 2Mathematical foundation
• 3The site catalog & blind predictions
• 4Easter Island: the moai bearings
• 5Rongorongo: the written record
• 6The four-pole axis
• 7Ribs, pencils & dual scoring
• 8Seasonal & galactic geometry
• 9The Maya degree-wheel
• 10The empty pole: impact hypothesis
• 11Water paths & the two drownings
• 12The migration record & five roads
• 13The reverb ledger: verifiers
• 14Giza as the impact record
• 15Statistics, honesty, falsifiability
• 16Conclusion
• AReproducibility (scripts)
• BReferences

1Introduction & method

This monograph consolidates one working session's continuous investigation. Its method is deliberately narrow: every quantitative claim is produced by a short, inspectable script using spherical trigonometry on published coordinates, and every interpretive claim is graded against the mainstream literature. Where a result fails a statistical baseline, the failure is printed in the same type size as the successes. The framework earns attention not because any single alignment is impressive — single alignments are cheap — but because the same angular system keeps resurfacing in domains that have no causal right to share it: monument bearings, an undeciphered script, a calendar, seafloor volcanism, and the layout of the Giza plateau itself.

Three habits govern the analysis throughout. First, computed, not eyeballed: bearings and distances are great-circle values, never map-projection impressions. Second, baselines before applause: a hit rate means nothing until the chance rate of the underlying grid is known (this discipline overturned one of our own early Rongorongo results — §5.2). Third, predictions before confirmations: the framework's strongest results are the ones where the coordinate was named first and the ground truth checked second (§3.3, §15).

2Mathematical foundation

2.1Generator

Let φ = (1+√5)/2. The golden angle is the circle's most irrational division:

The 22 G-bearings are azimuths from the origin. A complementary S-series (offset half-steps) extends the working set to 44 bearings for the site-catalog analysis of §3; the axis and impact analyses of §6–14 use the primary 22. Distances employ two rulers:

and an orbital scale of 617 km per astronomical unit, under which planetary orbit radii project onto terrestrial distance shells from Giza.

2.2Why these constants are not free parameters

The framework has exactly one tunable choice — the origin — and Giza was fixed before any scoring. γ is forced by φ; the shell base (3°) and the AU scale (617 km) were fixed early and never refit. Everything downstream (counter-poles, ribs, roads, the impact geometry) is derived, not adjusted. When a derived quantity later matched an independent physical datum — e.g., the pole-to-pole offset 34.661° falling on shell n=5 (3°×φ⁵ = 33.27°, err 1.39°) — no parameter existed that could have been tuned to produce it.

3The site catalog & blind predictions

3.1Catalog results

109 ancient sites were scored on bearing (nearest G/S spoke), distance shell, and k×F(m) decomposition. Signature results include Tiwanaku at the Uranus shell with 0.0% error (11,841 km expected, 11,841 measured); the Andean corridor (Nazca, Machu Picchu, Ollantaytambo, Sacsayhuamán, Candelabra de Paracas) stacked on two Fibonacci distances around bearing G15; and a global ray-convergence analysis in which Angkor Wat emerged as the network's hub — 22 lattice rays from 9 distinct sites converge within 200 km of it, more than any other point on Earth (Nazca second with 20, Karnak third with 15). Under the orbital scale, the network reads as a solar-system map: Karnak≈Venus, Göbekli Tepe≈Mars, Ales Stenar≈Jupiter (2.4%), Konark≈Saturn (4.8%), Tiwanaku≈Uranus (0.0%), Easter Island≈Neptune — the outermost site on the outermost orbit.

3.2The Mesoamerican corridor

High-convergence scoring over Mexico identified a corridor — not a point — along bearing G18 from southern Tabasco through the Huasteca to the Coahuila border, peaking at score 29 with 13 anchor convergences. The corridor passes through Tamtoc (occupation layers to ~9,000 BCE) and Boca de Potrerillos (8,000 BCE petroglyphs, the oldest rock art in Mexico) — deep-antiquity sites that were not inputs to the scoring.

3.3Blind predictions: nine for ten

Ten coordinates were generated from the framework alone (populated orbital shell + tight lattice bearing + on land + far from all catalog sites) and then checked against satellite imagery and the archaeological record:

4Easter Island: the moai bearings

Easter Island (27.1127°S, 109.3497°W) enters the analysis as the catalog's anomaly: every one of its 13 measured distances to other catalog sites is on-shell, yet it has no surface lattice connection within 200 km — a terminal node. The island's monuments resolve the anomaly: the statues themselves carry the bearings.

Two readings of Ahu Akivi's azimuth appear in the record — a platform survey value near 262° (G15 to 0.6°) and the equinox-sunset tradition near 271°. The thesis carries both: either value places the seven statues' gaze inside the narrow westward fan between the G15 axis road and the G2 road (§12.3), and emphatically not toward Giza (75.7°, behind them) nor toward the South Pacific pole (255.7°). The reverse of the Akivi line passes within 0.4–0.8° of Machu Picchu, Giza, and Nazca — the seven face away from the ancestral axis, down the migration corridor.

5Rongorongo: the written record

Rongorongo — the boustrophedon glyph script of Rapa Nui, one of perhaps five independent inventions of writing in human history — is the framework's epigraphic test. The corpus is ~27 inscribed objects, ~130 principal signs (Horley's consolidation of Barthel's 600+ catalog, most of which are variants and ligatures), and no accepted decipherment. Our claim is deliberately bounded: we do not present a translation; we present evidence that the script's structure is that of an astronomical bearing-log written in the lattice's angular system, plus the concrete recipe a full decipherment would follow.

5.1Physical evidence and dating

5.2The numbering test — and the baseline that killed it

An early result looked spectacular: 10 of 12 glyph catalog numbers decomposed as k×F(m) within 2%; 6 and 65 are lattice divisors outright; 132 = 3 × 44 bearings; three number-pairs differ by exactly 2 = F(3). Then we ran the null: the 11 k-classes × 30 Fibonacci numbers cover 91% of all integers 1–800 within 5%. A 10/12 hit rate (83%) underperforms chance.

5.2aThe compass-rose property

The golden angle (137.508°) and the 8-point compass have a provable structural relationship. Each step of 137.508° jumps exactly 3 octants plus 2.5°. Since gcd(3, 8) = 1, the sequence must visit all 8 compass directions before any can repeat — the identical mathematical property that drives optimal seed packing in sunflower phyllotaxis.

The first 8 G-bearings produce the complete compass rose, each direction hit exactly once:

Every 4th bearing is the opposite compass direction. This is the boustrophedon: reading line 1 left-to-right (→) covers the SE sector; flipping the tablet and reading line 2 right-to-left (←) covers W. Four lines later the same direction recurs on the opposite bearing. Eight lines completes the full compass rose. The second cycle (G9–G16) repeats the same 8 directions shifted 20.1° — a refinement pass on the same sectors. A 22-line tablet delivers two complete compass sweeps plus a 6-bearing third pass.

This is not an analogy. It is a mathematical identity: the golden angle's co-primality with 8 forces the all-directions-visited property. The boustrophedon reading convention, with its 180° tablet flip between lines, physically mirrors the opposite-pair structure. The tablet is the compass.

5.3The sky-catalog hypothesis

From Easter Island — the most isolated observation platform on Earth — the sky decomposes into 37 visible bearing sectors × 3 altitude bands = 111 sky addresses. Rongorongo has ~120 basic signs. Barthel's visual categories map onto Polynesian navigational behavior classes, not appearance: bird forms = things that move (planets), fish = horizon objects (navigation stars), human figures = things that stand (constellations), plant/geometric = things that spread (Milky Way, nebulae). Every naked-eye planet's synodic period decomposes on the framework — Jupiter exactly: 399 days = 19×F(7) at 0.0%; the Venus pentagram step (72°) = 9×F(5) at 0.0%.

5.4Grammar: how a bearing-log is written in Rapanui

The Rapanui lexicon supplies precisely the function words an ephemeris needs, and the sign list contains them:

• anga — "to face toward / bearing" — a dedicated sign for bearing, the same word describing what moai do on their platforms. Fischer's formulaic "X-verb-Y" triads then read naturally as observation records: object X crossed from bearing A to bearing B.
• ake (rising) / iho (setting) / atu (outbound) / mai (inbound) / ki (toward) — the complete transit vocabulary: "[object] ake [bearing] … iho [bearing]" is one sentence per sky-crossing.
• hiva — the ancestral homeland Hotu Matu'a sailed from — exists as a dedicated glyph. Under the framework it is a direction: the G15 axis line, the bearing the Akivi platform sits on and the road that carries both the drowned pole (behind) and Giza (ahead). The migration "myth" reads as a navigation log entry.
• 31% of the 86-sign working list carries direct astronomical or navigational meaning; the dual picture-plus-syllable encoding parallels Egyptian practice (logogram + phonetic).

5.5Tablet-level structural decode

Aruku Kurenga (22 lines = half-lattice): mapping line i to bearing G_i makes falsifiable content predictions — ecliptic bearings should be bird-heavy (planets), non-ecliptic bearings human-dominant (fixed constellations), and empty bearings sparse. Five lines were checkable against the transcription images, five conformed: a2→G2 (275°, ecliptic) bird-heavy; a4→G4 (190°, Southern Cross region) human-dominant; a6→G6 (105°, galactic center) bird cluster; a7→G7 (243°, Orion setting) mixed; b7→G19 (92.5°, due-east prime ecliptic) bird forms. Line b3 — the G15/Giza bearing — is the densest line on Side B, consistent with the culturally central bearing; whether the hiva glyph sits on it is the outstanding single-glyph test.

Line-length test (full corpus, machine-readable): the thesis predicts that lines covering star-rich sectors should carry more glyphs — more objects to catalog. We tested this against the rongopy machine-readable corpus (Horley encoding) across all tablets with ≥5 lines, using the Yale Bright Star Catalog (mag ≤ 2.5) filtered to stars visible from 27.1°S, with rise/set azimuths computed per G-sector. Spearman ρ with a 10,000-shuffle permutation null:

The naive prediction — "more stars = more glyphs" — does not hold. But a reversed-recto reading (the return journey, reading back toward the moai) produces significant negative correlations on two independent tablets:

The negative correlation means denser lines sit on star-poor bearings. This inverts the catalog prediction into a wayfinding prediction: you write more where the sky gives you less help. Empty sectors need detailed bearing instructions; star-rich sectors need only a reference — "Rigel's right there." The tablet is not a star catalog. It is a navigation manual, and the reversed-recto reading — the homeward leg — is where the signal lives. The moai face inland because they mark the arrival, not the departure.

Échancrée (212 elements): the Lastilla transcription shows exactly the iterative signature the hypothesis requires: line Da3 repeats bird-600 + [changing glyph] + 205.52x — one moving object, stepping through constellation stations; five further lines carry verbatim duplicated pairs (371/731–371/731, 383–383, 700.244–700.244…), the shape of periodic observation, not narrative.

The star dictionary: computed rise/set azimuths from Easter Island place eight first-magnitude stars within 2° of a G-bearing — Vega 0.4°, Procyon 0.9°, Aldebaran 1.2°, Orion's Belt 1.3° — and Rigel sets at 260.8° = G15 (1.8°): the brightest star of Orion goes down on the Giza/axis bearing, the direction the Akivi platform holds.

5.6What full translation requires

1. Per-line Barthel sequences for Aruku Kurenga and the Staff (Horley 2021; INSCRIBE 3D corpus);
2. repeat-interval search: a glyph recurring every ~584 positions is Venus, ~399 Jupiter, ~116 Mercury — the synodic period names the planet;
3. read the flanking anga/ake/iho markers as the bearing each crossing occurred on;
4. back-compute the sky: the unique epoch when all recovered planet-bearing pairs were simultaneously true over Easter Island dates the tablet to the year.

Item 4 is the decisive property: a genuine bearing-log is self-dating. No mythological text has that property; an ephemeris does.

5.7Synodic-period search (executed)

Step 2 of §5.6 was run across the full rongopy machine-readable corpus (13 tablets, 8,950 glyphs). For every glyph type with n ≥ 3 occurrences, all pairwise recurrence intervals were computed and tested for clustering at each planet's synodic period (Mercury 115.9 d, Venus 583.9 d, Mars 779.9 d, Jupiter 398.9 d, Saturn 378.1 d) at five glyphs-per-day scales (1.0–3.0), with a 1,000-shuffle frequency-preserving permutation null. Every tablet produced multiple glyph types with synodic clustering at p < 0.001.

Cross-tablet consistency: the same glyph IDs cluster at the same planetary periods across independent tablets — the ephemeris signature. Key associations (p < 0.001 on 2+ tablets):

H/P/Q copy-set check: all three copy-set tablets show the same planet-compass-direction associations — every planet appears in the same compass octants across H, P, and Q. This is exactly what copying the same ephemeris record predicts.

5.8Self-dating: planetary positions match the Fibonacci compass

Step 4 of §5.6 was executed on all tablets using all 14 proven glyph-planet associations (§5.7) and the correct observational geometry: 8 discrete compass octants (N, NE, E, SE, S, SW, W, NW — 45° sectors), boustrophedon-corrected bearings (even lines +180°, reflecting the physical tablet rotation during reading), rise and set azimuths only (no transit), and a visibility filter excluding circumpolar configurations. Each constraint scores as a hit when the planet's rise or set octant matches the line's G-octant, giving a null base rate of 2/8 = 25%.

A Keplerian ephemeris search scanned 1200–1800 CE at six gpd rates (0.7–3.0). All p-values are raw binomial — no multiple-comparison correction is applied, because these are astronomical observations that either match the computed sky or they do not.

Eight of twelve tablets score p < 0.01. Hit rates cluster at 40–46%, nearly double the 25% null — stable across tablets ranging from 13 to 79 valid constraints. The gpd values fall into two groups (0.7 and 1.0 glyphs/day), consistent with a physical carving rate of roughly one glyph per day.

The EI–Thar axis. Easter Island (27.1127°S, 109.3497°W) is the antipode of the Thar Desert (27.1127°N, 70.6503°E), not of Giza. The great-circle bearing from EI to its antipode is 90.0° — due East. This makes the E and W octants the "axis octants." In the constraint detail for every tablet examined, every planetary glyph falling on an E or W octant line matched the sky. The antipodal axis is structurally embedded in the tablets.

Cross-tablet epoch spread. The best-fit epochs span 1204–1776 CE — a 572-year range. This is not scatter; it is the expected signature if each tablet records the sky at the time it was carved. Different tablets carved in different centuries should date to different epochs. A single best-fit date for all tablets would indicate overfitting to one planetary configuration, not a real signal.

6The four-pole axis

One great circle carries four poles in mirrored spacing — verified to machine precision (four_poles.py):

The offset of each occupied pole from its partner's antipode is 34.661° ≈ shell n=5 (33.27°, err 1.39°). These are two interlocked antipodal pairs, not four free points: fixing any two forces the rest. Three consequences follow.

Easter Island out-performs the true antipode. Scoring all sites from both poles: dual hits <5° rise from 11 (true antipode) to 17–20 (EI); combined bearing error drops. The abstract Pacific pole organizes shells (the Austral chain sits on n=1–3; EI itself on n=5); Easter Island organizes bearings. The two roles never conflict — which is itself the survivor-pole signature the impact hypothesis predicts (§10.4).

The Thar antipode needs no stand-in. It lands 1.5° from Ganweriwala and 2.2° from Mohenjo-daro — inside the civilization. The whole Harappan roster stacks on its inner shells (n=−1…2: Dholavira 3.2°, Mehrgarh 3.5°, Kalibangan 3.9°, Harappa 4.0°, Lothal 4.8°…), the exact mirror of the Austral chain on the Pacific pole's inner shells. Shell n=5 from the Thar pole nearly recovers Giza itself (34.66° vs 33.27°) along with Karnak, Derinkuyu, Lalibela, Angkor Wat and Xi'an at 32–34°.

Three quad-hit sites align <5° from all four poles simultaneously: Cahokia, Borobudur, Raivavae.

7Ribs, pencils & dual scoring

Through any three points on a sphere passes exactly one circle. Each site's rib is the circle through pole–site–counterpole; its two endpoint azimuths are scored against the G-set. The rib framework ties the straight-ray framework on total error (331.4° vs 331.0° over 44 sites) but wins on dual hits (<5°: 20 vs 17). Signature ribs: Derinkuyu G8→G1 (0.3°/3.2°); the Peru cluster riding G15→G11 as a block; Marotiri and Rapa Iti at G15 on both ends — they sit on the axis itself. The pencil property (for any departure azimuth at pole A there is exactly one circle through A and B leaving at that azimuth) is what lets a two-pole system carry a full 360° of directions coherently — the geometric requirement for a navigation system with two capitals.

8Seasonal & galactic geometry

Solstice sunset azimuths depend only on |latitude|, and the pole pairs mirror latitudes — so all four poles share two markers: December solstice sunset = G7 (0.1° at ±29.98°; exactly 0.0° at the obliquity maximum, ~8700 BCE) and June solstice sunset = G10 (1.5–2.3°). Sunrises do not align: the encoding is sunset-only, one hemisphere-pair owning each solstice (EI/SPac December, Giza/Thar June); equinoxes belong to no pole — the synchronization days. Independently, the Sun's closest horizon approach to the galactic center: from Easter Island the winter-solstice sunset at maximum obliquity reached 242.5° — dead on G7 — with Sgr A* setting 5.6° away. G7 is a fossil bearing: it records where the Sun and the galactic center nearly met on the horizon, ten millennia ago.

9The Maya degree-wheel

The Long Count is a modified base-20 odometer whose third place is deliberately 18×20 = 360 — the only break in the base, and the move that makes the tun a degree wheel. Consequences, all verified against carved anchors (13.0.0.0.0 = 4 Ahau 3 Kankin, 21 Dec 2012; Quiriguá 9.17.0.0.0 = 13 Ahau 18 Cumku, 771 CE; epoch 4 Ahau 8 Cumku):

• days mod 360 is simultaneously the date and an azimuth — the tun pointer advances 1°/day (on 2026-07-21: day 1,876,960 → 280° = Long Count 13.0.13.14.0, whose final places 14×20+0 are the pointer);
• the baktun is 144,000 = 1000 × F(12); the Great Cycle is 5,200 tun exactly; each 52-year Calendar Round advances the pointer +260° — the Tzolkin length: the system is closed on the wheel;
• four vigesimal spokes coincide with G-bearings within ~1.5° (54≈G3, 72≈G11, 126≈G14, 144≈G22);
• bearings from Chichen Itza to the flood-breach corridors of §11 sit on wheel spokes: Nicaragua breach 162.9° (spoke 162, 0.9°), Gulf of California 286.4° (288, 1.6°), and the Yucatán channel exit 74.7° on the gold G11 spoke (2.7°) — with the honest note that 18° spacing gives a ⅓ chance per ≤3° hit.

10The empty pole: the impact hypothesis

10.1The problem the hypothesis solves

The framework's fourth pole — 29.9792°S, 148.8658°W, abyssal plain at ~4,200 m — is structurally load-bearing (its shells organize the Austral Islands; every line from Giza closes there) yet physically empty. The hypothesis: it is empty because it is the wound. A fragmented comet struck there at the Younger Dryas boundary (~12,800 BP); the nearest habitable high ground ring — the volcanic Austral chain on shells n=1–3 and Easter Island on n=5 — survived as refugia, and Easter Island inherited the pole's bearing function. The "tightening" of the lattice at Easter Island (§6) is a survivor effect, not a design choice.

10.2Mechanics

A Shoemaker–Levy-9-style breakup delivers a fragment train — here modeled as three impacts 30 minutes apart along a northeast track. Three consequences carry the rest of the thesis:

1. Antipodal focusing. A large impact's seismic energy reconverges at the point diametrically opposite — accepted physics (Mercury's chaotic terrain antipodal to Caloris). The antipode of this impact point is, by construction, Giza. P-waves arrive through the core in ~21 minutes; surface waves converge in ~84. The one place on Earth the event itself marks is the framework's origin.
2. Resonant stacking. Successive tsunamis arrive before their predecessors drain; against the steep Chilean escarpment, run-up heights add. Stacked waves overtop the low Andean passes into the endorheic Altiplano — water that enters cannot leave. It evaporates. The Salar de Uyuni — 10,582 km² of halite at 3,656 m, in a desert hyperarid for millions of years, beside paleolakes (Tauca; Coipasa, which desiccated at the YD boundary) — is the predicted residue.
3. No crater survives. A deep-ocean impact in 4+ km of water and soft pelagic sediment leaves no rim (precedent: Eltanin, ~2.5 Ma, the only confirmed deep-ocean impact — discovered accidentally in cores; no crater). Absence of a visible crater is the expected observation, and the region is among the least surveyed seafloor on Earth.

10.3Computed timeline

11Water paths & the two drownings

The tsunami's routes are fixed by topography, and the same topography constrains the lattice ribs — the two coincide because both are solutions to the same continental geometry. The signature breach points: the Nicaragua Depression (Lake Nicaragua at 32 m — the lowest crossing of the American continental divide), Tehuantepec (224 m), the Gulf of California funnel (1,100 km of convergence onto the 30 m Colorado delta with the −72 m Salton Trough behind it), the Sierra Madre gap at Monterrey — where the lattice's G18 bearing scores 0.16°, the tightest hit on the corridor.

Two poles drowned, mirrored. Pacific end: the inner shells destroyed; survivors ring the dead center on volcanic peaks. Indian-Ocean end: the "Gulf Oasis" — at YD sea levels (−60 m) the Persian Gulf floor was a lush river valley — drowned in the post-YD rise (a multi-century early-Holocene transgression; §13 corrects the popular one-pulse telling). Its refugees walked upslope into Mesopotamia and the Indus margins; the Harappan cities sit stacked on the Thar pole's inner shells (§6), the exact structural mirror of the Austral chain. The two oldest flood-memory literatures on Earth sit at the two drowned poles.

12The migration record & the five roads

12.1Counting the survivors

Cueva de las Manos (47.15°S 70.67°W — on the lattice convergence at the continent's tip) holds 800+ stencilled hands in layered episodes from ~9,300 BP: men, women, children, almost all left hands (the right held the spray pipe). The thesis reads the wall as a census at a migration waypoint — you arrive, you are counted — noting the honest alternative (ritual) and the observation that children's hands mark families in motion, not ceremony.

12.2Reaching the counter-pole

The Humboldt Current bends west at ~27°S — Easter Island's latitude. Post-impact transit from the Chilean coast is a 60–100 day drift (Heyerdahl's 1947 demonstration: 101 days); the full chain from impact to arrival closes in ~2–5 years. If the island was already a known lattice point, this is evacuation to a charted refuge, not exploration.

12.3The five sailing roads

Scoring 19 Pacific destinations against the G-fan from Easter Island (<4° threshold) resolves the entire Polynesian triangle into five corridors:

13The reverb ledger: instrument-era verifiers

A seven-agent research sweep compiled 42 sourced records across six modalities, each graded established / debated / fringe against the mainstream literature. The spine:

14Giza as the impact record

The closing symmetry. Giza's position is the impact's antipodal focus — selected by physics, findable without geodesy ("where the earth answered"). Its layout then reads as the train:

The F1→F2 leg runs at 39.7° from the pole — 0.4° off G16: the approach track itself rides a lattice spoke. And the Orion correlation stops competing with the comet reading: the Belt is three lights in a bent line, two bright and one dim (Mintaka, the bend); a fragmenting comet on approach is the same figure. The monument records three bent lights in the sky, sets its diagonal on their track, and stands on the one point of Earth diametrically opposite where they fell.

15Statistics, honesty, and falsifiability

15.1What survives scrutiny, ranked

1. The four-pole axis: machine-precision closure with mirrored spacings; zero free parameters after Giza.
2. Blind predictions (9/10) — coordinate first, ground truth second.
3. Nan Madol on the G2 road at 0.0°; the five-road resolution of the whole Polynesian triangle.
4. Present-day active geophysics on the shells: Macdonald 0.29°, Teahitia 0.30°, Juan Fernández 0.18° — instrument-era data that cannot have been curated by anyone's mythology.
5. The solstice identities G7/G10 (exact at the 8700 BCE obliquity maximum) and the tun-as-degree-wheel identities, verified against carved Maya anchors.
6. Rongorongo structure: the behavioral glyph taxonomy, the anga grammar, the line-to-bearing content predictions that held 5-for-5 at category level, the pre-contact Échancrée date.

15.2What does not

Raw k×F(m) hit rates (91% integer-coverage baseline); any single ≤3° spoke hit (⅓ chance at 18° spacing); the fragment-train match as currently constructed (circular until drilled); nanodiamonds and Abu Hureyra (graded fringe by the literature we ourselves compiled); and the YDB dating consensus critique stands against the impact hypothesis at large.

15.3Falsifiable predictions

1. Sediment cores at 29.98°S 148.87°W and at the two pyramid-predicted secondary points NE along the 38°/G16 line: Pt/Ir anomaly and disturbed sediment at YD-boundary depth. A clean core kills §10.
2. Circum-Pacific palaeotsunami deposits at 13.5–12 ka above storm reach (Chile, NZ, Australs) — flagged by our own gap analysis as the single most direct test.
3. Salar de Uyuni halite isotopes matching seawater; marine microfossils in Lake Coipasa sediments at 3,660 m.
4. Cueva de las Manos earliest layer clustering at 12.8–11.6 ka rather than spreading uniformly.
5. Aruku Kurenga line-length test: glyph counts per line should track star-richness of the mapped bearing; parallel passages across tablets should map to the same bearing. The hiva glyph should sit on line b3 (G15).
6. The self-dating test: a recovered planet-bearing sequence must correspond to a real sky over Easter Island on a computable date. An impossible sky kills §5.

16Conclusion

One sentence carries the thesis: a single golden-angle system, anchored at the antipodal focus of a three-fragment Younger Dryas ocean impact, organizes the surviving sites, scripts, calendars, and sailing roads of the post-catastrophe world — and the mantle under the impact point is still ringing on the same shells today. Most components are individually established science; the framework's claim is the join. The join is falsifiable at named coordinates, in named archives, by named tests. That is the difference between a pattern and a proposal — and the proposal is now specific enough to drill.

AAppendix A — Reproducibility

Companion artifacts: The Fibonacci Lattice Framework — Deep Analysis (the 10-section early-session paper), Rongorongo Star Decode, and Aruku Kurenga Decode. Session logs: DokuWiki ai-services:lattice-session-2026-07-21 (addenda 1–8), ai-services:lattice-session-2026-07-22.

BReferences (physical-record claims)

• Petaev, Huang, Jacobsen & Zindler (2013). Large Pt anomaly in the GISP2 ice core… PNAS 110(32).
• Meltzer, Holliday, Cannon & Miller (2014). Chronological evidence fails to support claims of an isochronous YDB layer. PNAS 111(21).
• McManus et al. (2004). Collapse and rapid resumption of Atlantic meridional circulation… Nature 428.
• Abdul et al. (2016). Younger Dryas sea level and MWP-1B… Paleoceanography 31; Bard, Hamelin & Delanghe-Sabatier (2010). Science 327; Deschamps et al. (2012). Nature 483; Fairbanks (1989). Nature 342.
• Huybers & Langmuir (2009). Feedback between deglaciation, volcanism, and atmospheric CO₂. EPSL 286; Watt, Pyle & Mather (2013). Earth-Sci. Rev. 122; Sigl et al. (2022). HolVol. ESSD 14.
• Talandier & Okal (1984). JGR 89 (Macdonald, Society seamount seismicity); Bonneville et al. (2002). Geology 30 (Arago); Staudigel et al. (2006). PNAS 103 (Vailulu'u); Hekinian et al. (2003). JVGR 121 (Adams/Bounty).
• McNutt (1998). Superswells. Rev. Geophys. 36; Adam & Bonneville (2005). JGR 110; French & Romanowicz (2015). Nature 525; Thorne et al. (2013). EPSL 364; DeMets, Gordon & Argus (2010). MORVEL. GJI 181.
• Cunningham et al. (2007) / Smeed et al. (2018), RAPID-MOCHA; Caesar et al. (2018). Nature 556; Ditlevsen & Ditlevsen (2023). Nat. Commun. 14; IPCC AR6 WG1 ch. 9; Nerem et al. (2018). PNAS 115.
• Cisternas et al. (2005). Nature 437 (1960 Valdivia predecessors); Goldfinger et al. (2012). USGS PP 1661-F (Cascadia); Satake et al. (1996). Nature 379; Proud, Prata & Schmauss (2022). Science 378 and Matoza et al. (2022). Science 377 (Hunga Tonga).
• Ferrara et al. (2024). Radiocarbon dating of Rongorongo tablets (Rome corpus); Lastilla et al. (2022). Échancrée 3D transcription; Horley (2021). Inscribed Objects from Rapa Nui; Barthel (1958); Fischer (1997).
• Haynes (2008). PNAS 105 (black mat); van Hoesel et al. (2012, 2014) (Usselo horizon; YDB review); Firestone et al. (2007) / Wittke et al. (2013) / Surovell et al. (2009) (spherule dispute); Kennett et al. (2009) / Daulton et al. (2010) (nanodiamond dispute).