Sator Research Program

"The Sator Square does not reverse physical entropy; it creates a symbolic subspace where directional entropy is minimized by geometric redundancy."

---

Overview

This repository documents the Sator Research Program, a formal investigation of the Sator Square as a mathematical information structure, integrated into the TAMESIS / TRI / TDTR theoretical framework developed by Douglas H. M. Fulber.

The central objective is to transform the Sator Square from a linguistic curiosity into a rigorous mathematical object. Every claim is classified as either a proved theorem, a computational result, or a modelling choice — following the epistemological standards of the Clay Mathematician and Senior Theoretical Physicist personas.

What this research establishes (defensible without qualification):

A class of linguistic structures with maximal symmetry that induces information redundancy sufficient for partial error recovery behaviour.

---

Key Results at a Glance

Result	Value	Status
Symmetry group	$G \cong \mathbb{Z}_2 \times \mathbb{Z}_2$	Theorem (proved, EXP-06)
Degrees of freedom	$25 \to 9$ orbits	Proved (orbit counting)
Compression factor	$26^{-16} \approx 2.29 \times 10^{-23}$	Derived analytically
$\Delta H$ conditional	$H(M) - H(M|G) = 16 \cdot \log_2 26 \approx 75.2$ bits	Proved (EXP-07)
$|\Delta H_{\text{dir}}|$	$< 6.40 \times 10^{-16}$ bits $(\varepsilon_{\text{machine}})$	Computational (not exactly 0)
Monte Carlo rarity	0 hits / 500,000 trials	Empirical confirmation
$d_{\min}^{\text{orb}}$ (non-trivial)	2 (orbit of size 2)	Computed (EXP-07)
Upper bound $|\Omega''|$	$\leq |L_5^{\text{rev}}|^2 \cdot |L_5^{\text{pal}}|$	Proved (EXP-07)
Portuguese density	40 / 385 = 0.104	Empirical (curated lexicon)
Recovery rate ($t=1$)	$98.8% \pm 1.8%$	Empirical (EXP-05)

---

Theoretical Framework

Persona	Role	Source
Foundational Architect	Evaluates coherence of new mathematical objects	ajuste_fino/ajustefino-fund.md
Clay Mathematician	Absolute axiomatic rigor — no heuristic as proof	ajuste_fino/ajustefino-clay.md
Senior Theoretical Physicist	Physical consistency and regime boundaries	ajuste_fino/ajustefino-fisic.md

---

Experimental Results

EXP-01: Formal Validation

All symmetry constraints verified computationally (0 errors across 25 positions). The matrix $M$ primary constraints are $M = M^T$ (Word Square) and $M_{ij} = M_{4-i,4-j}$ (Central Symmetry). The Palindromic Center ($M_{2j} = M_{2,4-j}$) follows as a direct corollary of WS and CS.

---

EXP-02: Shannon Entropy Analysis

Directional entropy $H_d = 2.8839$ bits is equal across all four reading directions. The directional difference is bounded — not claimed to be exactly zero but bounded by machine epsilon:

$$|\Delta H_{\text{dir}}| < \varepsilon \approx 6.40 \times 10^{-16} \text{ bits}$$

Entropy Reduction Sequence:

---

EXP-03: Computational Rarity Proof

Monte Carlo ($N = 500{,}000$): zero word squares found. This is empirical confirmation of the analytically derived bound $p = |\Sigma|^{-16}$ — the simulation is not the origin of the probability, only its validator.

$$p = \frac{|\Sigma|^9}{|\Sigma|^{25}} = |\Sigma|^{-16} \approx 2.29 \times 10^{-23}$$

---

EXP-04: Multi-Language Generator

Portuguese curated lexicon ($|L_5| = 96$) yielded 40 valid structures. Density normalized by the structural search space:

$$\text{density} = \frac{|\Omega''|}{|L_5^{\text{rev}}| \cdot |L_5^{\text{pal}}|} = \frac{40}{385} \approx 0.104$$

---

EXP-05: Tamesis Bridge (Isfet / Ma'at)

The symmetry constraints act as implicit redundancy. Under single-position corruption, the orbit-based majority vote achieves $98.8% \pm 1.8%$ recovery. This is not a classical error-correcting code (no finite field, no linear structure), but a symmetry-constrained symbolic structure with effective rate $R = k/n = 9/25 = 0.36$.

Isfet / Ma'at Cycle:

---

EXP-06: Formal Proofs — Klein Group & CSP Bounds

The group $G = {e, T, R, TR}$ under composition satisfies all five Klein group axioms (closure, associativity, identity, self-inverse, commutativity) — verified computationally over all $4^3 = 64$ triples and confirmed by Cayley table. The fixed-point space has dimension 9:

$$M \in \text{Fix}(G) \iff \dim(\text{Fix}(G)) = 9$$

CSP backtracking over $|\Sigma|=3$ confirmed: exactly $3^9 = 19{,}683$ consistent matrices — no hidden constraints beyond the three formal ones.

---

EXP-07: Paper Completion — d_min, Upper Bound, Conditional Entropy

Pillar 1 — Minimum Orbit Distance $d_{\min}^{\text{orb}}$: Orbit sizes are ${1, 2, 2, 2, 2, 4, 4, 4, 4}$. This is not a Hamming minimum distance of a classical linear code, but an orbit-induced symbolic distance. Changing a single orbit changes exactly $|O_i|$ matrix positions simultaneously, so $d_{\min}^{\text{orb}} = 2$. Empirically confirmed over 5,000 random symmetric matrix pairs.

Pillar 2 — Formal Upper Bound (proved):

$$|\Omega''| \leq |L_5^{\text{rev}}|^2 \cdot |L_5^{\text{pal}}|$$

Proof: any Sator-like square relies geometrically on selections for rows 1, 2, and 3. Bounding the possible options conservatively, the selection relies on forming $w_1, w_2 \in L_5^{\text{rev}}$ and $w_3 \in L_5^{\text{pal}}$. A tighter bound requires the cross-constraint $C(w_1, w_2, w_3)=1$ to be satisfied. $\blacksquare$

Pillar 3 — Full Conditional Entropy:

$$H(M) - H(M \mid G) = 16 \cdot \log_2 26 \approx 75.21 \text{ bits}$$

This is a proved identity, not a measurement: it follows directly from the orbit structure, independent of any specific matrix values.

---

Project Structure

sator_research/
├── index.html              (paper HTML — Tamesis standard)
├── readme.md
├── checklist.md
├── roadmap.md
├── EXECUTION_REPORT.md
│
├── simulations/
│   ├── 01_sator_validator.py
│   ├── 02_entropy_analysis.py
│   ├── 03_rarity_proof.py
│   ├── 04_sator_generator.py
│   ├── 05_tamesis_bridge.py
│   ├── 06_formal_proofs.py
│   └── 07_paper_completion.py
│
├── figures/
│   ├── animations/
│   ├── entropy_maps/
│   ├── formal/
│   ├── rarity/
│   ├── symmetry_viz/
│   └── tamesis_bridge/
│
├── results/
│   ├── entropy/
│   ├── formal/
│   ├── rarity/
│   └── squares/
│
└── ajuste_fino/