The Berrett Model: Statistical Evaluation of a Rule-based Syllabic-style Transformation System for Rongorongo Glyph Data

Description: This paper presents the Berrett Model, a rule-based, deterministic transformation framework for the analysis of Rongorongo glyph sequences. The study applies a syllabic-style mapping system to glyph data and evaluates resulting structures using statistical comparison against permutation-based null models. The methodology integrates a fixed lexical mapping procedure (M₁), equivalence-class classification rules, and a controlled falsification protocol based on scrambled corpus generation. Statistical evaluation is conducted using lexical enrichment measures, signal separation metrics, and a null-model-based enrichment framework (LSAET v1. 0; Annexure F). The framework is designed to be fully reproducible and operates without semantic inference or linguistic interpretation of glyph meaning. Instead, it tests whether predefined lexical sets and anchor patterns exhibit statistically significant clustering under deterministic transformation rules compared to randomized baselines. Results demonstrate measurable divergence between authentic and permuted datasets under identical processing conditions, including differences in lexical match rates and structured tag density. The study does not claim linguistic decipherment or translation of Rongorongo, but instead provides a falsifiable computational framework for evaluating structural regularities in glyph sequence organisation. All transformation rules, datasets, and statistical procedures are fully documented to support reproducibility and independent verification.