Reproducibility Materials for Legal-Conceptual Equivalence in Literary Translation: Italian Criminal-Procedure Terminology in Polish, English and French Versions of Testimone inconsapevole

This repository contains copyright-safe reproducibility materials for a multilingual parallel-corpus study of Italian criminal-law and criminal-procedure terminology in the Polish, English and French translations of Gianrico Carofiglio’s Testimone inconsapevole. The underlying research corpus comprised 229 manually aligned Italian–Polish–English–French source segments. Application of a frozen controlled inventory of 96 Italian legal terms and legally salient procedural phrases identified 374 analytical term or phrase records distributed across 190 source segments. Ninety-one inventory categories occurred in the corpus. Each analytical occurrence was evaluated in the three target translations in terms of translation strategy, legal-conceptual equivalence, conceptual or procedural shift, severity and legal-system orientation. The public researcher-validated dataset contains all 374 final analytical records. Generative AI was used for structured annotation assistance and research-support functions; AI-generated classifications were treated as preliminary drafts rather than independent coder judgments. All analytical records were subsequently reviewed and accepted by the researcher. The repository contains the frozen controlled extraction inventory, coding framework, copyright-safe researcher-validated derived dataset, post-coding outputs, manuscript Tables 1–5, corpus bibliographic metadata, software requirements and two Python analysis scripts. The public analysis script reproduces the reported descriptive profiles, figures and paired statistical tests directly from the copyright-safe dataset. The aligned literary passages themselves are not redistributed. The Italian novel and its published Polish, English and French translations are protected by copyright. The public dataset therefore excludes continuous source and translation passages, matched textual passages and exact target-language renderings. Researchers wishing to reconstruct the textual corpus must obtain lawful access to the relevant published editions independently. The statistical comparisons are exploratory within-text comparisons of three translations of a single literary work and should not be interpreted as population-level estimates of Polish, English or French literary translation generally.