ArmEpiC – Armenian Epigraphic Corpus (ArtsakhEpiC Sub-Corpus).

ArmEpiC: Methodology and Data Description ArmEpiC, Armenian Epigraphic Corpus (ArtsakhEpiC Sub-Corpus), is a digital scholarly edition of ten Armenian lapidary inscriptions from Artsakh, diplomatically transcribed and encoded in TEI/EpiDoc (v9.7), together with the controlled-vocabulary authority files needed to interpret them. ArtsakhEpiC is the Artsakh regional sub-corpus of the wider ArmEpiC (Armenian Epigraphic Corpus) initiative, hosted by the EPFL Digital Humanities Laboratory (DHLAB). Scope of the dataset. The deposit includes ten TEI/EpiDoc inscription files (ART0001–ART0010), each a standalone document with diplomatic transcription, translations in Armenian and English, physical description, dating, and bibliography; twelve authority files covering places and toponyms (ListPlace), monuments and monument parts (ListMonument), preservation and condition states (ListPreservation, cross-referenced to the EAGLE preservation-state vocabulary and Getty AAT), materials (ListMaterial), scripts and hands (ListScript), techniques (ListTechniques), object types (ListObjectType), inscription functional types (ListInscriptionType), abbreviation types (ListAbbreviationType), chronology and dating methods (ListChronology), bibliography (ListBibl), and editors and contributors (ListEditors); ten photographic figures (Fig. 1–Fig. 10, one per inscription); a README; and a licensing statement. Standards and encoding. All XML files are TEI/EpiDoc 9.7, associated via <?xml-model?> processing instructions with the schema at https://epidoc.stoa.org/schema/9.7/tei-epidoc.rng (both RelaxNG structure and embedded Schematron rules). Authority files are aligned with CIDOC CRM and, where applicable, cross-referenced to EAGLE and the Getty Art & Architecture Thesaurus. Each authority entity carries a persistent URN of the form urn:armepic:<vocabulary>:<ID>, and every inscription's editorial responsibility statement references the corresponding editor URN in ListEditors. Editors and contributors. Hamest Tamrazyan (EPFL DHLAB) — editor-in-chief: project direction, template development, EpiDoc adaptation, terminology and SKOS modelling, authority-file creation. Gayane Hovhannisyan (European University of Armenia) — editor: translation of texts and terminology, EpiDoc adaptation, validation of terminology and SKOS concepts. Arsen Harutyunyan (Institute of Archaeology and Ethnography, NAS RA) — editor: epigraphic research and provision of inscriptional data, EpiDoc adaptation, photographic data collection. Emanuela Boros (EPFL DHLAB) — data curator: technical implementation support, data curation, project co-supervision. Emile Cornamusaz (EPFL) — contributor: technical implementation and automation. Licensing and versioning. The dataset is released under a Creative Commons Attribution–NonCommercial 4.0 International (CC BY-NC 4.0) license; reuse for research and teaching is welcome with attribution, and commercial use requires the editors' permission. This Zenodo deposit represents a fixed release; future revisions will receive new versions under the same concept DOI. Funding. This project has been funded by the National Association for Armenian Studies and Research (NAASR), the Knights of Vartan Fund for Armenian Studies, and the Swiss National Science Foundation (SNSF, grants IZSEZ0_212008 and IZSEZ0_216747). What changed in this version. This release corrects a broad set of issues found in the previously published deposit: four inscription files that were not well-formed XML; the editor authority file, rebuilt as a valid, schema-associated TEI document with persistent URNs matching those already referenced by the inscriptions; a license inconsistency across files, now unified to CC BY-NC 4.0 throughout; a missing ListPreservation authority file, now added, resolving preservation-state references used by six of the ten inscriptions; a dead schema URL in all ten inscription files, repointed to the working one already used by the authority files; and filename/asset naming inconsistencies. All 22 XML files were additionally validated against the live EpiDoc 9.7 RelaxNG schema, surfacing and fixing further schema violations across the inscription files (invalid <license> spelling, duplicate single-occurrence elements, misplaced header elements, an invalid content-model mixture in the source description, out-of-order manuscript-identifier elements, and a misplaced pointing attribute) and one authority file (a header ordering error). One inscription record, ART0009, additionally had its title and inscription-type classification corrected: it had been copy-pasted from another record and mislabeled the wrong monument (Gandzasar instead of Dadivank). Citation. Tamrazyan, H., Hovhannisyan, G., Harutyunyan, A., & Boros, E. ArmEpiC – Armenian Epigraphic Corpus (ArtsakhEpiC Sub-Corpus). Zenodo. https://doi.org/10.5281/zenodo.18206503