READ Latin
Research Environment for Ancient Documents

READ platform

READ (Research Environment for Ancient Documents) is an open-source platform, developed by Stefan BaumsAndrew Glass, Ian McCrabb and Stephen White for the study of ancient written documents.

It has been recently modified for the analysis of Latin epigraphic texts. Until now, Latin texts had not been included in the platform. In order to test the validity of READ for the purpose of analyzing documents relating to Latin epigraphy, a Latin version of the platform was hosted on the Ca' Foscari University server, and called READ Latin.

16 sample texts have been identified, representative of the typological variety of the epigraphic documentation in Latin, both genuine and forged, known directly or indirectly (handwritten tradition).

The result has been a consequent variety of critical and lexical mark-up, which made it possible to draw the first assessments regarding the ability of READ Latin to operate with texts belonging to a language so far not contemplated by the platform, to dialogue with the coding criteria in use at epigraphic datasets mainly used by specialists (EDR - Epigraphic Database Roma, EDB - Epigraphic Database Bari, EDF - Epigraphic Database Falsae), as well as to export data to EpiDoc.

The test allowed finding above all a positive element, not achievable in the current digital epigraphic panorama. READ Latin is in fact able to perform a punctual paleographic analysis, character by character (grapheme by grapheme).

Honorary inscription for Arrius Antoninus from Concordia.

One of the most interesting directions for READ Latin would be the implementation of the dataset by injecting and curating data retrieved from existing databases (e.g. EDCS Epigraphik Datenbank Clauss-Slaby). With READ as a VRE the curation of the dataset could effectively be achieved through a web-based community of junior epigraphists participating in a joint project, following a shared protocol. This would allow, in parallel, the implementation of a key new feature of READ: AI assisted palaeographic analysis. If launched on a significant textual corpus of Latin inscriptions, it would provide a potentially very fruitful outcome: the identification of recurring letter-patterns, that in the domain of serial production, as that of Greek and Latin epigraphy, would invaluably help us to track the genealogy and diffusion of epigraphic habits and of epigraphic trades throughout the Roman Empire.