A digital critical edition model for Priscian. Glosses, graeca, quotations

This paper discusses three key methodological aspects of a model for a planned digital critical edition of Priscian's Ars grammatica: 1. the glossae found in manuscripts and their relationship with the manuscript text; 2. the presence of Greek text and its importance for the recensio and the construction of a stemma codicum; 3. literary quotations from classical works. Such research is connected with the project of a new edition of the Ars, initiated by Michela Rosellini with her 2015 edition of the second part of the XVIII book of the work. 1. Glossae. Their inclusion in the edition poses specific modelling challenges, since they bear a complex relationship with Priscian's text: a glossa lives in a manuscript's page, so it refers to the text of that specific witness, not to the abstract text proposed by the editor, and as such it must be modelled in the edition. This can be achieved in the TEI XML encoding by linking each glossa to the manuscript it belongs to, and to the specific manuscript reading it comments upon. 2. Greek text. As Rosellini pointed out, while contamination is very common in Priscian's manuscripts, it is much less frequent for the Greek portions of the text, so the latter become key for the recensio. In the proposed edition model, Greek passages will be encoded on two layers, i.e. with both a normalized and a palaeographic transcription. As a consequence, not only "substantial" readings, but also "palaeographic" and "orthographic" variants will be recorded for those textual portions, thus providing scholars of Priscian's text with additional philological evidence. 3. Literary quotations. They will be marked with machine-readable and processable citations of their sources through the Canonical Text Services (CTS) protocol and the newly-developed Distributed Texts Services (DTS) specification.