Digital Editions and Text Processing

The paper analyses the causes of the persisting resistance of the scholarly community as a whole to support unreservedly the adoption of digital editions despite their superior functionalities. A possible reason for this drawback is identified in the difficulty of transferring to the machine a suitable part of the reader’s ‘competence’ and in the lack of an adequate data model to process textual information. The standard form of text representation does not does not provide a satisfactory data model to improve our critical engagement with the text through effective computational processing. An adequate understanding of the role played by markup in representing both syntactic and semantic relationships and abstractions can afford a dynamic model of the textual condition. The status of markup is ambivalent: like any other diacritical mark, it can be seen as part of the text, or as a metalinguistic indication of a certain textual feature; moreover, markup can designate both an operation and its value: as the result of a restructuring operation, it provides a textual variant; as an instruction to a restructuring operation, it provides an interpretational variant. A dynamic textual model based on these features is then proposed. The introduction of 'extended strings' as a new datatype for textual information, as proposed by Manfred Thaller, could assist in transferring all the injunctive and performative force of markup to basic low level processing and so enhance the functionality of text representation and digital editions.