Digital representation and the use of shared texts: the case of a theatrical prompt book

Abstract

The digitisation of cultural textual artefacts is focussed on conservation in digital archives and on the information retrieval tools for which they are designed. Many such artefacts, however, exist only in the form of discontinuous, unstable and stratified objects whose representation within this model presents problems for humanists and technologists alike. A less canonical vision of the past reveals that variation and instability are the constitutive elements of culture and their transmission is characterised by interaction and contamination. The digital representation thus needs an ethnographic perspective, in a cultural context interwoven with relations, exchanges and alterations. The typescript of the comedy by the author, who was also actor and director, the Roman Ettore Petrolini, which we examine here, exhibits the very characteristics that sum up this challenge. The comedy Peppe er Pollo (Peppe the Fool) was composed in the early 1920s and performed until the end of the 1930s, in standard Italian or in the Roman dialect. The Italianisations and alterations to the substance of the text probably by actors and other agents are very numerous. What we are dealing with is a network of written and spoken traces, which cannot be linked back to a single authorial intention. How to capture this complex stratification by different agents and how to facilitate interactions with users, readers, students and scholars is obviously a great technical challenge. We extend our earlier work on Multi-Version-Documents or MVDs for recording complex textual variation to the meta-textual properties editors typically use to structure or interpret the text. Instead of using embedded markup to record this information we represent it externally, so breaking down the humanistic information into separate categories of text, markup and images of the typescript. The markup, like the text, exists in many versions, and both are stored in separate MVDs so that the correct markup and text can be retrieved for each version. The two are then combined in the browser with links to the images of the relevant pages. This separation allows different sets of markup, e.g. structural vs. interpretative markup, to be freely mixed or interchanged, so increasing collaboration between scholars and facilitating information reuse. Conformance to embedded XML standards such as TEI (Text Encoding Initiative) is thus forced into an outer layer of the software system, as an import/export format, rather than integrated into its design.