Conference Agenda

The project Untangling the cordel seeks to study and revalue a corpus of Spanish chapbooks dating from the 19th century by creating a digital library (Leblanc and Carta 2021). This corpus of chapbooks, also called pliegos de cordel, is highly heterogeneous in its content and editorial formats, giving rise to multiple reflections on its encoding.

In this short paper, we would like to share our feedback and thoughts on the XML-TEI encoding of a Gospel pliego for its integration into TEI-Publisher.

This pliego is an in-4° containing 16 small columns with extracts from the Four Gospels (John's prologue, Annunciation, Nativity, Mark's finale and the passion according to John; i.e. the same extracts as those in the book of hours (Join-Lambert 2016)) duplicated on both sides. This printout had to be cut in half and then folded to obtain two identical sets of excerpts from the Four Gospels. Whoever acquires it appropriates the object for private devotions or protection: it is therefore not an object kept for reading (the text is written in Latin with small letters) but for apotropaic or curative use (Botrel 2021).

To put forward the interest of this pliego as a devotional object and not strictly as a textual object required much reflection concerning its encoding and its publication on our digital library. Indeed, depending on our choice of encoding, the information conveyed differs: should we favour a diplomatic and formal edition or an encoding that follows the reading?

To determine which encoding would be the most suitable, we decided to test two encoding solutions, one with <facsimile> and another with <sourceDoc>. The visualisation of the two encodings possibilities on TEI-Publisher will allow us to develop the advantages and disadvantages of each method.

The Bocabulario de lengua sangleya por las letraz de el A.B.C. is a historical Chinese-Spanish dictionary held by the British Library (Add ms. 25.317), probably written in 1617. It consists of 223 double-sided folios with about 1400 alphabetically arranged Hokkien Chinese lemmas in the Roman alphabet.

The contribution will introduce our considerations on how to extract and annotate linguistic data from the historical manuscript and the design of a digital scholarly edition (DSE) in order to answer research questions in the fields of linguistics, missionary linguistics and migration (Klöter/Döhla 2022).

Topic modeling (Blei et al. 2003, Blei 2012) as a quantitative text analysis method is not part of the classic editing workflow as it stands for a way of working with text that in many respects contrasts with critical editing. However, for the purpose of a thematic classification of documents, topic modeling can be a useful enhancement to an editorial project. It has the potential to replace the cumbersome manual work that is needed to represent and structure large edition corpora thematically, as has been done for instance in the projects Alfred Escher Briefedition (Jung 2022), Jean Paul – Sämtliche Briefe digital (Miller et al. 2018) or the edition humboldt digital (Ette 2016).

We apply topic modeling to two edition corpora of correspondence of the German-language authors Jean Paul (1763-1825) and Uwe Johnson (1934-1984), compiled at the Berlin-Brandenburg Academy of Sciences and Humanities (BBAW) and the University of Rostock (Miller et al. 2018, Helbig et al. 2017). In our contribution, we discuss how the results of the topic modeling can be usefully integrated into digital editions. We propose to integrate them into the TEI corpora on three levels: (1) the topic model of a corpus, including the topic words and the parameters of its creation, is modeled as a taxonomy in a separate TEI file, (2) the relevance of the topics for individual documents is expressed in the text classification section of the TEI header of each document in the corpus, and (3) the assignment of individual words in a document to topics is expressed by links from word tokens to the corresponding topic in the taxonomy. Following a TEI encoding workflow as outlined above allows for developing digital editions that include topic modeling as an integral part of their user interface.

The Catalogue of Heroines (Odyssey 11.225-330) presents a corpus of prominent mythological women as Odysseus recounts the stories of each woman he encounters in the Underworld. I undertook a TEI close reading of the Catalogue in order to center ancient women in a discussion of the Odyssey and determine how the relationships between the heroines contribute to the Catalogue’s overall purpose. In this short paper I demonstrate first my process: developing my own detailed feminist translation of the Catalogue, applying a TEI close reading to both my translation and the original ancient Greek, and creating a customized schema to best suit my purposes. Then, I detail my analysis of my close reading using cross-language encoding and a prosopography I developed through that reading, which reveals complex connections, both explicit and implied, among characters of the Catalogue. Third, I present the result of this analysis: that through this act of close reading I identified a heretofore unconsidered list of objects within the Catalogue and then demonstrated how these four objects of the Catalogue, ζώνη (girdle), βρόχοs (noose), ἕδνα (bride-price), and χρυσὸs (gold), reveal the ancient Greek stigma surrounding women, sexuality, and fidelity. These objects clearly allude to negative perceptions of women in ancient Greek society and through these objects the Catalogue of Heroines reminds its audience of Odysseus’ concerns regarding the faithfulness of his wife Penelope. Ultimately, by applying and adapting a TEI close reading, I identified patterns within the text that spoke to a greater purpose for the Catalogue and the Odyssey overall, that was able to export for further analysis of this prosopographical data. By the time of the conference, I will be able to present data visualizations that provide pathways that can assist other classicists to center women in ancient texts.