• Marco Leone de Castris University of Naples

University of Naples

This contribution offers a reflection on the achievements of the ‘MAGIC’ project in the field of Digital Humanities. It outlines the results obtained with regard to the methodology for the digitization of manuscripts and early printed books (incunabula and sixteenth-century editions), and examines how digital image enhancement has impacted philological research.

In particular it looks at the use of computer-vision algorithms designed to reduce bleed-through (the migration of ink from recto to  verso) and asks how these tools can improve the readability of marginal notes that are otherwise difficult to decipher, opening new perspectives for the study of textual transmission. The case study will be primarily on «postillati»  ‒ copies containing annotations, marginalia, and ownership notes ‒ belonging to a collection examined within the project, namely the substantial Torraca Collection preserved at the Accademia Pontaniana (Via Mezzocannone 8, Naples). 

The ‘MAGIC’ project was carried out through a collaboration between the Department of Humanities and the Department of Physics “Ettore Pancini” of the Università degli Studi di Napoli «Federico II», and resulted in the creation of an operational prototype of a ‘Smart Library’: an infrastructure capable of combining the material preservation of digitized items, scientific analysis, specialized metadata creation, and open digital access, thus reconfiguring the library from a place of custody into a dynamic instrument for the production, integration, and dissemination of knowledge.

The digital catalogue has been available since December 2023 at https://www.magic.unina.it/ and is continuously updated.

Digitization activities focused on three main corpora: the illuminated manuscripts of Dante’s Commedia; the incunabula and sixteenth-century books of the Accademia Pontaniana and of the Society of Sciences, Letters and Arts of Naples; and the Copia lettere and Copia dispacci of the Royal Palace of Caserta. 

As mentioned above, our case study concerns the Torraca Collection of the Accademia Pontaniana, consisting of 192 editions from the fifteenth and sixteenth centuries (6 incunabula and 186 sixteenth-century editions). The corpus is highly heterogeneous: it includes mainly works of Latin literature (59) and Italian literature (43), followed by history (33), philosophy (12), religion (10), and law (6).

The volumes were digitized using high-resolution planetary scanners, followed by image processing (adjustment of saturation, brightness, color balance, and verification of image integrity) in accordance with IFLA guidelines for the handling of rare books. The master files, produced in TIFF format with JPG derivatives, were converted into F.I.T.S. Thanks to macro-data processing, and through pixel classification and spatial coherence–based filtering, it was possible to increase text legibility by reducing the aforementioned bleed-through effect and thus to recover glosses and marginal commentaries, with significant implications for philological analysis and for the description of layers of use.

A specific case is presented: a copy of the edition Opera de lo elegante poeta Thibaldeo ferrarese, Venice, Bindoni, Francesco Bindoni & Maffeo Pasini, 1535, which contains a small series of annotations by a Petrarchan scholar.

 

Bibliography

G. Petrella, Scrivere sui libri. Breve guida al libro a stampa postillato, Roma, Salerno editrice, 2022.

G. Petrella, Patrimonio “diffuso”. Gli incunaboli delle biblioteche di Napoli. Storia e catalogo, Roma-Padova. Editrice Antenore, 2025.

S. Sbordone, Le edizioni del ‘400 e del ‘500 nell’Accademia Pontaniana di Napoli, in «Quaderni dell’Accademia Pontaniana», 51, 2007.

• Frances Nolan University College Dublin

University College Dublin

Based on research from the Research Ireland-funded Law versus Practice project,  which uses digital tools and methods to identify and map women’s property ownership in early modern Ireland, this paper will discuss the creation of a database from an early modern printed source titled A list of the claims. Published in Dublin in 1701, the List is a printed schedule of claims submitted to the board of trustees for the Irish forfeitures between 1700 and 1703. The board was created by the Act of Resumption (1700), which was passed by the English Parliament to nullify William III’s ‘exorbitant’ grants of Irish land to military generals and personal favourites. The royal grants comprised estates that were forfeited by Irish supporters of James II, after their defeat by William and his forces in the War of the Two Kings in Ireland (1689-91). Crucially, the Act of Resumption was retroactive; this meant that any person with a title to all or part of a forfeited estate that predated 13 February 1689 (the date of William III and Mary II’s accession) could submit a claim. The volume of submissions was considerable: the List of claims comprises over 3,100 claims, almost all of which were heard and adjudicated upon by a court of claims, which was presided over by the forfeiture trustees. 

The List is a valuable record of property-ownership in seventeenth-century Ireland. It is also a significant source of biographical, legal and geographical information, containing details of over 4,000 named individuals, the legal basis of their ownership (e.g. deeds, marriage settlements, wills) and the geographic location of the property they claimed (i.e. townlands, parishes, baronies and counties). Using a pre-existing dataset (c based on the printed List and contained in an Excel spreadsheet), the Law versus Practice team are parsing biographical, geographical and legal data to create a relational database through MySQL. This database will ultimately enable the team to trace and, using GIS, map female property ownership across time and place. At present, the team has identified 771 women and girls, who will be assigned unique identifiers (UIDs), using the Virtual Record Treasury of Ireland (VRTI) Knowledge Graph (KG) person schema as a template. 

While the LOC dataset is a valuable foundation for the creation of a database, it is not without issues: first, the dataset is a reproduction of the original printed List and thus contains errors, contractions and inconsistencies in the spelling of names and placenames; second, numerous individuals share the same name and live or own property in close proximity to each other; and third, data is not structured to distinguish gender, or to identify instances where women have remarried and changed their name. These issues require manual intervention, through cleaning and research to establish context. 

Using the LOC dataset and the Law versus Practice database as a model, this paper will explore the potential of (and the potential problems attending) datasets and databases created from early modern sources in the Irish context.  

• Bethany Warner International Institute of Social History

International Institute of Social History

 

“We have decided at the meeting of 1 June to impose a heavier punishment on that people, to strike fear into them, so that those of the Company's slaves who run away and are recaptured will be flogged for the first time and branded on the cheek; the second time, a brand on the other cheek, and the third time, in addition to flogging, to cut off the nose and ears, insofar as they may not have committed a crime during their escape that is punishable by death”.

 

The recent digitalisation of the UNESCO Memory of the World-listed Dutch East India Company (VOC) archives has opened the door to new kinds of research into Dutch colonialism and the lives of those subject to it. These archives contain vast information on the inner workings of the VOC and the societies in which they operated. Major infrastructure projects, such as Globalise, have worked to increase the accessibility and searchability of these archives using machine learning techniques for handwritten text recognition (HTR). On a smaller scale, projects at the International Institute of Social History have transcribed another 15 inventories of the VOC archives to make them accessible for research, equalling around 2.5 million individual pages.

The resulting corpus of around 7.5 million transcribed pages provides an opportunity for computational approaches to research. In this presentation, I propose to explore how the physical body was impacted by the process and experience of enslavement in the context of the VOC. This topic has been researched in the context of the British Caribbean by Stefanie Hunt-Kennedy, however, there is no comparable research in the Asian context. Hunt-Kennedy argues that in the Caribbean context, “the disabilities produced by slavery and endured by the enslaved became a sign of slaveowner power and a physical marker of Africans’ enslaved status” while simultaneously, “such marks also told a narrative of enslaved people’s personhood, survival, and their refusal to accept their enslavement”. I will explore the validity of this statement in the context of Asian slavery and the dynamic between enslavement, the physical body, and resistance. The opening extract from the Cape of Good Hope describes, in the Dutch colonial context, how slaves who attempted to resist were potentially subject to significant physical punishment, a demonstration of the power of the VOC, but also an indication of active desertion and resistance. 

While this decision could be individually identified using the Globalise transcriptions viewer, in this research, I will make use of digital humanities techniques to explore newly searchable transcriptions of the VOC archives at scale. Specifically, I will use lexical searching in combination with natural language processing techniques, including named entity recognition, to create a dataset of potential observations of moments of resistance by enslaved peoples. Furthermore, I will explore data visualisation to analyse trends in the kinds of bodies which were commodified as slave labour, punishments received, and locations of resistance. I will discuss the challenges of working with automatically transcribed material, which are often rife with spelling errors and layout mistakes. This will result in both a research database and a starting point for close reading, to delve into the archives in a targeted way to explore the relationship between resistance, the body, and enslavement in VOC controlled societies.

---

1. Nationaal Archief, Den Haag, Boekhouder-Generaal te Batavia, access number 1.04.18.02, Inventory Number: 10458, pp. 61. Cape of the Good Hope, 1712: Register of Letters, Books, and Other Papers. Translated from Dutch

2. ‘Archieven van de Verenigde Oost-Indische Compagnie’, accessed 2 March 2026, https://unesco.nl/memory-of-the-world/archieven-van-de-verenigde-oost-indische-compagnie.

3. ‘Globalise Project Overview’, GLOBALISE, accessed 1 May 2025, https://globalise.huygens.knaw.nl/project_overview/.

4. Stefanie Hunt-Kennedy, ‘Disability History in Slavery’s Archive’, in Early Modern Medicine, 1st edn, by Olivia Weisser (Routledge, 2024), 261, https://doi.org/10.4324/9781003094876-19.

5. ‘Globalise Transcriptions Viewer’, accessed 2 March 2026, https://transcriptions.globalise.huygens.knaw.nl/.