PEMM was launched in March 2018 to catalog and translate the Ethiopian compilation text that contains these stories, called Täˀammərä Maryam (Miracles of Mary). The project is aimed at creating a resource for all students and scholars interested in folklore and to the Ethiopian Orthodox Church community. It provides Ethiopians with digital access to their patrimony, while raising general awareness about the beauty, breadth, and variety of these vital works of early African literature.

Drawing on manuscripts from the exceptional collection of Täˀammərä Maryam in Princeton’s Rare Books and Special Collections, PEMM collects information about these miracle tales to enable better scholarship on their contents across regions, languages, and time. Its base is the painstaking work of miracle story identifications that William F. Macomber did in the 1980s, itself compiling the work of previous scholars.

Conceived and directed by Prof. Wendy Laura Belcher, PEMM began in 2018 and was developed in partnership with Princeton’s Center for Digital Humanities, along with PEMM’s first project manager Evgeniia Lambrinaki, second project manager Blaine Kebede, and technical lead Henok Alem; as well as the catalogers and translators Jeremy Brown, Mehari Worku, and Dawit Muluneh. Many others contributed as part of the team of scholars, programmers, volunteers; community and scholarly partners; and funders. 

In five years, PEMM has created a database with over 33,14,563 pieces of painstakingly collected metadata about representations of the world's most storied person and the power of one woman to heal the world and deliver justice in both Arabic and Geʿez (classical Ethiopic). It has documented the development of the Marian story genre from the very beginning to its flourishing on the African continent, all the way into the present. In terms of content, PEMM has 33,399 pieces of metadata on 1,000 amazing and surprising stories; 38,110 pieces of metadata on 1,000 manuscripts; 108,144 pieces of metadata on 2,500 beautiful paintings; 2,086 pieces of metadata on 500 archives, and 68,125 pieces of metadata on 3,000 translations, as well as 1,724,512 pieces of metadata in story instance and 39,092 pieces of metadata on keywords.

PEMM is a shared digital library providing metadata in a centrally available system, allowing users to search across collections from multiple institutions. This benefits the 100+ institutions in the PEMM database by extending the reach of their digital collections. 

PEMM began in Google Sheets but now uses a PostgreSQL database managed with the DirectUs Content Management System. The data is hosted on AWS Aurora. This data is uploaded annually to Zenodo, with the last update in August 2023 posted at Zenodo 2023. PEMM code is archived in Github, at Princeton PEMM. Singleton web pages are created using DirectUs's default APIs; the dynamic pages have custom APIs stored in the Github DirectUs-script repository. The site was created using Next.js, Javascript, Mapbox, and OpenSearch. If available on the host site, PEMM uses IIIf ("a set of open standards for delivering high-quality, attributed digital objects online at scale") to display paintings. Where it is not available on the host site, PEMM uses images it has collected and stored on its AWS server.

Origin

In 2013, Belcher attended a conference on hagiography at Ruhr-Universität Bochum, Germany. She presented a paper titled “What Would Mary Do? Thirteenth-Century French and Fifteenth-Century Ethiopic Marian Miracles in Comparative Perspective.”

In preparing the paper, she realized that she couldn't conclusively say even basic things about the Ethiopian Miracles of Mary book or its stories. Much scholarly speculation had attended them--particularly their origin, dating, quantity, topics, and so on--but no quantitative research. 

Belcher decided to start a project that would collect such quantitative data. Prof. Steve Delamarter told her that William F. Macomber had done much work in the 1980s on an unpublished  hand list of stories and manuscripts. This would form the basis of the PEMM project, although expanded nearly ten fold. 

Research Questions

PEMM was designed to answer some important research questions. 

• Dating: When is the earliest attestation of each tale (in which century does each first appear)? Are some stories more popular in some centuries than others? In which century is each tale the most popular? Do some stories fade in popularity over time? What are the themes of the Ethiopian Marian miracle stories?
• Provenance: Where is the earliest attestation of each tale (i.e., in which Ethiopian region does each first appear)? Are some stories more popular in some regions than others? In which region is each tale the most popular? The least popular?
• Themes: What are the themes of the Ethiopian Marian miracle stories? Which themes are most common (i.e., appear in the most stories)? Which themes most commonly appear together in a tale (e.g., women and healing)? When and where is the earliest attestation of each theme? Are some themes more popular in one region than another? Do manuscripts have overarching themes (i.e., do some manuscripts have stories mostly on healing, or kingship?)
• Characters: Are men or women protagonists more common? Young or old? Which character professions are most common (e.g., monks or farmers)?
• Order: Is the order of Miracles of Mary stories within manuscripts random? Is there a standard order of stories (e.g., the same 33 first and in the same order; foreign first, indigenous second; early ones first, later ones second)? Does that order change by century or region? Which stories do most manuscripts begin with? Which stories do most manuscripts end with?
• Relationships: Which stories appear most often with which other stories in the same manuscript? Where in manuscripts is each story most likely to appear (e.g., early, middle, late)? Is there a standard collection of stories (e.g., the same 33 told in most manuscripts)? Does the standard change by century or region?
• Recensions: How did individual Ethiopian Marian miracle stories change over time and region? Did they get longer or shorter?
• Origin: What is the origin of each Ethiopian Marian miracle tale? Which are indigenous and which are foreign (i.e., told first in Europe or the Middle East)? Do foreign Marian miracle stories fade in popularity over time (i.e., told more often in the 1400s and less often by the 1900s)? Are they more popular in some Ethiopian regions than others? Do foreign and indigenous stories vary by theme (e.g., more foreign stories are about knights)? Do foreign and indigenous stories vary by placement (e.g., foreign stories tend to appear at the beginning of manuscripts)?

Phases

PEMM’s first and second phase were made possible by the Princeton Center for Digital Humanities team and funding.

During the 2018-19 academic year, the CDH provided both a Dataset Curation grant and a Public Humanities grant. This funded the time of the CDH Development and Design Team. PEMM began by converting into computable data the extraordinary work done by William F. Macomber in the 1980s, who catalogued 643 stories in 175 manuscripts in Gəʿəz and typed up that information in an unpublished handlist, with many hand emendations in pen. With CDH help, Belcher and Lambrinaki scanned it, cleaned up with the program Sublime Text, and put in spreadsheet form. This funding also resulted in an outreach program addressed at engaging the Ethiopian Orthodox community. 

The second phase was supported by a CDH Research Partnership grant. This funded the time of the CDH Development and Design Team to work with Belcher to create a robust data structure to store and connect the data, and collaborate on building prototype web interfaces and data visualizations. This was completed in January 2020.

The third phase was funded by the Princeton Council of the Humanities and executive director Kathleen S. Crown, through the David A. Gardner Innovation Grants for New Projects in the Humanities, as Belcher and her team created content and performed analyses of it. The phase was to be completed in June 2021. This phase was also supported by the University Committee on Research in the Humanities and Social Sciences grant for student workers to keyword and translate stories.

PEMM’s fourth phase was made possible by two major grants from the National Endowment for the Humanities, awarded for work from fall 2021 through summer 2024. The NEH Scholarly Editions and Scholarly Translations Grant funds the team of experienced researchers with rare language skills to catalog stories in parchment manuscripts, translate stories into English, and write short introductions to them. The NEH Digital Humanities Advancement Grant funds a public-facing open-access web application and data portal to share the stories in, images about, translations of, and scholarship on this crucial body of medieval African literature and to build upon our innovative prototype tool for searching in Gəˁəz.

Other important funders throughout were the Princeton Department of African American Studies, previously directed by the Eddie S. Glaude and now by Tera Hunter, as well as the Program in Gender and Sexuality Studies (directed by Wallace Best), the Program in African Studies (directed by Emmanuel Kreike and now Chika Okeke-Agulu), the Center for the Study of Religion (directed by Jonathan Gold), and the Department of Comparative Literature (directed by Thomas Hare).

PEMM's fifth phase was made possible by a Princeton Alliance for Collaborative Research and Innovation (PACRI) grant to work with the Howard University Center for African Studies on outreach to middle school and high school teachers and students, as well as the public.