Reclaiming Their Time: Developing The Heresies Project as an Intersectional Digital Feminist Resource

The Heresies Project aims to produce an open-access digital resource dedicated to the Heresies journal -- a pioneering work of feminist art and criticism published from 1977 to 1993. The journal featured the work of some of the most important Second-Wave Feminist thinkers, authors, and artists of the 20th century, including Adrienne Rich, Audre Lorde, Joanna Russ, Harmony Hammond, Lucy Lippard, Joan Braderman, and others. The journal, and the radical feminist collective that produced it, had a profound impact on the development of intersectional feminist analysis of society’s most confounding problems, including gender expression and identity, sexual and racial inequality, environmental crisis, and the marginalization of art and humanities in societal discourse. And yet the journal, which was published in small print-runs and distributed independently by members of the Heresies Collective, are hard to find in physical form. Recently, 26 of the 27 issues were digitized as part of JSTOR's Reveal Digital program and made available as open access image-only PDFs, a first step in making Heresies more accessible to today's feminist scholars. The irony is that while Heresies has had a profound impact on 21st-century intersectional feminist thought, there are many feminist scholars and practitioners who have never seen a copy.

The Heresies Project team at Bucknell University began its work in 2021, with the intention of producing an open-source accessible digital edition of the issues. The team also began to produce a personography of the women who published in the journal and an annotated bibliography of the works and references to other sources cited by the authors. By undertaking these three aspects of the project, we are making a significant contribution to the socio-cultural understanding of feminist knowledge production. In the process of taking on this labor, the team encountered three significant obstacles, which will be addressed in this presentation:

1. The image-only PDFs with which the team had to work were difficult to transform into machine-readable texts, due to the complex, at times idiosyncratic design of the journal (Figure 1). We used the Transkribus1 text recognition engine to capture the text components, but the complex nature of the pages required a significant amount of post-processing before we could undertake any kind of encoding or text analysis. Furthermore, we needed to find a solution which would enable us to ‘see’ the original page copy while we were checking the text for accuracy (Figure 2).
2. While textual scholars working on the project were keen to focus on TEI-XML encoding for structural and semantic encoding and analysis, the art historians on the team were frustrated by privileging text over image, and were keen to find ways to annotate the images, while at the same time presenting the multimodal work in the issues as they were designed and intended. In the process of developing a personography that could be shared via Linked Open Data to the Semantic Web, the team discovered how few of the women who created and contributed to the Heresies journal have any kind of data “footprint” in existing authorities like Wikipedia, VIAF, or Getty; those who do have a footprint often have but a base stub of an entry, without any kind of biographical detail or properly researched information.

Building the Heresies project in the LEAF (Linked Editing Academic Framework) virtual research environment provides the team with the opportunity to edit and annotate the texts in TEI-XML, publish high-resolution images of the full issues, and tag named entities (people, organizations, and entitled works of art, literature, and criticism). Using LEAF’s web-based tools has allowed us to streamline the transformation from Transkribus output to valid TEI, capture named entities using LEAF’s entity look-up tool (Figure 3), and publish web versions of the encoded texts. Our approach has been to attend to one issue at a time, rather than to batch-transform all 27 issues. This approach allows us to focus on the theme of each issue, refining and expanding the information we are capturing. It also provides the opportunity to undertake research about the women who contributed to each issue and, where possible, provide information about them to the Semantic Web, bringing our Heresies project into dialogue with other feminist DH projects, like the Yellow Nineties project and Orlando. However, this focus also creates complexities with regard to the human labor required to produce the project. The team is made up of faculty, staff, and students. To date, three ‘generations’ of undergraduate students have contributed to the project, working for a year or two as paid researchers before graduating from the university and passing on their experience and expertise to the next ‘generation’. This approach to crowdsourced knowledge production has significant challenges. In fact, three of the co-authors of this presentation will have graduated from Bucknell by the time the conference happens in August 2024.

Our presentation will focus on two aspects of the project:

1. The continuing refinement of the workflow to optimize rich, multimodal analysis and presentation of the issues as an open-access digital scholarly resource.
2. The development of a data model that allows us to “mint” and augment the entities revealed within the pages of the journal, and how we will use LEAF’s integrated bridge to the Semantic Web, that will not only reintroduce many of these women to feminist discourse but will also reveal the social networks of contributors and collaborators to the larger understanding of 20th century cultural production.