Digital Medieval Studies—Practice and Preservation 9781802700152

This project-based publication aims to bridge the gap between digital and conventional scholarly activity and to communi

163 75 4MB

English Pages 130 [128] Year 2022

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Digital Medieval Studies—Practice and Preservation
 9781802700152

Citation preview

Collection Development, Cultural Heritage, and Digital Humanities For further information and publications, please see: www.arc-humanities.org/search-results-list/?series=collection-developmentcultural-heritage-and-digital-humanities

DIGITAL MEDIEVAL STUDIES PRACTICE AND PRESERVATION

edited by

Laura K. Morreale and Sean Gilsdorf

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. © 2022, Arc Humanities Press, Leeds

The authors assert their moral right to be identified as the authors of their part of this work.

Permission to use brief excerpts from this work in scholarly and educational works is hereby granted provided that the source is acknowledged. Any use of material in this work that is an exception or limitation covered by Article 5 of the European Union’s Copyright Directive (2001/29/EC) or would be determined to be “fair use” under Section 107 of the U.S. Copyright Act September 2010 Page 2 or that satisfies the conditions specified in Section 108 of the U.S. Copy­ right Act (17 USC §108, as revised by P.L. 94-553) does not require the Publisher’s permission.

ISBN (HB): 9781641894463 ISBN (PDF): 9781802700152 www.arc-humanities.org Printed and bound in the UK (by CPI Group [UK] Ltd), USA (by Bookmasters), and elsewhere using print-on-demand technology.

CONTENTS

List of Illustrations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

Introduction: The Medievalist, Digital Edition LAURA K. MORREALE and SEAN GILSDORF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Chapter 1: Beginnings: The Labyrinth Medieval Studies Website DEBORAH EVERHART and MARTIN IRVINE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Chapter 2: New Approaches to Old Questions: Digital Technology, Sigillography, and digisig JOHN MCEWAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Chapter 3: Corpus Synodalium: Medieval Canon Law in a Digital Age ROWAN DORIN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Chapter 4: Teaching Constantinople as a (Pixelated) Palimpsest J. W. TORGERSON. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Chapter 5: Life on—and off—the Continuum LISA FAGIN DAVIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Appendix: Permanent Links to the Catalogued Assets of Profiled Projects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

LIST OF ILLUSTRATIONS

Figure 1.1: Image of the floor labyrinth at Chartres Cathedral (ca. 1215–1221 CE), reworked. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Figure 1.2: Martin Irvine and Deborah Everhart at Georgetown University with an early Labyrinth page . . . . . . . . . . . . . . . . 19 Figure 3.1a: Search results for the word matrimonium (marriage), showing absolute frequencies over twenty-five-year intervals. . . . . . . . 62

Figure 3.1b: Search results for the word matrimonium (marriage); clicking any result brings up the full text with the search term highlighted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Figure 3.2: Search results for the word matrimonium (marriage); clicking any result brings up the full text with the search term highlighted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure 3.3: Map showing frequency of references to matrimonium (marriage) in diocesan statutes; clicking any jurisdiction brings up its associated texts with the search term highlighted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Figure 4.1: Constantinople as Palimpsest concept, with map for the topic Monumental Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Figure 4.2: Constantinople as Palimpsest, ca. 2015. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Figure 4.3: Clustered cisterns near the Imperial Court. . . . . . . . . . . . . . . . . . . . . . . . . . 90 Figure 4.4: Clustered cisterns in the regions of Blachernae and Petra. . . . . . . . . . . . 90

Figure 4.5: Cistern clusters around the monastery of Constantine Lips. . . . . . . . . . 90

Introduction

THE MEDIEVALIST, DIGITAL EDITION LAURA K. MORREALE and SEAN GILSDORF

If you are a scholar of the medieval world, a morning of research and writing might

unfold in the following way. After starting up your laptop, you open a document using Microsoft Word or some other word processor. You scroll through the pages, perhaps pausing to fix an errant footnote or typo. Once the writing process begins, you might select a few .pdfs of primary or secondary source readings, open them on your computer’s desktop, then search for a passage or term by using a menu or keystroke command. If the text is something you decide to quote in your article or chapter, you may be able to highlight it, press “control-c” (copy), and then press “control-v” (paste) to incorporate the relevant passages from the source material straight into the document you are writing, thereby avoiding concerns about introducing spelling or other errors into the writing process, no matter how obscure or unfamiliar the text. You’ll then note the page number from the .pdf, enter it into a footnote (the format of which has been checked and rechecked on the website of the journal or publisher to which the researcher is submitting), and move on with the writing process. What if a file is missing from your personal laptop or Zotero library, or you’ve misplaced the details needed to cite the work or page number? This means a virtual trip to the library to consult the online catalogue, to dip into computer-based repositories like Internet Archive or HathiTrust, or even to reference a digitized manuscript so that the argument is made whole and all sources accounted for. Unfortunately, there also may be missing pieces that make such a virtual trip to the library impossible: you cannot remember the author of an article, the editor of a collection, or the shelf mark of the manuscript that would allow an easy search through the digital archives. No need to worry; there are field- and subject-specific databases and online bibliographies that collect these references and make them accessible to fellow scholars, all a click or two away. A quick Google search also may turn up the right information, and if all hope is lost, a call can be sent out to colleagues via email, Twitter, Facebook, listservs, or other forms of instant messaging. People (sometimes ones you’ve never even met before) respond to your pleas, perhaps within minutes, offering up citations, missing page numbers, and even emailed copies of articles to get you back to the reading, thinking, and writing that is so fundamental to our field. Digital tools and approaches inform our pedagogy as well. Twenty-five years ago, you might have hired an information technology (IT) professional to build and manage a website for you, if you felt you needed one at all. Now, many of us maintain a web presLaura K. Morreale, Global Medieval Studies Program, Georgetown University; Sean Gilsdorf, Committee on Medieval Studies, Harvard University.

2

Laura K. Morreale and Sean Gilsdorf

ence via course pages on our school’s content management platforms. We upload materials, provide links, add new pages, and subscribe users to the curated web-spaces that nearly everyone who teaches is required to have. We must become savvy consumers of web-based knowledge, content, and tools in order to populate our course pages, add chat rooms, incorporate YouTube videos, or integrate tools, exhibits, or workspaces for our students.1 This is what being a scholar and teacher looks like today: online, instantly and unceasingly accessible, and in constant contact with colleagues and repositories of knowledge.2 Yet despite our field’s deep immersion in online modes of researching, writing, and teaching, many medievalists still feel uncomfortable calling themselves digital scholars, and thus are quick to dismiss ground-breaking digitally-inflected medieval scholarship as outside their purview and not relevant to the way they work.3 Whatever the reasons, there is a gap between what we think we are doing as medievalists and what we are actually doing as we consult manuscripts, collate texts, compile subjectspecific bibliographies, teach and mentor students, stay up-to-date with the latest work in our sub-field, and create new scholarship in a computer-based knowledge economy.4 This collection of invited essays features authors whose computer-based scholarship is oriented towards fundamental goals of medieval humanities research: they identify and make available field-specific resources, locate and organize sources from the medieval past, contextualize, analyze, and compare medieval materials, and present what they have learned from these efforts to students in the classroom and colleagues in public venues. Because digital humanities scholarship is governed by the same values, practices, and goals of humanistic inquiry more broadly, its products also are recognizable to all medievalists, and there are many journals and essay collections that map out how individual digital projects extend the scholarly activities first practiced in the analogue world. Our goal in this volume, however, is to ensure both that this scholarship is taken seriously as an identifiable and articulated process, and just as importantly that it can exist as a durable intellectual product. The chapters that follow thus are neither project reports nor stand-ins for the digital efforts of their authors.5 Rather, they are critical reflections on what those authors have learned in the process of engaging in computer-based research and teaching, and demonstrations of how such work can be recorded, disseminated, and preserved in a form commensurate with more traditional forms of Medieval Studies scholarship. In this introduction, we outline the journey many medievalists make when setting out to work in the digital space, some common challenges they face, and new ways that they can make their work not only more comprehensible to peers now but also available to scholars in the future. 1  Davis, Gold, and Harris, “Curating Digital Pedagogy in the Humanities.”

2  This point is made effectively in Birnbaum, Bonde, and Kestemont, “The Digital Middle Ages,” S3.

3  There are still few clear-cut examples of the efficacy of digital methodologies; this is addressed directly in Hiltmann et al., “Digital Methods in Practice.” 4  For a reflection on how digital methodologies rely upon core aspects of traditional humanities work, see Moravec, “Digital Historical Practices.”

5  For examples of project reports, see Davis, Mahoney-Steel, and Turnator, Meeting the Medieval in a Digital World, or the many project-based articles featured in the journal Digital Medievalist.



Familiarity and Novelty

The Medievalist, Digital Edition

3

We might describe the current situation, with so many medievalists fully immersed in online knowledge exchange yet still hesitant to engage with born-digital research in their field, as a case of growing pains. This process of transformation, affecting not only the tools we use to carry out our research but also the very process of that research, can sometimes feel rapid and disorienting—due at least in part to the perception of digital scholarship as a province of the hard sciences, where innovation is the watchword. While much of the rhetoric surrounding born-digital work emphasizes what is new and pathbreaking, therefore, many of us accept change in fits and starts, with an accompanying sense of discomfort, and we rely heavily upon previous practices before engaging in, or even accepting, new modes of working and learning. This is because, as with any technology, new digital platforms, methods, and approaches always must be rooted in and integrated with older, more familiar and comfortable ways of learning and knowing in order to be comprehensible, much less successful. Millennia ago, one of the most momentous transformations in learning technology—the shift from scrolls to codices as the dominant format for longer-form written communication and textual preservation—epitomized this integration of the familiar and unfamiliar, adapting longstanding scribal habits and practices within a new system of informational storage and presentation first embraced (like all technological innovations) by a cadre of early adopters before gaining wider acceptance in the following centuries.6 One can see a similar process at work in the creation of glossed Bibles in the twelfth century ce, as a centuries-old textual form (the Biblical codex) was updated and transformed with what were then high-tech innovations, such as apparatus keyed to main text via complex sigla (the distant ancestors, in a sense, of our modern hyperlinks). Perhaps most famously, the shift in Europe from manuscript to print, despite its frequent characterization as revolutionary, was one in which older forms and formats played a critical role.7 In addition to the ongoing role of manuscripts themselves as sources, models, and adjuncts to printed volumes, for many years printed books had to operate according to manuscript visual and operative logic in order to be accepted and understood, an impulse still seen today in the e-reader, whose creators and programmers quickly realized that users still wanted to be able to turn the page, if only in a simulacral form. Much more recently, the introduction of computers and computing into the scholarly research process has followed a similar path, one in which digital elements have been mixed in or appended onto predominantly analogue materials and outputs. In many instances, scholars have applied computer-based tools such as algorithms, visualization software, or data mining to well-established questions or topics. This approach was one taken by a number of medieval social historians, including David Herlihy and Christiane 6  The classic account remains Roberts and Skeat, The Birth of the Codex, from 1983; cf. the perceptive comments of McCormick, “The Birth of the Codex and the Apostolic Life-Style.”

7  For a recent and valuable overview of this issue in its English context, see Boffey, “From Man­ uscript to Print.”

4

Laura K. Morreale and Sean Gilsdorf

Klapisch-Zuber in their 1978 study of the Florentine catasto of 1427, or Stephen White a decade later in his analysis of kin approval (laudatio parentum) in eleventh-century French donation charters.8 In both cases, computers were used to carry out work that previous generations of scholars either had done painstakingly by hand, or had avoided doing due to the time and effort it required. In turn, medievalists increasingly turned to digital technology in order to expand or enhance works of traditional (i.e., print-based) scholarship. Good examples of this are the Online Catasto project at Brown University, produced by Robert Burr Litchfield and Anthony Moelho to present in generally accessible and searchable form the data derived from Herlihy and Klapisch-Zuber’s earlier research,9 or the cd-rom disks containing maps, images, and sound files that became common appendices to print textbooks in the early 2000s, and that since have been supplanted by analogous publisher-produced websites. Even as digital scholarship matures, however, and the research that we do becomes increasingly born-digital, it remains the case that all such work still must build and rely upon earlier technologies and epistemic structures, and be legible in terms of previous scholarly practices and modes. To be sure, every medievalist today is a digital scholar in important ways. Just as significantly, however, even the most tech-savvy among us continues to be part of a scholarly enterprise with deep and enduring analogue roots. We as a field have come to our current state of digital integration slowly, adopting and rejecting certain technologies along the way insofar as they do or do not meet our research needs.10 Many of us have witnessed this change without taking much notice of it; at times we consciously changed our ways of working, while at other times we did so unthinkingly, but always in step with developing norms, in concert with colleagues who themselves were learning to work in the digital medium and with reference to the working models of the past.

Thinking Digitally, Documenting Narratively

In light of all this, it should come as no surprise that digital argumentation in Medieval Studies still relies upon only a limited number of methodologies, even as computerbased tools are constantly evolving. Among these are the ability to aggregate, analyze, and filter large amounts of data; to access, disseminate, and publish materials broadly; to support simultaneity and user-driven interaction; to link materials across the digital landscape; to collocate words and images within a corpus; and to represent space, whether real or imagined, visually.11 Most computer-based scholarship calls upon one or a combination of these affordances, so that the outcomes are readily recognizable 8  Herlihy and Klapisch-Zuber, Les Toscans et leurs familles; White, Custom, Kinship, and Gifts to Saints. 9  Moelho and Litchfield, The Online Catasto Project.

10  The continued hesitancy about accepting digital modes of scholarship has been under discussion for quite some time. See, for example, Lässig, “Digital History. Challenges and Opportunities for the Profession,” and Edmond, “Collaboration and Infrastructure.” 11  TaDiRAH (Taxonomy of Digital Research Activities in the Humanities), downloadable at http://



The Medievalist, Digital Edition

5

to their audiences. Searchable databases and portals, online image galleries and exhibits, mapping and timeline projects, and specialized resource collections are all familiar products and are increasingly viewed as the valuable and authoritative contributions to Medieval Studies they are. Even though emerging technologies continue to refer to organizational or epistemological constructs from previous scholarly models, there is no denying that the digital medium remakes our sources and has remade us as scholars.12 As the essays in this volume demonstrate, the tools developed by what are now several decades of digital humanists allow us to ask questions we previously could not, and while a digital approach does not replace or render invalid traditional forms of inquiry, it shifts the discussion from the kinds of questions we routinely pose within any particular scholarly tradition to those that new technologies support and promote. As the creators and audiences of digital scholarship become more sophisticated consumers of digital tools and methods and push their boundaries, the objects of historical inquiry, the questions we ask of them, and the conclusions that might yield all are transformed. Technology and content are symbiotic in creating the final product. Moreover, what has been missing up to now in the ever-changing environment of computer-based scholarship is a sense of permanence, which in turn has limited how these important forms of knowledge might operate within our current systems of scholarly exchange.13 Technological challenges integral to the online world—including frequent platform upgrades, obsolete or incompatible programs, or outdated hyperlinks that render materials inaccessible—all make scholars reluctant to cite, employ, or depend exclusively upon born-digital materials. For digital scholarship to be fully integrated into the scholarly ecosystem, there must be a tangible deliverable product and a way to signpost the work done. The failure to take such steps often has led to the dismissal of born-digital work as too inherently unreliable to sustain long-term scholarly attention, despite its clear value and quality. Although work continues on computerbased technological responses to the lack of long-term sustainability, we propose that narrative, one of the oldest and most human forms of information technology, provides a fail-safe approach to counter the loss of knowledge that can occur in digital Medieval Studies.14 Because people often need stories to understand concepts that are new to them, describing digital work and outcomes in a narrative format is a particularly effective way to anchor such scholarship. tadirah.dariah.eu/vocab/index.php, accessed November 10, 2021 offers a helpful classification system for digital methods.

12  A point made forcefully in Albritton, Henley, and Treharne, Medieval Manuscripts in the Digital Age.

13  See Nancy Partner’s “Report of the Delegate to the American Council of Learned Societies,” in the “Proceedings of the Annual Meeting of the Medieval Academy of America, 2016,” Speculum 91, no. 3 (July 2016): 881–85, where she sounded the alarm to medievalists concerning the loss of much born-digital knowledge. Partner repeated this warning at the Academy’s 2018 meeting in Atlanta, Georgia. 14  Owens, The Theory and Craft of Digital Preservation.

6

Laura K. Morreale and Sean Gilsdorf

Therefore, aside from Lisa Fagin Davis’ retrospective essay, all of the projects profiled in this volume have been archived according to the Digital Documentation Process (ddp), a set of cataloguing norms that preserve a project’s digital building blocks, provide a comprehensive history of how the work came to be and how it progressed, and outline its constituent parts.15 The historical component, called the Archiving Dossier Narrative (adn), gives inherently ephemeral online scholarship a story of its own, providing a context in which to understand how each of the digital objects created for the project supported the work during its active phase.16 Project archivists chronicle their efforts from the beginning, describing how these arose from a nagging research question, a gap in the scholarly record, a need to challenge students in a classroom environment, or any other impetus.17 Furthermore, the adn plots out the impact of the work not only in terms of computer-generated analytics, but also in the form of conference presentations, articles, blog posts, or other scholarly outputs emerging from the borndigital efforts of the project initiators. The digital objects catalogued alongside the adn might include spreadsheets, images, text files, mapping coordinates, audio files, or any other component brought to the online presentation of the work. To preserve the collected data and interpretive achievements of online projects, therefore, the ddp marries the storytelling capacity of the adn with a rigorous process for storing and preserving the project’s varied digital components. The following contributions flesh out how a digital approach shaped their authors’ projects and the conclusions they drew from them, and demonstrate what can be gained both retrospectively and in an ongoing way from actively reflecting (through documenting and cataloguing) on what we learn in the process of doing digital scholarship. Personal experience is fundamental to the histories of each of these initiatives, in part because so much of what digital scholarship can teach us comes not with the conclusion of the work, but rather while we are doing it. Taken as a whole, these articles contribute meaningfully and persistently to the short but already rich history of online Medieval Studies, both by recounting what was done and by suggesting the potential lifespan of medieval digital scholarship. We begin, therefore, at the dawn of medieval digital scholarship, and end with a project still in process, one that continues to grow as new students make their own contributions. Deborah Everhart and Martin Irvine’s account of the Labyrinth opens the collection by plotting out the earliest days of digital Medieval Studies, and the groundbreaking decisions that led to much of what we take for granted in humanities computing today. The history of the Labyrinth project, the first and most enduring subject15  Fostano and Morreale, “The Digital Documentation Process.” See also Morreale, “Medieval Digital Humanities and The Rite of Spring.”

16  Links to the Archiving Dossier Narratives of projects featured in this volume can be found in the Appendix. 17  The Archiving Dossier Narrative for the resource collection Middle Ages for Educators, for example, explains that the project was a direct response to the COVID-19 pandemic, which suddenly forced educators online en masse in the spring semester of 2020: Eisenberg, McDougall, and Morreale, “Middle Ages for Educators 1st Edition (June 2020) Archiving Dossier Narrative.”



The Medievalist, Digital Edition

7

specific resource portal for Medieval Studies as well as the first humanities website on the web, reveals how medievalists first came to computer-based learning—eagerly, but also tentatively. Much of what the project achieved had to be invented wholesale, even if what was accomplished seems to be second-nature to us now. This includes the very basic notions of non-hierarchical inquiry and investigation: the metaphor of the thread wending its way through the labyrinth helped users break away from the parent-child relationships of previous computer-based systems, and introduced the idea of a userdriven system of discovery that has become the norm for online research and learning in the humanities and elsewhere. The digisig project, created and maintained by John McEwan, is not a portal like the Labyrinth, but rather an online finding aid created to support the work of medievalists working in sigillography. McEwan’s chapter demonstrates that computer-based systems of storing and linking knowledge can be far superior to print-based research tools, simply due to the nature of the sources being studied. Because the material evidence needed for the study of seals, seal matrices, and seal imprints often is widely dispersed, the computer’s ability to link common pieces of information across indexed systems allows scholars to isolate all the information on one seal in one place, rather than in several printed volumes that contain only one piece of the whole puzzle. Sigillographers and other scholars using sigillographic sources now can consider evidence more comprehensively because they are able to collect and examine their materials efficiently and exhaustively using the computer-based modes of inquiry that digisig allows. Rowan Dorin’s introduction to the Corpus Synodalium project likewise highlights the benefits of using digital tools to aggregate materials. The inspiration for his mapping and transcription project, however, came not from a sub-specialty’s methodology but rather from a research question that had been similarly thwarted by the limitations of printbased resources. Dorin’s musings on how far and in what format one late thirteenth-century conciliar decree—regarding the banishment of foreign moneylenders from certain locales—was received and circulated inspired him to seek out a born-digital solution to his query. His early investigations revealed that previous scholars had scarcely looked beyond the printed versions of the decree, instead taking it for granted that the ruling on banishment was both consistently and evenly disseminated. Since computer-based tools can collect and collate multiple versions of one text and plot them according to their geographic provenance far more easily than can print technologies, a digital database and map allowed him to attack his research question much more directly. Dorin’s profile of the successes as well as the practical challenges of collecting, transcribing, and then mapping ecclesiastical legislation reveals how knowledge creation so often occurs in the medieval digital humanities. In the confrontation between medieval and modern systems of information management, choices always have to be made, and it is often in weighing the alternatives that we arrive at a real sense of scholarly understanding. Jesse Torgerson’s chapter on the classroom-based Constantinople as Palimpsest initiative celebrates the pedagogical potential of the digital medium, and especially its power to challenge long-held truths about how teaching and learning should occur. In an essay that reveals sincere appreciation for both his students as co-learners and his collaborators as key to the project’s success, Torgerson embraces the revolution that online

8

Laura K. Morreale and Sean Gilsdorf

methodologies have brought to teaching, the mainstay of all humanities work. Aside from the immediate benefits of bringing the mapping project into the curriculum— students were engaged in their learning in a way that was more creative and therefore perhaps more committed than in previous models—the tools allowed students to make connections in the materials that he himself had not seen, since they stood outside of more entrenched historiographies on the imperial city. Torgerson’s appeal and invitation to fellow medievalists rests meaningfully on the medieval models of Eusebius and Accursius, who embraced new forms of knowledge sharing and thereby changed the intellectual worlds we ourselves study. In her closing contribution to the volume, “Life on—and off—the Continuum,” Lisa Fagin Davis chronicles her own slow journey into the world of computer-based methodologies, taking readers step by step and project by project as she added digital proficiencies to her medievalist skill-set. This journey is one likely familiar to many of us, since it was largely undertaken in an ad hoc fashion rather than as a formal plan of study, accomplished over the course of many years, with its various steps taken in response to ideas fellow medievalists put forth in conference presentations, one-off workshops, or cocktail-hour conversations. She generously recognizes how the building of her career, as well as her identity as a digital medievalist, has depended not only on how successful her many projects have been as on how meaningfully they engage with others in the field. Dr. Davis’ article re-introduces community as a fundamental part of both traditional and online scholarship, and reminds us how well digital tools are suited to creating and sustaining scholarly exchange.

Coming Together

The crucial matter of community brings us back, finally, to the purpose and genesis of this collection. For each of its contributors, digital tools and methods have served above all as a way to create, sustain, and expand scholarly conversations. As Everhart and Irvine make clear in the opening chapter, the Labyrinth was envisioned as a radical tool for broadening access to sources and resources across the disciplines of Medieval Studies, and for allowing medievalists across the country and even around the world to meet, learn from, and teach one another in new and exciting ways. Torgerson’s work on Constantinople as Palimpsest likewise grew out of a deep love for, and dedication to, teaching students how space and place shaped human interactions in the past. Far from being an exclusive, “techy” enterprise, his web-based teaching platform has offered dozens of students the opportunity to make history themselves through the medium of digital mapping and database creation. As such examples make clear, this book is not intended as an introduction to the newest and shiniest tools being used by researchers today, or a deep dive into the theory and philosophy of virtual learning and knowledge creation. These questions and issues have been addressed critically and extensively by many other scholars in the past decade. Instead, it is meant as a kind of “meet-and-greet,” featuring a group of colleagues from a range of disciplines working on a variety of projects. Like many of us, each of them remembers how scholarship functioned prior to the wholesale adoption (and even reli-



The Medievalist, Digital Edition

9

ance upon) digital tools. As a result, each of their essays provides valuable perspectives on what “thinking digitally” is, how it has evolved with the introduction of more and more digital tools into our work, and how they (and we) might bridge the gap between what came before and what comes next. Crucially, each of the projects described in the following pages has been catalogued by its creators, preserved in a stable and citable form that allows each of us, and future generations, to re-visit their work and appreciate its contribution to Medieval Studies more broadly. The histories chronicled in each chapter thus are not simply tales of a hero gaining and applying technological knowhow, important though this may be. Indeed, while many digitally-proficient scholars have worked valiantly to communicate the benefits and limitations of digital methodologies to their colleagues, many of us remain unsure or anxious about following their lead, suspicious of their motives, or frustrated by our own perceived lack of technical savvy. We hope that this volume, and the stories it has to tell, will make “digital humanities” a bit less threatening, and encourage scholars of all stripes to avail themselves of the exciting possibilities that digital tools and methods can provide to their own work as researchers, teachers, and colleagues.

Bibliography Albritton, Benjamin, Georgia Henley, and Elaine Treharne, eds. Medieval Manuscripts in the Digital Age: Digital Research in the Arts and Humanities. Milton: Taylor and Francis, 2020. Birnbaum, David J., Sheila Bonde, and Mike Kestemont. “The Digital Middle Ages: An Introduction.” Speculum 92, no. S1 (October 2017): S1–S38. Boffey, Julia. “From Manuscript to Print: Continuity and Change.” In A Companion to the Early Printed Book in Britain, 1476–1558, edited by Vincent Gillespie and Susan Powell, 13–26. Cambridge: Brewer, 2014. Davis, Matthew Evan, Tamsyn Mahoney-Steel, and Ece Turnator, eds. Meeting the Medieval in a Digital World: Medieval Media Cultures. Leeds: Arc Humanities Press, 2018. Davis, Rebecca Frost, Matthew K. Gold, and Katherine D. Harris. “Curating Digital Pedagogy in the Humanities.” Digital Pedagogy in the Humanities, 2020. https://digitalpedagogy. hcommons.org/introduction/. Edmond, Jennifer. “Collaboration and Infrastructure.” In A New Companion to Digital Humanities, edited by Susan Schreibman, Ray Siemens, and John Unsworth, 54–65. Malden: Wiley-Blackwell, 2016. Eisenberg, Merle, Sara McDougall, and Laura Morreale. “Middle Ages for Educators 1st Edition (June 2020) Archiving Dossier Narrative.” BodoArXiv Works, doi:10.34055/osf.io/4yedu. Fostano, Katherina and Laura K. Morreale. “The Digital Documentation Process.” The Digital Documentation Process, January 31, 2019. https://digitalhumanitiesddp.com/, accessed June 7, 2021. Herlihy, David and Christiane Klapisch-Zuber. Les Toscans et leurs familles: Une étude du “catasto” florentin de 1427. Paris: EHESS, 1978. Hiltmann, Torsten, Jan Keupp, Melanie Althage, and Philipp Schneider. “Digital Methods in Practice: The Epistemological Implications of Applying Text Re-Use Analysis to the Bloody Accounts of the Conquest of Jerusalem (1099).” Geschichte und Gesellschaft 46, no. 1 (June 2021): 122–56.

10

Laura K. Morreale and Sean Gilsdorf

Lässig, Simone. “Digital History. Challenges and Opportunities for the Profession.” Geschichte und Gesellschaft 46, no. 1 (June 2021): 5–34. McCormick, Michael. “The Birth of the Codex and the Apostolic Life-Style.” Scriptorium 39, no. 1 (1985): 150–58. Moelho, Anthony and R. Burr Litchfield. The Online Catasto Project. http://cds.library.brown. edu/projects/catasto/, accessed July 4, 2021. Moravec, Michelle. “Digital Historical Practices.” Journal of Women’s History 33, no. 2 (2021): 163–67. Morreale, Laura. “Medieval Digital Humanities and The Rite of Spring: Thoughts on Performance and Preservation,” July 18, 2019. BodoArXivWorks, doi:10.34055/osf.io/7p2t6. Owens, Trevor. The Theory and Craft of Digital Preservation. Baltimore: Johns Hopkins University Press, 2018. Partner, Nancy. “Report of the Delegate to the American Council of Learned Societies.” Speculum 91, no. 3 (July 2016): 881–85. Roberts, Colin and T. C. Skeat. The Birth of the Codex. Oxford: Oxford University Press, 1983. TaDiRAH (Taxonomy of Digital Research Activities in the Humanities). http://tadirah.dariah. eu/vocab/index.php, accessed November 10, 2021. White, Stephen D. Custom, Kinship, and Gifts to Saints: The Laudatio Parentum in Western France, 1050–1150. Chapel Hill: University of North Carolina Press, 1988.

Chapter 1

BEGINNINGS: THE LABYRINTH MEDIEVAL STUDIES WEBSITE DEBORAH EVERHART and MARTIN IRVINE

This chapter describes the origin, development, and significance of the Labyrinth—the first Medieval Studies website, as well as the first such site devoted to the Humanities—from its conception in 1992 to the present day. While the Labyrinth is still a living website as of the time of this publication (available at http://labyrinth. georgetown.edu), its story is one as old as the web itself. And, like the web itself, the story of the Labyrinth involves complex sets of connections and collaborations. There have been many, many contributors, both direct and indirect, to the success and value of the Labyrinth and its resources. The pages that follow are an attempt, admittedly incomplete, to chronicle those connections, and the complicated process by which they created a new tool for scholarly collaboration and creation. We encourage others to document the history of other electronic resources, and use this chapter to help them weave an even richer tapestry of scholarly history. The Beginning

In the spring of 1992, Deborah Everhart had become increasingly excited about the opportunities for connecting the work of medieval scholars using the many networking technologies that were becoming more widely available, including the nascent World Wide Web. In a letter to a friend on March 4, 1992, she wrote, “I couldn’t stop thinking about the networking we were discussing … professors of literature might be a key target … The time is right now.”1 The email went on to describe what this network (which she suggested naming “The Labyrinth”) might include: online journals, works in progress forums, conference listings, job lists, literature servers, image servers, music servers, a dictionary of literary terms; “the possibilities are endless,” she wrote, “and I think the profession is ripe for the switch to networking.” She was certain not only that the network could be built, but that it could be on line before the May 1994 meeting of the International Congress on Medieval Studies in Kalamazoo. In October 1992, Deborah met Martin Irvine at the Medieval Academy’s Committee on Centers and Regional Associations (CARA) meeting at Stanford University, where it became clear that numerous people (including Martin) were creating technology-rich projects ready for the kind of collaboration and integration that she had been envi1  Deborah Everhart, email to an unnamed friend, March 4, 1992.

Deborah Everhart, Credential Engine; Martin Irvine, Georgetown University.

12

Deborah Everhart and Martin Irvine

sioning. A few months later, in the spring of 1993, Deborah and Martin began working together to learn the technological architecture of the web and develop the Labyrinth. In May 1993, we presented an introduction to this work: “The Labyrinth: A Universal Electronic Information Network for Medievalists” at the 28th International Congress on Medieval Studies, Kalamazoo, Michigan. By December of 1993, we had uploaded the first Labyrinth web files onto a server at Georgetown University, thus creating the first website in the humanities. The Labyrinth officially launched on May 2, 1994 with a demonstration at Georgetown University, followed by a public event that same week (the “Labyrinth Computer Demonstration”) at the 29th International Congress on Medieval Studies in Kalamazoo. The rest, as they say, is history—a history recounted in the following pages.

Pre-Web Technologies in Medieval Studies

In the early 1990s, there were many electronic resources available that were both specifically and generally useful for Medieval Studies. In order to use these, however, one first needed to know that they even existed. You then needed to know how to use very cumbersome technologies, and the resources were useless to you unless you could understand and follow the detailed technical instructions for gaining access to a given resource that often included a login. In many cases, moreover, although resources were electronic, they were not available online. These basic challenges are reflected in early guides to scholarly computing such as Edwin Duncan’s “Medievalist’s E-Resources Directory,” which describes pre-web online resources and attempts to clarify how to access them.2 Duncan lists resources that medievalists certainly valued: eighteen medieval studies academic discussion forums and their archives, many of which had long histories and large numbers of subscribers, as well as sites available via Telnet, including the OTA (Oxford Text Archive), SEENET (The Society for Early English and Old Norse Electronic Texts), the Chaucer On-Line Bibliography, and the Dartmouth Dante Project. While these sites had significant bodies of materials, however, prior to the web one needed to have both scholarly expertise and technical expertise to use them. Therefore, in addition to providing instructions on (for instance) how to subscribe to an academic network, listserv commands, how to Telnet, or how to use FTP (File Transfer Protocol), Duncan also had to include separate detailed technical instructions for each site that he mentioned, such as these for the Oxford Text Archive: Probably the largest repository of accessible electronic texts on medieval topics is the Oxford Text Archive (or OTA). From the OTA you may obtain for no charge almost any of the major English or continental literary classics simply by filling out a form and signing a statement in which you swear to use the text for scholarly purposes only and not for

2  Everhart and Irvine, The Labyrinth, http://hdl.handle.net/10822/1061738: labyrinth/medieval_ databases/eduncan-dir1.html. Here and passim, Here and passim, references to Labyrinth files include the stable URL for the project’s archived materials at Georgetown University, https:// repository.library.georgetown.edu/handle/10822/1061738 followed by the path/file designators for the individual HTML files therein.



Beginnings: The Labyrinth Medieval Studies Website

13

class texts or for commercial use. Once the OTA people receive your application, they will send you electronically up to five requested texts, which you can then download onto your hard disk into your own word-processing software. Of course, you may never need an electronic version of a text, but if you do, this service can be invaluable. For more information send to [email protected] or to [email protected] the following message: “get humanist filelist” or, if that presents a problem, send an e-mail message requesting more information to [email protected] …3

In short, while the stage was set—ambitious and valuable electronic resources were being developed by medievalists, many of whom were already adept at working online— pre-web technologies made their work more difficult and cumbersome than what the web would soon offer.

The Vision Becomes a Reality The Power of Metaphors

When we first started describing the opportunities of the web to medievalists in 1993, very few people had any idea how a “web” of online resources would work, or why it would be valuable. We needed to communicate even the most basic principles; hence the usefulness and the beauty of the metaphor of the labyrinth. Since all medievalists know the story of Ariadne and Theseus, the winding passages of the labyrinth in classical mythology, and the thread that brought Theseus back to his love, we used it as a metaphor for visualizing both the complexities of the web and the reliable “way home” that would be a key component of our website. We used this in our presentation and publication descriptions of the Labyrinth, as well as in a web page explaining the name and navigation of the site: Ovid says that Daedalus built a house in which he confused the usual passages and deceived the eye with a conflicting maze of various wandering paths (in errorem variarum ambage viarum) (Metamorphoses 8.161):

… so Daedalus made the innumerable paths of deception [innumeras errore vias], and he was barely able to return to the entrance: so deceptive was the house [tanta est fallacia tecti] (8.166–68).

The Labyrinth project on the World Wide Web is designed to allow you to make your own Ariadne’s thread through the maze of information available on the Internet. And you will always be able to find your way back by choosing the “Return to Labyrinth Home Page” link at the end of each Labyrinth document.4

The metaphor not only helped people understand website navigation, but also appealed to journalists, provoking headlines like “On the Net with Ancient Heroes” (Chronicle of Higher Education, 1995)5 and “Out of the archives, onto the Internet” (Stanford University 3  Ibid.

4  Ibid.

5  Thomason, “On the Net with Ancient Heroes,” The Washington Post, November 2, 1995, C7.

14

Deborah Everhart and Martin Irvine

News Service, 1995).6 Such media pick-ups contributed to the Labyrinth’s immediate popularity and recognition: since others were telling and promoting our story, we didn’t need to do all that work ourselves. We also realized that aesthetics matter. Therefore, we used elegant visuals that were appealing to all audiences, but resonated particularly with medievalists. The key image that became the unofficial “brand” of the Labyrinth website is the labyrinth in the floor of Chartres Cathedral (figure 1.1). We also used complementary tile and brick patterns for backgrounds, icons, and navigational markers (see for Figure 1.1: Image of the floor labyrinth at example the Labyrinth images “laby-sm2. Chartres Cathedral (ca. 1215–1221 CE), gif,” “maze.gif,” and “navbutton.jpg”), all of reworked, which became the unofficial “brand” image for The Labyrinth. which combined to give the site a coherent style as a “place” with medieval character. Finally, in a prescient example of the importance of user experience design, we created consistent navigation footers with exactly the same text and mnemonic visuals on every page, giving our website explorers the “Ariadne’s thread” that we had promised would help them find their way “home” from anywhere in the site. At the time, this consistent footer had to be manually written into every page, but we were sticklers for detail in both the HTML code and the beauty of the site. Contextualizing Web Concepts

This visualization of a labyrinth also gave us a way to communicate the difference between earlier technologies and the web. Unlike earlier internet protocols such as Gopher, the web is not a hierarchy; since it does not have a “top” or a “root,” it was important to assure people that they would not get “lost” in this new online experience. But most people were also “lost” with regard to the intellectual significance of this new form of communication and publication. Simple metaphors were not enough to explain many crucial innovations, since many complexities of the web that we take for granted today were novel in 1993. One of our earliest documents explaining the web to medievalists, “The Labyrinth: An Electronic Information Network for Medieval Studies” works through a number of challenging key concepts. It’s worth noting that our decision to refer to this as an “article” defied traditional boundaries at the time, since it was never part of a formal publication, but was very widely circulated, referenced, and remixed by us and others. The article emerged from the innovative Interscripta project, a topical 6  “Out of the archives, onto the Internet,” Stanford University News Service, 16 August 1995; https://news.stanford.edu/pr/95/950816Arc5114.html.



Beginnings: The Labyrinth Medieval Studies Website

15

online forum created in 1993 by Patrick Conner, William Schipper, and us. As the forum authors explained at the time, Each six-week discussion on Interscripta focuses on a specific topic in a forum much like a round table discussion. Topics are proposed by scholars, then the advisory board of Interscripta reviews and chooses topics, schedules discussions, and oversees the general operations of the forum. Each topical discussion is moderated by an individual scholar, and at the close of the discussion, the moderator shapes the material into an article which is distributed to all participants for review and commentary before its final revision. Finished articles are then submitted to the on-line journal Interscripta for expeditious publication.7

The Interscripta article about the Labyrinth was, to our knowledge, the first to explain the web and its potential value for medievalists. It is worth emphasizing that the article was not meant to offer technological explanations or guidance, but rather to provide an intellectual context for new ways of communicating and publishing online that were emerging at the time, and to demonstrate how webs of resources could enable new forms of scholarship. Intertextuality and cross referencing always have been familiar to scholars, and hypertext provided a new way of communicating these connections: Data available to us electronically presents itself in a form extraordinarily different from that of the printed book; rather than following the strict linear order of lines on a page and pages in a book, electronic information communicates the intricate links between and within texts, the constantly shifting meanings of words and other bits of information as they appear in different contexts. Hypertext, the technological method whereby context after context can expand out from a single word, provides us with a new way of maneuvering and visualizing information, promising to supplant the printed book and rewrite not only our understanding of textuality, but also our concepts of authorship, pedagogy, and ultimately theory itself.8

As we realized, the connectedness that the web made possible was unprecedented and difficult to grasp. We used metaphors such as the “desktop”, and comparisons to traditional research methods, to explain its impact: The organization of diverse resources under one easy-to-use hypertext interface is revolutionizing the way we and our students do research. W3 technology allows searches across databases and facilitates customized manipulation of data. Files residing on servers around the world thus become part of one’s desktop resources. Fast, powerful indexing programs allow users to process massive amounts of data with a single command. These tools for sorting and analyzing data not only save time, but also allow users to perform searches and make connections that would not have been possible before.9

Nevertheless, we did not shy away from how decentring these technologies could be, even fundamentally changing our concepts of “mastery” and “authorship” due to the speed with which data now could be produced, shared, and manipulated in a rapidly expanding number of venues. As we noted, since “scholars may no longer operate under

7  Everhart and Irvine, The Labyrinth, http://hdl.handle.net/10822/1061738: labyrinth/article. html. 8  Ibid.

9  Ibid.

16

Deborah Everhart and Martin Irvine

the illusion that they have mastered all relevant information in their field … we can adopt a new metaphor, that of the skillful navigator, the explorer, the user of sophisticated tools.” The internet, we continued, both metaphorically and practically deconstructs the illusion of the academic master and the authority of the author. Because the manipulation of information in the Web is user- centered, the new scholar does not follow the author’s guidelines, does not read texts chronologically or entirely. The Web scholar makes decisions and maneuvers data in ways directly relevant to his or her own research and intentions, decentering the traditional concept of authorship by producing a new hypertext of connections among fragments of authors’ works … And the concept of authorship is further decentered by group practices, as new forums like Interscripta facilitate collaborative writing and blur the distinctions of identity and intellectual property.10

It is no surprise that even in 1993 scholars were struggling with the impact of online electronic resources and how they would affect their own research methods, autonomy, and publication strategies, just as they do today. Since relatively few people were reading resources online at the time we made the Interscripta article available, however, we also spread our message in more traditional scholarly venues. In May 1993, we spoke at the 28th International Congress on Medieval Studies in Kalamazoo which was the largest annual gathering of medieval scholars from all disciplines, offering the ultimate audience for professional outreach and recognition. The key points of our presentation, “The Labyrinth: A Universal Electronic Information Network for Medievalists,” were largely the same as those in the Interscripta article, and were well received by a large audience. The Labyrinth became a hot topic at that year’s conference, which set the stage for the technology launch of the Labyrinth in December 1993, and its live demonstration at Kalamazoo in 1994. The Web Expands to the Humanities

Thanks to the work of medievalist visionaries who paved the way for new technologies, combined with our rapid rollout of the Labyrinth at the end of 1993, Medieval Studies became the first field in the humanities to have its resources connected via the web. We would have had a scant amount of content to put in the first Labyrinth web pages had it not been for Gopher, listserv, e-text repositories, Telnet library access, and other medieval and scholarly resources already available online. Prior to 1993, however, the use of the newly invented web protocols was already expanding in the sciences. Perhaps the best known of these was the World Wide Web (W3) interface developed at CERN, the European Particle Physics Laboratory in Geneva, as a fast and convenient way for physicists to share data and work collaboratively. Inspired by such initiatives, we knew that access to relevant scholarly resources organized and connected using hyperlinks could profoundly change the exchange of information and the speed of research in the humanities as well. In this vein, we noted, the Labyrinth could not only provide an organizational structure for Medieval Studies, 10  Ibid.



Beginnings: The Labyrinth Medieval Studies Website

17

but … also serve as a model for similar, collaborative projects in other fields of study, thus opening up new possibilities for the academic community of the twenty-first century … For medievalists, the Labyrinth’s main menu will be the natural starting point, but the Labyrinth’s reciprocal connections to other disciplinary servers will allow medievalists to branch out into other fields of study and … connect medievalists to the rest of the global Web community.11

Previously, other non-web-based medieval studies servers had to be accessed separately. By connecting them, the Labyrinth made them more accessible and promoted online scholarship through their virtual combination with other resources. This was a highly effective use of hyperlinks connecting distributed sources that could still remain autonomous, allowing users to consult a potentially unlimited range of materials without requiring those materials to be kept in one location. Of particular value was the Labyrinth’s ability to provide access to many different types of digital resources—texts, images, bibliographies, discussion lists, library catalogs, and so forth—while maintaining the autonomy of those resources and the projects and archives that contained them. For example, the Labyrinth offered users a link to the Dartmouth Dante Database (ddd), an early public-access compendium of Dante’s works as well as Dante scholarship, without interfering with the database itself. While users still could access sites like the ddd directly, the Labyrinth now made this valuable resource known and available to a much wider potential audience, who did not need to know in advance that the ddd even existed, much less how to establish a remote connection to it. The Labyrinth thus served as a heuristic system, allowing its users to find and explore materials that were previously unknown or inaccessible to them. The amount of these now-networked materials rapidly increased in the following years, as did the number of people employing the Labyrinth for their teaching and research. Other scholarly websites in the humanities quickly borrowed from our model, seeking our expertise and developing rich scholarly collaborations that connected medieval studies resources to other fields—among them, the Electronic Archives for Teaching the American Literatures website directed by Randy Bass,12 which in 1994 became one of several “Centers of Excellence” besides the Labyrinth developed at Georgetown University. To be sure, not everyone understood our goals. For example, despite strong references, our 1993 application for a National Endowment for the Humanities (neh) grant was rejected—funding that would have enabled us to add corpora of primary texts in Old English, Middle English, Old French, Old Norse, Old High German, and Latin to the Labyrinth “library.” Although feedback from the anonymous reviewers was generally positive, they were concerned that the proposal was too ambitious (admittedly true) and that medievalists might not use these resources (emphatically not true, as it turned out).13 Fortunately, administrative leaders at Georgetown University, including a number with humanities backgrounds, saw the value of the Labyrinth and funded our work 11  Ibid.

12  Bass, Electronic Archives for Teaching the American Literatures.

13  In an ironic twist, the neh ultimately recognized the Labyrinth (in 1997) as “one of the best sites on the Internet for education in the humanities” (EDSITEment award letter from Sheldon

18

Deborah Everhart and Martin Irvine

in various ways from its inception.14 In addition to a start-up grant from the university for the initial demonstration project, the Labyrinth also benefited from the effort of interns funded by Georgetown’s Medieval Studies Program, from the generous assistance of Georgetown IT staff members in developing and expanding the website, from the support of Georgetown librarians in the archiving process, and from Deborah’s own appointment as Georgetown’s first webmaster. Early Web Technologies

When the Labyrinth was created, the technologies available for building websites were extremely limited and difficult to master—not surprising, given that in early 1993 there were only fifty or so websites in the entire world.15 We wrote the Labyrinth’s HTML code by hand in a text editor, saved it in plain text files, and uploaded them to the Georgetown Unix server via FTP command line interfaces. The files needed to be created in organized sets with a predefined folder structure, so that we could keep resources organized and anticipate the growth of the Labyrinth. Links between these files also were written by hand in HTML code, and then tested manually once the files had been uploaded; any typo or change to the file structure had to be identified and corrected through the same manual process. Links to servers and resources outside our set of HTML files often needed to be accompanied with step-by-step instructions to help people understand how to access and navigate the Gopher or Telnet site to which we were linking; these instructions themselves often changed, however, generating other types of errors that we needed to track down and address manually. In turn, accessing the web was a major challenge in its own right. In early 1993, the small number of people accessing the web via personal computers did so primarily through command-line browsers. Use of the web was so new and rare that we had to explain to people how to access the Labyrinth, even offering one-on-one advice to potential users about processes that we now take for granted, such as how to get to a starting point web page. We also showed people how to use the web by demonstrating simple but powerful affordances in Labyrinth web pages, most notably hyperlinks—a simple but highly useful navigation device that most people had not experienced before. In our introductory article, we included footnotes with clickable hyperlinks, from which the reader could return to the main text by selecting “return to article”—a practical and immediate demonstration of how to navigate through the Labyrinth itself.16 We were, in essence, teaching the medieval scholarly community how to use the web. Hackney, Chairman of the National Endowment for the Humanities, to Martin Irvine and Deborah Everhart, May 6, 1997).

14  Worthy of special mention and thanks in this regard are Robert (Bob) Lawton, then Dean of Georgetown College, John (Jack) DeGioia, then Associate Vice President and Chief Administrative Officer for the Main Campus, and Richard (Dick) Schwartz, then Dean of the Graduate School, all of whom understood and supported our vision from the beginning.

15  For a concise overview of dates and developments in early web history, see Connolly, “A Little History of the World Wide Web.” 16  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/article.html.



Beginnings: The Labyrinth Medieval Studies Website

19

Figure 1.2: Martin Irvine and Deborah Everhart at Georgetown University with an early Labyrinth page

In late 1993, the release of personal computer versions of the Mosaic web browser, with its graphical user interface, made the web much more popular, owing to the ease of navigation and the ability to incorporate images within internet pages. Even before most people had access to a graphical web browser, however, we fully expected that the Labyrinth would be a media-rich environment, extending well beyond the simple top-of-page and navigational images used in the first Labyrinth website. Our plans for the Labyrinth thus included providing access to a wealth of digitized images, including digital images of manuscript illustrations and text, medieval art, architecture, maps, and objects from daily life and material culture, anticipating the critical role that digital imaging and processing would play in fields such as manuscript studies. These plans quickly became realities, as the rapid evolution of web browsers and the adoption of the web not only for scholarly but also for commercial uses ushered in a whole new array of opportunities. By 1995, the Labyrinth supported hypermedia, hybrid online/offline activities, and how-to help for non-technical people to develop their own web-based resources. In a very short time, the Labyrinth had moved from an ambitious vision to a rapidly growing and increasingly popular online resource.

Growth and Adoption

In our neh proposal, written in August 1993, we estimated that at least eight hundred active users would be accessing the Labyrinth site in the first few months following its anticipated launch in December 1994. As we discovered after the site went public (in May 1994, more than six months earlier than originally planned), our neh proposal estimate was wildly off the mark. User statistics compiled during the Labyrinth’s first month

20

Deborah Everhart and Martin Irvine

of operation, and documented in an August 1994 Coalition for Networked Information (CNI) workshop project report, revealed 9300 web requests to the Labyrinth from educational, governmental, military, and commercial sites in twenty-five different countries. In only its first month, we concluded, the Labyrinth already was “reaching an enormous audience, and our correspondence with an incredibly diverse range of interested people demonstrates the great potential for developing World Wide Web disciplinary servers for other subject areas.”17 Just one year later, in August 1995, the Labyrinth received its millionth file request. By 1998 it was averaging over six thousand daily hits. The immediate success of the Labyrinth was due not simply to our ability to understand and act upon current technological trends. Just as critically, it did not have to start from scratch. When the Labyrinth went live, medievalists had developed a large number of enormously valuable electronic resources already, including e-texts, databases, and vibrant online communities, created in a variety of formats and housed on servers requiring different logins and protocols. They were primed, in other words, to adopt new, web-based technologies that made it easier to access and connect these previously disparate resources. The Labyrinth in particular provided three key advantages: its use of open standards for hypertext links; its ability to organize resources from different sources; and its provision of open, common access methods. Even in its initial iteration, the Labyrinth provided easy-to-understand menus and navigation tools that made it easy for users to access existing as well as new resources that previously had existed in isolation, including the Oxford Text Archive, the MALIN History Text Archive, Curia Irish Text Archive, SEENET texts, the Oxford English Dictionary, the Vulgate Bible, Studies in the Age of Chaucer Bibliography, the Dartmouth Dante Database, the University of Virginia Electronic Text Center, the Bodleian Library, and a number of medieval-themed Listserv archives. By coordinating, collecting, and organizing such materials, the Labyrinth made it easier for new users to find and access them, just as the existing user base of those resources provided a new audience for the Labyrinth itself. Supporting Web Development in Medieval Studies

Of course, building the Labyrinth was not a simple matter of creating menus of existing resources. Our goal, rather, was to create new resources ourselves at the same time as we supported collaborative development by others. Besides the many explanatory resources in the Labyrinth, we also contributed scores of our own manuscript transcriptions, translations of primary texts, teaching materials, and scholarly writings, and helped others to contribute and collaborate. Since there were few resources in the early 1990s for those who wanted to create web sites or pages, we also ended up teaching many people how to do so and supporting their subsequent efforts. In particular, we provided explicit, step-by-step guidance for specific audiences, including medieval professional associations, as well as on-site demonstrations, seminars, and consultations for scholars and institutions interested in developing similar projects in Medieval Studies as well as other humanistic fields. As we noted at the time, 17  Everhart and Irvine, “The Labyrinth: A World Wide Web Disciplinary Server, Project Number 04 – 1994.”



Beginnings: The Labyrinth Medieval Studies Website

21

the sharing of such information will facilitate the coordination of disciplinary servers in the humanities, such that we may avoid duplicated efforts and incompatible systems. We hope that our experience in developing the Labyrinth will be of broad use throughout the humanities as our profession navigates the transition from traditional forms of research and pedagogy to the technological possibilities of the coming decades.18

At the March 1995 meeting of the Medieval Academy of America, we presented a paper addressing the long-term issues of scholarly development on the World Wide Web, arguing that Medieval Studies organizations and associations should develop World Wide Web resources in their own scholarly communities, resources which in turn could be coordinated and made easily accessible through the Labyrinth. To facilitate this development, we organized a set of materials, including general internet and web introductory resources and Deborah’s Georgetown University Guide to World Wide Web Development. We even created a “fill in the blanks” HTML file that could be used as a template for organizations, who simply needed to substitute their information for the form’s general statements before either uploading the form to a web server at their institutions or e-mailing it to us for inclusion on the Georgetown web server. We also wrote a letter to Medieval Studies organizations, explaining the importance of web development and providing both intellectual context as well as technical guidance on how to create and store files in the new HTML format.19 Many of these organizations responded positively to our initiative, and some initially hosted their web pages on the Labyrinth server (perhaps most importantly, the Medieval Academy itself, whose website was hosted and maintained by us on the Labyrinth server from 1997 until 2000). In 1999, the Labyrinth directory of Medieval Studies organizations, societies, and centres contained 125 entries, of which 70 had websites.20 Our expertise in web development made many other professional contributions possible. Deborah led web development, policy, and processes at Georgetown University from 1994 to 2000, going on to lead product strategy and invent critical features of learning management systems, portfolios, badges, and digital credentials. Martin wrote Web Works (1996), an introductory book that taught thousands of people how and why to use the web. Martin went on to found Georgetown’s Communication, Culture, and Technology graduate program, the first interdisciplinary program of its kind. In many ways, then, our formative experiences in early web development served as a foundation for careers that have promoted and advanced technological innovation far beyond the boundaries of Medieval Studies itself.

18  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ article.html. 19  See Appendix to this article.

20  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ organizations/index.html.

22

Deborah Everhart and Martin Irvine

Adoption by Medievalists and the General Public The rise in popular interest in the web in the early 1990s, aided by the launch of imagefriendly browsers such as Mosaic, coincided with our ability to show people something interesting. The intriguing intersection of medieval and tech drew attention not only from the community of medievalists, but also from scholarly publications and the popular press, leading to catchy headlines such as one from the Washington Post in 1995, “On the Net with Ancient Heroes: Perturbations, pleasures and predicaments on the information superhighway.” As the article’s author declared, “The Middle Ages have been hyperlinked,” before elaborating on some of the materials that could be accessed via the Labyrinth, including Beowulf manuscripts, drawings of King Arthur, maps of the Hundred Years War, and photographs of Viking rune stones.21 The Labyrinth’s scholarly importance already had been acknowledged in leading educational publications such as the Chronicle of Higher Education, which in May 1994 observed how A Dante scholar might use the Labyrinth to connect to an Italian computer to see the text of The Divine Comedy and then connect to a computer at Dartmouth College to read criticisms of that and other works … The project also offers a collection of pedagogical tools … [including] “Interscripta,” a mailing list of 400 people that is devoted to topics of interest to medievalists.22

A year later, Stanford University News Service noted the Labyrinth’s importance to medievalists, as well as the leading role of Medieval Studies within the nascent field of what would come to be known as digital humanities: In the rush to the World Wide Web, researchers in matters medieval and Anglo-Saxon apparently are leading the scholarly pack … Since it was established a year ago, Labyrinth has served more than one million files … to some 60 different countries … and already has been rated among the top 5 percent of web resources in an independent survey conducted by Point Communications Corp … Stanford’s [George] Brown, a specialist in Old English and Anglo-Latin who has published on a wide range of medieval topics, says that the isolation that many medievalist scholars feel has fueled the current electronic drive. “Most of us got onto computers out of sheer necessity,” Brown says. “And once you’re on the computer, it’s pretty easy to take the next step of networking.”23

As Deborah told the article’s authors, “We’ve always been driven. In medieval studies you can’t just be interested in text. You have to be interested in languages and art history and lots of other disciplines. So adding one more new discipline isn’t a problem for us.”24 Within a few years, the Labyrinth was included in a large number of guides and directories of web resources, including those compiled by W. W. Norton, Encyclopedia Britannica, the National Endowment for the Humanities, and the World Wide Web Virtual Library, hosted by the W3 Consortium at MIT. Attention from these many audiences translated into traffic on the Labyrinth. These visitors, moreover, were not just 21  Thomason, “On the Net with Ancient Heroes.”

22  DeLoughry, “Georgetown U. Offers Gateway to Resources for Medievalists,” Chronicle of Higher Education, May 11, 1994, A21. 23  Stanford News Service, “Out of the archives, onto the Internet.” 24  Ibid.



Beginnings: The Labyrinth Medieval Studies Website

23

casual passers-by; people who visited the Labyrinth were able to quickly access rich and diverse resources, leading them to stick around and spend significant amounts of time on the site. Even as the World Wide Web grew to include hundreds of thousands of websites, offering a myriad of ways for people to spend their time online, use of the Labyrinth continued to grow, averaging over six thousand hits per day from early 1998 well into 2000.

Building Communities and Leveraging the Web

While popular attention was welcome, our primary audience for the Labyrinth was the community of Medieval Studies scholars and students, whose needs we intended to serve by offering a rich and diverse set of scholarly, pedagogical, and professional resources. Building upon the wide range of electronic resources and online communities already available to medievalists in the early 1990s, we thus strove to explicitly create structures for community and professional engagement. Critically, the heuristic nature of the Labyrinth’s collected resources helped people find not only Medieval Studies sources and materials, but also collaborators and organizations that would help them achieve their goals, in the process levelling the playing field by providing access to as much professional information as possible. Indeed, equitable access to resources was one of the most important outcomes of the Labyrinth, and one of our early guiding principles. As we explained in our 1993 article, This project and others like it will greatly enhance research and pedagogy by making an enormous amount of data universally available, easy to access, and convenient to use. The practical implications of hypertext networks are manifold. The Web is already fostering unprecedented collaboration among scholars, providing discussion forums which are channels for efficient communication and allowing researchers to undertake joint ventures that would not have been possible before. Students, both graduate and undergraduate, often participate in these discussions, furthering not only their own education but also the efforts of their educators. The Web democratizes the process of sharing ideas and information.25

From the outset, moreover, we envisioned a wide range of community resources, including information on professional organizations, conference announcements and calls for papers, academic newsletters, bibliographies, journal tables of contents, job bulletins, fora on work in progress, virtual meeting places for those working on joint projects, pedagogical materials, and job bulletins. While not all of these desiderata were realized, many did end up coming into being on other sites or networks; moreover, the Labyrinth and later Medieval Studies websites became vital sources of opportunities for collaboration and idea sharing, serving a large and diverse group of students, teachers, and researchers.

25  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ article.html.

24

Deborah Everhart and Martin Irvine

Publishing

One of the most important professional achievements of the Labyrinth was to provide a home, organizing principles, and a model for Medieval Studies scholarly organizations on the web. In turn, bibliographies—both individual and as part of larger databases— became more accessible by being listed and linked on the Labyrinth’s Bibliographies page, which aggregated information about bibliographies that many people would have had difficulty finding, including many that required passwords on proprietary sites that used older technologies such as Telnet and Gopher. Wherever possible, we provided not only links, but also instructions for logging in to these resources, saving scholars hours of frustration. While we never realized our ambitious goal of including tables of contents of journals on the Labyrinth, we did include a directory of publishers responsible for books and journals in Medieval Studies, which while not exhaustive nonetheless provided a vital starting point for scholars wishing to survey the state of the field as well as to find possible publishing venues for their own research. Within only a few years of its launch, the Labyrinth itself came to be seen as a valuable place in which to discover and disseminate scholarly publications; as early as the May 1996 International Congress of Medieval Studies meeting in Kalamazoo, two works—William Schipper’s “The Labyrinth Guide to Manuscript Repositories” and the Bryn Mawr Medieval Review—were publicized as accessible via the Labyrinth’s new Scholarly Publications page, which would grow to include links to and records of a growing number of monographs, journals, reviews, and websites.26 Conferences and Discussion Forums

By aggregating access to sites and forums, the Labyrinth also encouraged and enabled scholars to collaborate, develop and post their work online, attract interest in their projects from publishers, and even attend online conferences. One notable example of the latter was Cultural Frictions, a hybrid online/face-to-face conference hosted by Georgetown University in October 1995 which invited participants to investigate the ways that cultural discourse is deeply implicated in the origins of its production. How are the objects we study entwined with the modes of their critical articulation? What does cultural studies offer medieval studies? … Post-modern theory is also beginning to notice the impact of the new networked hypermedia environment of the World Wide Web on literary studies and the humanities, and the Web as a new context for cultural studies is both a topic for discussion as well as the medium for transmitting this discussion worldwide.27

In contrast to the traditional conference model, Cultural Frictions allowed active and acknowledged participation and interaction through online discussion as well as physical attendance at the campus venue, while presenters were able to publish their papers 26  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ professional/scholarly_pubs.html. 27  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/conf/ cs95/index.html.



Beginnings: The Labyrinth Medieval Studies Website

25

in an online collection of conference proceedings hosted on the Labyrinth, one that encouraged feedback and comments from readers around the world via a web-based comment form. As the organizers noted at the time, this conference not only was the first event devoted to the topic of Medieval Cultural Studies, but also the first worldwide, interactive conference in any humanistic field. Pedagogical Resources

Both the web and prior technologies such as e-text databases provided unprecedented opportunities to incorporate medieval source material into courses and student research. This fostered a community of innovation among Medieval Studies teachers, with access to and aggregation of resources supported by the Labyrinth. At the outset, we anticipated how pedagogy could evolve with these new tools: Labyrinth [pedagogical] forums will provide not only shared information among scholars teaching similar subject matter, but also on-line seminars and other opportunities for distance learning … We will offer sample syllabi and course outlines, study questions and writing assignments, e-anthologies, bibliographies, aids for learning medieval languages, even multi-media teaching platforms … based on contributions from users, representing the widely differing strategies, agendas, and teaching tools of faculty from around the world.28

As we had envisioned, the Labyrinth soon offered access to online courses, course materials, language learning aids, and entire corpora of primary texts in the Labyrinth Library, serving as the only point of access to a number of primary and secondary resources for teaching, research, and scholarship.29 As the web became increasingly complicated and littered with material of dubious quality and reliability, moreover, the Labyrinth continued to be the site that teachers and their students—both medievalists and those seeking a gateway to medieval resources—could trust. In recognition of this fact, the Labyrinth was added to the National Endowment for the Humanities’ EDSITEment directory, a tool designed to help teachers “streamline your time online by connecting you quickly to websites with real educational value, websites screened by a rigorous academic review process and endorsed by a distinguished panel of educators and parents.”30

Scaling and Sustainability Models

Although the Labyrinth was designed for collaborative contributions and distributed development, the unforeseen scale, speed, and complexity of its growth required web management structures that had not yet been invented. Previous technologies such as Gopher depended upon a centrally controlled hierarchical structure, one more suited to top-down management than to collaborative expansion in numerous directions.

28  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ article.html. 29  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ pedagogical/pedagog.html. 30  National Endowment for the Humanities, EDSITEment promotional flyer, 1997.

26

Deborah Everhart and Martin Irvine

Online databases and e-text corpora typically had a single owner or governing organization. Comparable commercial ventures were organized according to product and profit structures. Faced with the challenges of unprecedented scaling, we were forced to forge ahead, aided in our efforts by the help and encouragement of numerous colleagues. Scholarly Contributions

From the beginning, we insisted that the Labyrinth remain a scholarly, non-commercial endeavour, even when the site’s rapidly growing user base offered an opportunity for advertising revenue. While the Labyrinth might well have benefitted from the increased resources that such an income stream could provide, we ultimately decided that commercialization would just as likely dilute and even corrupt the scholarly contributions upon which the site depended and which drew users to it in the first place. Without commercial sources of income, and in the absence of substantial grant funding or institutional budgets, the Labyrinth thus lacked a paid staff, depending principally instead upon the unpaid labour of its creators themselves as well as in-kind technical and intern support from Georgetown University. Given these circumstances, the Labyrinth never could have succeeded without the time, effort, and expertise given to it by many colleagues, in particular its international advisory board of technologists and medievalists, who in addition to providing invaluable practical and theoretical guidance also agreed to weave their own scholarly work into the Labyrinth’s rich tapestry. From the beginning, board members helped to collect and manage resources for parts of the Labyrinth that were within their own area of expertise, like William Schipper, who helped to curate the subject page for Anglo-Saxon Culture.31 We expanded this model by recruiting volunteer subject area experts and page editors like R. Allen Shoaf, who for several years edited the Scholarly Publications page, compiling online journals and websites relevant for online publishing.32 Interns from the Georgetown University Medieval Studies program also collected, curated, and maintained resources under our guidance, while in some cases even undergraduate students had an opportunity to contribute to the creation of the Labyrinth. In each case, we followed a simple but useful model by which numerous people collected resources and feedback and manually edited individually-curated web pages. The resulting pages were naturally (and charmingly) different and varied in the structure of their content, although each featured a consistent Labyrinth navigation structure. Nevertheless, as the Labyrinth grew to include hundreds of HTML and image files for pages providing access to thousands of resources across the web that were constantly moving, changing, and disappearing, this original approach—and the link maintenance that it required—proved to be neither sustainable nor scalable. 31  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ subjects/british_isles/anglosaxon/.

32  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ professional/scholarly_pubs.html.



Beginnings: The Labyrinth Medieval Studies Website

27

The Labyrinth as a Web-Based Catalogue

As more and more Medieval Studies and cognate academic websites matured, the Labyrinth’s role shifted towards one of cataloguing resources rather than hosting multiple individual pages. For example, the Medieval Academy ceased using the Labyrinth as the hosting site for its organizational website; more generally, scholars increasingly were able to publish resources on their own institutions’ websites, rather than relying on the Labyrinth for this service. By 1999, therefore, it was time for a logical and technological restructuring. In order to shift to this new cataloguing role and clean up the disparate subject pages, we turned to a newer technology designed for indexing structured data: the web database server, specifically Adobe’s ColdFusion. This enabled us to store all of the links in a database, organized with categories and other metadata, and generate web pages “on the fly” based upon the user’s selection of metadata or application of search parameters. The automatic generation of a web page based on a user’s specific requests is something that we take for granted today; after many years of managing individual pages manually, however, we found this new technology to be transformative. Unlike its predecessor, the ColdFusion Labyrinth site essentially had only two pages: the home page, where users initiated queries, and the results page, which listed the links that matched those queries. The home page also included links to predefined results for each of the top-level organizational categories, allowing people to have a heuristic experience similar to that of browsing the pages on the earlier Labyrinth site. This was particularly important as a teaching tool, helping students explore and learn about Medieval Studies without prior knowledge of the relevant categories. Transitioning to ColdFusion required defining those categories and establishing a metadata structure to give people useful and intuitive ways to find what they needed. For this, we went back to an idea that had been impractical in the early days of the Labyrinth due to lack of web content: the Library of Congress Classification (lcc). Although our original plan had incorporated lcc, for many years Medieval Studies web content was too “lumpy”, with a few categories full of resources and many other categories empty of any content. Moreover, the Labyrinth had evolved by adding what was new, not by systematically cataloguing what existed, exacerbating the tension between a tight lcc structure and a loose practice of sense-making. The ColdFusion site used categories and subcategories based on lcc, as well as types of materials, as an organizing structure, drawing upon metadata in the ColdFusion database. Users could search for materials either by entering keywords or by using any of these metadata. The “types of materials” filter, for example, allowed users to search for Archaeological Materials, Architectural Materials, Audio, Bibliographies, Catalogues and Handlists, Course Materials, Databases, Discussion Lists and Forums, Glossaries, Images, Maps, Materials for Children, Organizations, Primary Texts, Secondary Texts and Articles, and Video. By this point, there were sufficient Medieval Studies resources on the web to include materials in all of the categories; indeed, we even undertook a massive expansion by adding hundreds of new links. To be sure, the prior Labyrinth resources and the nature of the web in general resisted our efforts at clear classification. For example, many sites were collections spanning multiple categories, making it difficult to correctly categorize

28

Deborah Everhart and Martin Irvine

the resources they contained. Likewise, it was difficult to capture metadata about the types of materials on any given site. Nonetheless, the improved management of Labyrinth content provided by the ColdFusion database far outweighed any disadvantages it posed. The ColdFusion server made maintenance far easier, providing a fast and efficient way to scan for broken links. It also made it easier to “crowdsource” content through a form that allowed users to offer corrections to existing links and suggest new ones. This web form captured the relevant information in the database structure, so that we could simply review and approve changes and additions. In theory, we had reached a point where the Labyrinth could grow and be maintained largely through semi-automated public input rather than manual scholarly curation. As it turned out, however, public input was not the solution we had hoped it would be.

The Wild, Wild Web

Twenty-five years after he invented the web, Tim Berners-Lee reflected on how it had empowered and connected people around the world, transforming “the way we talk, share and create.”33 Yet when asked, “What was one of the things you never thought the internet would be used for, but has actually become one of the main reasons people use the internet,” his response was unexpected: “Kittens.” None of us expected kittens. All of our high-minded visions for the scholarly communities on the web and equitable, organized access to the best resources for everyone from beginning students to advanced researchers are dramatically overshadowed by the popular uses of the web.34

The growth of the web into a cacophony of materials from innumerable sources does not, of course, preclude serious scholarly uses. It does, however, require serious web users and developers to gain the skills they need to discern the source, authenticity, veracity, and value of new, ever-changing types of resources. For us, this meant that the Labyrinth needed more intense editorial vetting processes, at a time when we in fact had far fewer human resources to apply to the project. There were many important questions to address: Should the Labyrinth include popular and non-scholarly resources, frequently suggested by users via our public web form? How should we justify excluding sites? How much time and effort could we apply to evolving and documenting our editorial process? Should we include commercial sites? As commercial interests and advertising became dominant on the web, even on scholarly sites, a link in the Labyrinth became a prized source of increased traffic and in some cases increased income for the site’s owner. This unintended outcome complicated our decisions about what to include in the consistently non-commercial Labyrinth. Despite numerous overtures, we staunchly resisted commercializing the Labyrinth. Consequently, the Labyrinth has never charged for access or resources, never collected member or subscription dues, never billed anyone for posting scholarly resources 33  Berners-Lee, “I am Tim Berners-Lee,” Reddit discussion, March 12, 2014. 34  Ibid.



Beginnings: The Labyrinth Medieval Studies Website

29

or even entire corpora on the site, never included advertising, and never had commercial sponsors. While some or all of these options might have made the Labyrinth more sustainable, finding an acceptable balance between commercial and scholarly interests was a task we found too difficult to undertake. At the same time, the need and demand for reliable access to vetted Medieval Studies resources continued to increase. When a raw web search for “King Arthur” returns millions of hits and prioritizes products and movies, it becomes worthless for scholarly research and misleading and even dangerous for students. As a contrasting example, the Labyrinth’s collection of Arthurian Studies resources continued for many years to play a crucial role, offering a reliable and trustworthy starting point for students, teachers, and writers, with teachers in particular pleading for us to keep the Labyrinth up-to-date and available. Simple Stability

As the demands of vetting resources and the increasing likelihood of link rot threatened to overwhelm us, we were caught between a rock and a hard place. Faced with this dilemma, we chose the rock, narrowing the Labyrinth down over a number of years to a simple but relatively stable site. The hard place, which would have been much more valuable, would have required developing a new governance structure for the Labyrinth. A stable and committed community effort could have restructured the Labyrinth, defined clear editorial processes, staffed the work of vetting and curation (potentially with volunteers), and revived the Labyrinth as a growing, thriving site. That didn’t happen, although in theory it still could. Our careers have evolved away from Medieval Studies, and although we still value this community, we are not the right people to initiate or lead this kind of complex, long-term community effort. A simple version of the Labyrinth remains available at the time this chapter is being written (2021), and many of its collected resources are still valuable. We transitioned to a WordPress web site in 2014 when the ColdFusion server was retired, the Labyrinth having outlived many web technology structures over the decades. The WordPress site largely follows the structure of its ColdFusion predecessor, but without a web database infrastructure. Although the WordPress site receives only minimal maintenance, it continues to provide web users access to these archived resources. It is, in other words, a slice of history.

An Enduring Tribute and a Foundation for the Future

While the WordPress site reflects prior versions of the Labyrinth site, we also knew we needed to capture and archive this rich history and related resources. The process of archiving the Labyrinth has been a daunting undertaking. With so many layers of a connected and complex history, it was hard to know where to begin. It was even harder to decide where to draw some sort of boundaries between what is included in the archive and what is not. The web doesn’t work that way, nor do our collaborations and rich professional relationships. Ultimately, then, the archive (and this chapter itself) have been bounded in ways that are counter to their nature. We sincerely apologize for the many important people, activities, and resources that are not included.

30

Deborah Everhart and Martin Irvine

Archived Materials

The logic of what is included begins with the Labyrinth files themselves, which on their own are massive and complex. Even making decisions about which files to archive was challenging. The Labyrinth web files that are written in basic HTML we have kept in their original format, as a set representing the most expansive Labyrinth HTML site. They can be viewed with any web browser or text viewer. The ColdFusion files, also in their original format, could be reinstated on a ColdFusion server, but they are not humanly readable. The WordPress files are still available and usable as a website at http://labyrinth.georgetown.edu. All of these Labyrinth web pages changed continuously, so the captured collections represent specific points in time: • The latest and most complete set of manually written HTML web files and images, from 2007, when the URL http://www.georgetown.edu/labyrinth was retired. • The latest and most complete set of ColdFusion files, from 2014 when the ColdFusion server was retired.

In addition to our own original copies of the Labyrinth files, we captured third-party documentation of the Labyrinth site from the Internet Archive at numerous sample points from 1997 (the earliest Internet Archive record of the Labyrinth) to 2021. This Internet Archive documentation is included in the archive as PDF files. Along with the Labyrinth files themselves, we compiled documentation explaining the history of the site and describing its significance. Here again, there was a continuous flow of activities, citations, and influences which we attempted to represent through three documents, which are included in the archive as PDF files: 1. This chapter itself, providing a narrative account of the Labyrinth. 2. An Archiving Dossier Narrative.

3. A Labyrinth History and Documentation (1992–2021) document that includes: • An abstract

• Curricula vitae for Labyrinth creators Martin Irvine and Deborah Everhart

• List of contributors (necessarily incomplete) • Timeline

• WorldCat catalogue references

• Publications and presentations related to the Labyrinth by the creators • Citations and references to the Labyrinth (necessarily incomplete)

• List of the artifacts archived, including the Labyrinth files and third-party documentation in the form of Internet Archive captures

Further detail about the archive can be found in the Labyrinth History and Documentation document. All of these materials are available from the Labyrinth website at Georgetown University (labyrinth.georgetown.edu), and are cataloged by Georgetown University Library with the permanent identifier http://hdl.handle.net/10822/1061738.



Conclusion

Beginnings: The Labyrinth Medieval Studies Website

31

At the time of this writing, we could only rest easy with this partial, imperfect compilation by reassuring ourselves that the Labyrinth website is still live, and that we can continue to add historical documentation there and also create additional archives, “had we but world enough and time.” We remain proud of the Labyrinth, its role in advancing Medieval Studies, the humanities, and scholarly collaboration on the web, and the contributions we and our colleagues have made to providing equitable opportunities for discovering and exploring the rich and complex world of our medieval heritage. We hope that this chapter, and the broader archive of our work, will provide a permanent record of one important episode in the history of the web, and lay the foundation for future evolutions in Medieval Studies.

Appendix

Letter from Deborah Everhart and Martin Irvine to Medieval Studies organi­ zations on the aims and importance of web development (1995)35 We would like to invite you to join us in this collaborative venture to extend the accessibility and intellectual content of medieval studies resources on the Web. Web networking technology was designed to be a fast and easy way for communities of researchers to share information and work collaboratively. From this origin, the Web has grown with extraordinary speed to include educational and government organizations, non-profit groups, commercial entities, and private individuals. We began building the Labyrinth, a World Wide Web disciplinary server for medieval studies, over a year ago, responding to a need to organize existing electronic resources and to plan new initiatives. Since the grand opening of the Labyrinth on May 2, 1994, the Labyrinth Web server has processed over 400,000 file requests from users in 56 countries. Currently, the Labyrinth server handles approximately 3,500 file requests per day. These statistics give evidence for wide interest in medieval studies information on the Web, and they also suggest the potential of greatly expanded outreach for medieval studies organizations linked through the Labyrinth. The growing importance of the Web in all areas of professional and personal life recommends that organizations in our field develop their Web presence as soon as possible. Medievalists have long been in the forefront of technological innovation, and Web development is not likely to be an exception to this pattern. For example, your organization’s home page on the Web could provide a centre for current and prospective members to find information quickly, share ideas, and coordinate projects. More importantly, your society could serve as a focal point for the development of electronic resources in your field, making a valuable contribution to the intellectual core of the Web. Everything from membership information to new scholarly projects and multimedia resources can be developed and shaped on the Web.

35  Everhart and Irvine, The Labyrinth, Georgetown University Library Repository, labyrinth/ professional/howto/letter.html.

32

Deborah Everhart and Martin Irvine

The Labyrinth can facilitate this development by providing a shared centre for online resources, and in turn, professional societies can help us make the Labyrinth a worldwide web for medieval studies, according to its original design. Now that more and more of our colleagues are using the Net and Web as part of everyday academic life, we expect an explosion of Web development in medieval studies. The existing organizational structures and the collective expertise of professional societies in our field will provide a strong foundation for promoting these developments, and the Labyrinth can continue to provide a central “meeting place” and structure that unites the various fields of medieval studies on the Web. HTML files are simply ASCII files with markup in angle brackets, and anyone with word processing experience can create Web files. Once you have created Web files, they can reside either on our server or on a server chosen by your organization. These files can act as an opening structure for an infinitely expandable body of materials organized, edited, and maintained by your organization and its membership. This collaborative model will give you the independence to shape Web resources according to your own needs and the intellectual life of your community, and it will also give your resources enormous visibility through the centralized access of the Labyrinth. Through collaborative development, the global medieval studies community can contribute to and continue to benefit from the rich diversity of the Labyrinth.

Bibliography Anon. [DM]. “Out of the archives, onto the Internet.” Stanford University News Service, August 16, 1995. https://news.stanford.edu/pr/95/950816Arc5114.html. Bass, Randy. Electronic Archives for Teaching the American Literatures. https://faculty.georgetown.edu/bassr/tamlit/. Berners-Lee, Tim. “I am Tim Berners-Lee. I invented the WWW 25 years ago and I am concerned and excited about its future.” Reddit discussion, March 12, 2014. https://www. reddit.com/r/IAmA/comments/2091d4/i_am_tim_bernerslee_i_invented_the_www_25_ years/. Connolly, Dan. “A Little History of the World Wide Web.” W3C. https://www.w3.org/History. html. DeLoughry, Thomas J. “Georgetown U. Offers Gateway to Resources for Medievalists.” Chronicle of Higher Education, May 11, 1994. https://www.chronicle.com/article/GeorgetownU-Offers-Gateway/92073. Everhart, Deborah, and Martin Irvine. The Labyrinth. https://repository.library.georgetown. edu/handle/10822/1061738. —— . “The Labyrinth: A World Wide Web Disciplinary Server, Project Number 04 – 1994: Coalition for Networked Information 1994 Workshop Final Report.” Coalition for Networked Information. https://www.cni.org/resources/historical-resources/new-learning-communities/workshops-1994-1995/1994-workshop-final-report/project-number-04-1994. Irvine, Martin. Web Works: The Norton Guide to the World Wide Web. New York: Norton, 1996. Thomason, Robert. “On the Net with Ancient Heroes.” Washington Post, November 2, 1995.

Chapter 2

NEW APPROACHES TO OLD QUESTIONS: DIGITAL TECHNOLOGY, SIGILLOGRAPHY, AND digisig JOHN MCEWAN

In England in the later Middle Ages, people from a wide range of social and vocational environments authenticated and validated their documents with seals. “By the early thirteenth century,” Harvey and McGuinness argued, the seal-using group “comprised everyone who had free land or property to convey or other business to be agreed in formal writing.”1 The seal impressions most people appended to documents are typically small objects—a few centimetres in diameter—and single sided.2 Grand seals, such as those used by kings, could be larger and include text and images on two sides.3 Mechanically speaking, seals were, and remain, a simple technology; to create seal impressions, a type of stamp, called a seal matrix, was used.4 The matrix was pressed into a plastic material such as wax, and the impression could then be attached to the object to be secured. Because seals were difficult to copy, medieval document-makers relied upon them for some protection against forgery; the best way to copy a seal was to create a new matrix, a task that required considerable skill.5 Despite the simple technology behind these tools, the three-dimensional nature and complex combination of text and image make seal impressions difficult to record or reproduce in other media, and therefore challenging to study. Millions of seals survive from medieval Europe, but because they are numerous and by nature hard to record, they present scholars with particular challenges. The digisig project, begun in 2013, initially aimed to address a long-standing challenge for sigillographers. Hundreds of repositories in England hold medieval seals, but in 2013 there was no single resource that could tell a researcher where any particular seal was located.6 The first version of digisig addressed this problem. But once the prototype system was in place, it proved to have further unanticipated capacities. digisig enabled researchers 1  Harvey and McGuinness, A Guide to British Medieval Seals, 77; see also McEwan, “Formation of a Sealing Society.” 2  McEwan, “Does Size Matter?,” 108.

3  Harvey and McGuinness, A Guide to British Medieval Seals, 27–42; Steane, The Archaeology of the Medieval English Monarchy, 22–30. 4  New, Seals and Sealing Practices, 8–13.

5  Heslop, “Twelfth-Century Forgeries as Evidence for Earlier Seals.” On the manufacture of seals, see McEwan, “Making a Mark in Medieval London.” 6  McEwan, “The Past, Present and Future of Sigillography.”

John A. McEwan, Walter J. Ong Center for Digital Humanities, St. Louis University.

34

John McEwan

not only to discover seals, but to analyze, compare and combine information from multiple reference works.7 These capacities may transform sigillography by enabling scholars to consider long-standing problems in the history of sealing practices, based on a fuller understanding of the evidence. The development of systems such as digisig underlines the complex interactions between search and discovery systems and the scope of historical knowledge. In the late twentieth century, the emergence of digital technology tantalized sigillographers with the prospect of improving their reference works, but it also gave them new approaches to established research questions. Sigillographers long have struggled to record the appearance of seal impressions. In the early modern era, antiquarians such as Nicholas Charles (d. 1613) and William Dugdale (1605–1686) sketched seals they studied, while other researchers provided even more information.8 Christopher Hatton’s Book of Seals, for example, is an elaborate reference work from this same period which offers colour illustrations of seals and the documents to which they were attached.9 This work remains important to this day in part because some of the seal impressions depicted in the volume have since been damaged or lost, so that the elaborate illustrations provide the best evidence of for these seal impressions. However, illustrations alone could not fully capture the three-dimensional qualities of seal impressions or seal matrices. To address this problem, nineteenth-century seal scholars and enthusiasts made casts of seal impressions and then fabricated moulds to make copies availible to reference collections. Conservator and businessman John Doubleday, who was associated with the British Museum, for example, made casts of seals in the Museum and for the Museum, but also sold them to other institutions and to the public.10 The process of casting that Doubleday used, however, could damage seal impressions. In 1922, Hilary Jenkinson described one method used at the Public Record Office (at The National Archives; hereinafter TNA) to take a cast of a seal that required the person making the cast to “oil the seal with olive oil and take a mould by means of liquid plaster applied with a brush.”11 Such practices may have been effective, but could also damage the fragile wax impressions. Photography, on the other hand, offered a noninvasive way to record the appearance of seals, and by the end of the nineteenth century, seal catalogues regularly included photographs.12 By the end of the twentieth century, 7  For lists of sigillographic reference works, see Harvey and McGuinness, A Guide to British Medieval Seals, 120–21, and New, Seals and Sealing Practices, 129–30. 8  McEwan, “Tout and Seals.”

9  Sir Christopher Hatton’s Book of Seals, ed. Loyd and Stenton; Harvey and McGuinness, A Guide to British Medieval Seals, 23.

10  The anonymous “Miscellaneous Communications from an American Naval Officer,” 74–78. For information on some of the major casting projects undertaken in French and Belgian archives, see Coulon, Le Service Sigillographique et les collections d’empreintes de sceaux des Archives Nationales, 24, and Libert, “The Seal Casts Collection,” 256.

11  Jenkinson, A Manual of Archive Administration, 64. See also Villela-Petit, “Les techniques de moulage des sceaux,” 515. For a description of a modern method of casting seals, see Melvyn, “Seal Repair, Moulds and Casts.”

12  Birch, Catalogue of Seals; Greenwell and Blair, Catalogue of Seals in the Treasury of the Dean and Chapter of Durham.

New Approaches to Old Questions

35

digital cameras revolutionized the photography of seals, since the new technology made it affordable to photograph large numbers of seals and share the images online.13 Two other methodologies, reflectance transformation imaging and photogrammetry (based on photographic technologies, to create dynamic digital visualizations with threedimensional qualities) offered new ways for cataloguers to convey information about the appearance of seals to the public.14 The seals used in Medieval Europe to protect against the forging of documents have now, thanks to digital technology, become works that scholars can record precisely and study remotely and online. Although the technology now exists to record faithfully the appearance of seal impressions without damaging the originals, scholars still need tools to locate the impressions that interest them. The number of surviving seal impressions is staggering. In England, Clanchy suggested that once small landowners began recording exchanges in writing, by 1300, “it seems reasonable to conclude that at least hundreds of thousands, and perhaps millions, of peasants’ charters were made.”15 Only a fraction of these documents survive, but when they are added to those that survive from major landowners, ecclesiastical administrations and the royal government, the number of records becomes imposing. Many of these records were validated or authenticated with seals and a good portion still preserve their seal impressions. Harvey, for example, estimated that TNA could hold some two hundred and fifty thousand seal impressions, of which perhaps fifty thousand are medieval.16 TNA has one of the largest collections in England, but many other repositories also hold important collections. The sheer number of examples creates enormous challenges for scholars attempting to search among them, and this challenge is further heightened by the nature of seals and the conditions in which they survive. Although medievalists rarely consider print technology in much of their work, seal impressions are in fact a form of printed material, for a person or corporate entity could use the same seal matrix to place impressions on multiple documents. In the course of directing the affairs of England, for example, King Henry III’s administration produced large numbers of documents that were validated with the king’s great seal, and today, the London Metropolitan Archives alone holds five examples of Henry III’s first great seal.17 Furthermore, the National Library of Wales has one in the Penrice and Margam Estate Records,18 Westminster Abbey has at least two,19 there are four in the National Archives,20 and Durham Cathedral has five in the Regalia series.21 The British Library 13  New, “Digital Imaging of British Medieval Seals.”

14  McEwan, “Reflectance Transformation Imaging.” 15  Clanchy, From Memory to Written Record, 50.

16  Harvey, “Computer Catalogue of Seals in the Public Record Office,” 29.

17  McEwan, Seals in Medieval London, no. 9. COL/CH/01/011, COL/CH/01/012, COL/CH/01/013, COL/CH/01/017/A, COL/CH/01/017/B. 18  Aberystwyth, National Library of Wales, PENRICE/134.

19  London, Westminster Abbey Muniments, Deeds 6318 and 17356.

20  London, National Archives, DL 10/87, E 329/464, E 42/524, E 42/529.

21  Greenwell and Blair, Catalogue of Seals in the Treasury of the Dean and Chapter of Durham, no. 3025.

36

John McEwan

reports that it holds eighteen, including modern casts (which are themselves copies of medieval seal impressions), located in such series as Cotton Charters, Harley Charters, Additional Charters, Topham Charters, Wolley Charters, and Doubleday Casts.22 Scholars seeking to understand Henry III’s seal and its history thus would need to examine all the impressions of a seal, because each one can add to the seal’s composite story. Seal impressions are fragile objects; only a fraction survive in their original condition, most impressions are damaged, and many are missing sections or are partly illegible. So, if a single complete legible seal impression is not available, researchers might still be able to reconstruct the appearance of the seal based on an examination of multiple impressions. The combination of the vast number of surviving seal impressions, the existence of multiple copies of some seals, and the damaged state of most seal impressions makes it harder for scholars to determine precisely how many seals survive and to establish their contents. Because scholars need tools both to visualize seals and to discover and search for them, the adoption of digital tools to record the appearance of seals has occurred in concert with digital systems to manage information about them. Scholars rely on the the catalogues and finding aids maintained by repositories to discover and search for seals. The catalogues and finding aids are the foundation for sigillographic knowledge, and therefore must be reckoned with. Many generations of archivists have worked to record the seals in archival collections, but the work is still incomplete, which is a serious problem for scholars. Ellis pointed out that sigillography “cannot be fully or systematically studied until our records repositories have compiled, and made accessible, catalogues of all the seals”.23 Despite the importance of cataloguing, the process in many countries, including England, has proceeded slowly. However, the process was disrupted in the late twentieth century by scholar’s increased access to and reliance upon computer-based technologies. From an early date in the history of computers, scholars recognized their potential to change the form of sigillographic reference works. In 1975, archaeologist Françoise Digard published a work describing the electronic information system that he and his collaborators had built for cylinder seals. At that time, computers were bulky and expensive tools, so few scholars had access to them, but Digard explained that computers enabled methods of interacting with sigillographic information that differed from those provided by a printed reference work. An electronic reference work did not have to be read in a linear fashion, like a text, or searched through the handful of indexes that could be included in a printed work. Instead, the database enabled scholars to perform searches that involved any combination of terms, “and not only a selection of terms, or more generally on isolated terms, like in the case of a printed index.”24 Moreover, Digard and his collaborators were conscious that they were creating a system that 22  Birch, Catalogue of Seals, nos. 100–17.

23  Ellis, Personal Seals, 1:vii.

24  “Et non pas seulement sur quelques-unes d’entre elles, ou plus generalement sur des termes isolés, come dans le cas des index imprimés.” Digard, Abellard, and Bourelly, Répertoire analytique des cylindres orientaux, 1:7.

New Approaches to Old Questions

37

could be incrementally improved. The contents of a printed publication were fixed, but the electronic system that Digard prototyped could be continually “enriched.” From an early date, scholars recognized that they could use electronic systems to create reference works that performed some of the key functions of their analogue counterparts, yet more efficiently and effectively. Scholars also recognized that computers made it possible to create reference works with features that were unparalleled in their analogue counterparts. Until the late-twentieth century, reference works were designed with the limitations of human beings in mind: both those of the cataloguers who wrote the reference works, and the scholars who consulted them. However, computers could assist cataloguers in the creation of elaborate indexes and complex classification systems. In 1993, Martine Dalas-Garigues published a description of an elaborate index devised for a digital catalogue of the seal cast collection in the Archives Nationales.25 The collection included about one hundred thousand casts, so it was too large for anyone to search by hand. The cataloguers relied on computers to help them create an index but they also foresaw that users would rely on computers to help them search the index. Scholars also investigated the development of new types of classification systems. In 1981, Michel Pastoureau wondered whether computers would enable “des principes de classement plus subtiles et plus satisfaisants.”26 One such classification system was developed by the Seals in Medieval Wales (simew) team, using a hierarchical system, with multiple levels, to help retrieve seals with particular motifs.27 A human can struggle to maintain and manipulate a multilevel hierarchical classification system, but it is a trivial task for a computer. Once cataloguers could assume that computers would be available, they could devise new sigillographic reference work formats. By the 1990s, as computers were becoming more commonplace, fully formed electronic information management systems for seals began to appear. In England, seal expert Paul Harvey and technology expert T. M. Chalmers worked together to design a system that would prove to be a turning point in the field. Working in collaboration with TNA, they used Informix Corporation’s Smartware II, an early office suite for desktop computers that included database software. Harvey took a holistic approach to designing the sigillographic aspects of the system, using computers to facilitate the creation of sigillographic information as well as to present it to the public. Harvey wanted the computer to help ensure the information in the system was entered with “rigorous consistency,” but would also enable scholars to search the data by document, date, name, design, legend, place, size, or shape.28 The result was an electronic information management system with tools to assist cataloguers and scholars.29 In 1996 Harvey published an article that documented the system and outlined its capabilities for archivists, but in practice, it proved difficult to implement. Smartware II 25  Dalas-Garrigues, “L’inventaire informatique de sceaux.” 26  Pastoureau, Les Sceaux, 61.

27  McEwan, “The Challenge of the Visual.”

28  Harvey, “Computer Catalogue of Seals in the Public Record Office,” 31. 29  Chalmers, Reference Manual.

38

John McEwan

was a standard piece of software at TNA when Harvey began work on his project, but it was subsequently superseded by Microsoft Windows.30 This did not render Harvey’s work unusable, but TNA itself no longer supported the Smartware format. By 1996, the project had been transferred into a Microsoft format (MSAccess), but TNA does not seem to have made it publically accessible.31 Instead, the future that TNA envisaged was online. In 1999, TNA reported that they had just completed the conversion of their paper catalogues into an electronic form and they aimed to have a full electronic catalogue accessible via the internet in 2001.32 Although Harvey’s seal catalogue was created in the digital medium, it was left behind when TNA made the leap to offering access to its reference materials online. Instead, TNA made a printed work based on Harvey’s digital catalogue, entitled Duchy of Lancaster: Deeds, Series L, DL25, Catalogue of Seals, that was made available in the archive reading room. Stripped of its electronic interface, the printed version could be searched in a limited number of ways. As Harvey’s case demonstrates, in the late twentieth century, scholars surged ahead in building electronic information systems for seals, but encountered problems with providing public access to those same systems when respositories shifted their focus to providing online services. The digisig Sigillography Resource (digisig) project, begun in 2013, built upon both Harvey’s work at TNA and the Seals in Medieval Wales (simew) project;33 in fact, the conclusion of simew was the immediate prompt for digisig. During the research phase of simew, team members surveyed roughly two and a half thousand medieval seals drawn from multiple collections in several repositories, and compiled a database.34 While simew was in progress, the data was stored in a database running SQL Server 2008, and when it was completd in 2012, simew published its findings and ran outreach events to bring those findings to the public. Making the dataset itself publically accessible in an electronic format via the internet, however, was not foreseen in the original planning phases and therefore outside the project’s scope.35 Only a few years earlier, after about a decade during which time TNA had made Harvey’s catalogue publically accessible only in a printed format, the institutions put it online, thereby creating a digital resource that enabled many of the query types Harvey had originally envisaged.36 TNA’s experience with Harvey’s electronic catalogue therefore underlined that making a sigillographic 30  Harvey, “Computer Catalogue of Seals in the Public Record Office,” 36.

31  Chalmers, Reference Manual, 1.

32  Tyacke et al., The 40th Annual Report of the Keeper of Public Records, 2–3. For a view of the website in its earliest days, see: https://webarchive.nationalarchives.gov.uk/19981203123849/ http://www.pro.gov.uk:80/ad2001/procat.htm. 33  McEwan, “The Past, Present and Future of Sigillography.”

34  Schofield et al., “Seals in Medieval Wales and the Welsh Marches”; McEwan, “Seals in Medieval Wales,” 13.

35  For a printed list of the seals, see the appendix to Schofield et al., Seals and Society. For a catalogue of the public exhibition related to the project, see McEwan and New, Seals in Context.

36  For a view of the query entry form, see http://web.archive.org/web/20110202214520/ h t t p : / / w w w. n a t i o n a l a r c h i v e s . g o v. u k / d o c u m e n t s o n l i n e / b r o w s e - r e f i n e . asp?CatID=47&searchType=browserefine&pagenumber=1&query=*&queryType=1. Harvey’s seal

New Approaches to Old Questions

39

reference work available online was indeed possible. Although simew might have emulated the online search system that TNA had created, it did not have the support of a repository to finance or sustain the project; simew was a university-led research project that had drawn together information about seal impressions located in many different repositories, meaning there was no single repository to serve as the simew dataset’s natural home. The predicament of the simew dataset raised the question of how digital scholarly and research sigillographic datasets could be made publically accessible over the long term without the enduring support of a repository. An online resource dedicated to only simew would struggle to endure, but I wondered if a resource that included simew and additional sigillographic reference works, might have more success. In 2013, sigillographers who worked on English seals still needed a tool to determine which repositories held copies of particular seals. TNA and the British Library had large seal collections, but so did many county record offices, libraries, cathedrals, and even hospitals. Some of these repositories’ finding aids were accessible exclusively. For example, Westminster Abbey’s extraordinarily rich card catalogue, which enabled scholars to search for several different types of seals, was physically located only in its search room. Although some repositories had published their reference works and others had made resources available online, researchers still had to consult each work in turn. The archival community recognized these inefficiencies, and about a decade earlier, the National Archives of Scotland, the British Museum, the British Library, and the National Library of Wales and TNA discussed the practicality of creating a union catalogue for seals in the United Kingdom. Sadly, that project did not come to fruition.37 Thus, the financial difficulty of paying for a web resource and scholars’ need for a means to facilitate searches across multiple repositories were arguments for creating a tool for searching multiple sigillographic reference works. Moreover, technology developed for simew could be repurposed to serve as a foundation for the project, for the simew project had built a system to manage information about seals in many repositories. If I transformed the simew dataset into an online resource, added more sigillographic reference works, and enabled people to search it for seals in various repositories, then perhaps digisig would develop an audience that could sustain it, at least until the arrival of the archival of the community’s long-envisaged union catalogue of seals. In 2013, there were few models for systems that facilitated searching multiple sigillographic reference works, so this facet of the project broke new ground. One of the preliminary challenges was gathering a representative sample of sigillographic reference works. Harvey’s catalogue and the Portable Antiquities Scheme dataset existed in a digital format, but many others did not, so at the beginning of the digisig project, I digitized samples of printed reference works from the British Library, British Museum, and TNA.38 descriptions were subsequently adapted into records for “Discovery,” the current main catalogue, with loss of data. 37  New, “Digital Imaging,” 11.

38  Birch, Catalogue of Seals; Ellis, Monastic Seals; Ellis, Personal Seals; Tonnochy, Catalogue of British Seal Dies in the British Museum. Each of these works was only partially digitized and

40

John McEwan

The books were scanned and a digital image of each entry created. As a result, when researchers wanted to see entries, digisig could display them as they appeared in the original reference work. However, to enable researchers to perfom searches on the texts of those entries, they had to be turned into a machine readable format, so I passed the scans through text recognition software. When the sample of sigillographic reference works was digitized and compared, some overlap in their contents appeared. For example, when Ellis created his catalogue of seals in TNA, he found that in some cases he was recording seals of which Birch had earlier recorded other impressions (some of which were casts of impressions held in TNA) in the British Library. Then Harvey, in turn, catalogued some seal impressions that had already been recorded by Ellis.39 Cases where multiple catalogues described the same seals are exceptional, for most seals are represented in the historical record by a single impression or matrix. However, since seals are a form of printed material, impressions of any particular seal can turn up in many places. In the design of digisig, one of my most important considerations was how to present to researchers cases where there were multiple descriptions of the same seals. Since digisig’s primary aim was to help researchers locate seals in repositories, resolving discrepancies among the descriptions of particular seals found in the various sigillographic reference works was an additional problem outside the scope of the project. By contrast, as digisig was in development, another large-scale sigillographic project was underway in France that in fact did address the challenge of creating authoritative seal descriptions for seals. The sigilla project, which focused on seals in France, aimed to provide researchers with access to large numbers of seals.40 The sigilla project used its considerable human resources to prepare authoritative seal descriptions. With the limited time I had to devote to the project, I could not revise seal descriptions, but I could identify duplicate seal descriptions. They can be hard to detect because they arise from a variety of circumstances, but these determinations are based on the evidence in the reference works themselves. Cataloguers can and often do indicate that a seal description is a revision of another description. As already mentioned, Ellis’s catalogues of seals in TNA identified cases where he was describing a seal that had previously been described by Birch.41 Some seals, such as those of kings, are so well known that they have conventional names and so can be easily recognized in different reference works.42 Still other duplications can occur when different cataloguers describe the same seal impressions. A good example is offered by simew, which recorded some material from the British Library. The duplicate seal descriptions were discovered when Birch’s cataremains a work in progress. Further works were subsequently added, such as Greenwell and Blair, Catalogue of Seals in the Treasury of the Dean and Chapter of Durham and McEwan, Seals in Medieval London. The system also includes unpublished work of the author, including surveys of seals in the collections of the Huntington Library, California.

39  For instance, compare Birch, Seals, vol. 1, no. 3981; Ellis, Monastic Seals, M761: https:// discovery.nationalarchives.gov.uk/details/r/C16099054. 40  The project maintains a website at: www.sigilla.org.

41  For example, see Ellis, Personal Seals, 1:50, no. P608. 42  Wyon and Wyon, The Great Seals of England.

New Approaches to Old Questions

41

logue was digitized and the document reference numbers of the seals were compared in Birch and in simew.43 In all three cases, the identification of duplicates is performed by myself, based on information supplied by the reference works themselves. Then the digisig website can notify researchers of cases in which multiple descriptions are available, but scholars are left to evaluate these multiple descriptions themselves. Asking researchers to read multiple descriptions of the same seal might be considered a burden by some researchers, especially those with little experience with seals, but there are legitimate reasons why multiple seal descriptions exist. When cataloguers record seals, as already mentioned, they gather together the evidence for a particular seal and compose a description. One cataloguer might have access to different evidence for a seal than another cataloguer. For instance, one might have the left side of the seal, and another the right side. A further complication is that when cataloguers record medieval seals, they are engaged in an act of interpretation. A seal presents graphical and textual information, and there can be multiple valid descriptions of that information. A cataloguer can choose to prepare a very detailed description that includes information about both the main and secondary elements (such as ornamentation in the background), or a cataloguer can focus more narrowly on the main element.44 Furthermore, scholars in the history of art and iconography have not yet reached a consensus on the significance of every motif that can appear on a medieval seal.45 Consequently, a cataloguer cannot always provide a researcher with the meaning of an image in its original historical context. When the contemporary meaning of an image is debatable, cataloguers can offer a literal description. Then, when a scholarly consensus has been reached on the meaning of an image in its historical context, a cataloguer might return to the seal and prepare a new description. When a sigillographic information system offers access to multiple sigillographic reference works, those works can contain contrasting descriptions whose differences are the result of cataloguing conventions, variable reconstructions of fragmentary seals, disagreement in interpretation, or a combination of those factors. In these situations, acknowledging differences in interpretation of the evidence is the correct course of action for a system such as digisig, whose main aim is to enable researchers to discover and locate seals. Composing new seal descriptions takes human resources, but it also takes time, for each case has to be considered individually. digisig needed a substantial dataset of seals at its launch to demonstrate that its system could help researchers locate seals. When it was made publicly accessible in 2015, digisig provided access to descriptions of 27,817 seals. This large number was possible because the seal descriptions in the various sigillographic reference works incorporated into the dataset were taken verbatim and not revised. Following its launch, digisig then continued to grow swiftly. At the time of writing in 2021, the dataset included 42,637 seals, which are represented in 58,799 seal

43  Birch, Catalogue of Seals, vol. 2, nos. 5944–45; Schofield et al., “Seals in Medieval Wales and the Welsh Marches,” 135, no. 55. 44  McEwan, “The Challenge of the Visible,” 1004.

45  McEwan, Seals in Medieval London, 181.

42

John McEwan

impressions, matrices or casts, and are described in 44,977 distinct seal descriptions.46 If digisig aimed to provide a single authoritative entry for each seal, the authoritative descriptions would need to be drafted on entry into the system, and then revised and updated as new information came to light or scholarly opinion shifted, which would severely inhibit growth. Instead, digisig can accommodate new reference works as they become available, without revisions to other descriptions in the system. Over time, digisig will become a progressively richer resource. As digisig does not try to settle discrepancies between the works, but exposes them to scrutiny, it will become an excellent source of information on the history of sigillography. Generations of sigillographers have contributed to the documenting of the sigillographic legacy of the Middle Ages, and each contributor has built on the work of their predecessors, but often with a new emphasis and fresh insights. For example, in outlining the importance of his catalogue, Harvey argued that it represented both an important continuation of Ellis’ work, and also an attempt to bring new attention to the seals of people of lesser standing, which Ellis overlooked.47 digisig, by design, requires some sigillographic knowledge to use most effectively, but the system can grow in response to new seal cataloguing work, and it makes accessible to scholars information on the history of the cataloguing of particular seals. In the autumn of 2015, the prototype of digisig was launched online.48 I built the front end of the website using PHP, HTML, CSS and Javascript, and the backend was a MySQL database. In addition to offering researchers seal descriptions, the website also provided photographs of seal impressions and seal matrices.49 To query the dataset, digisig asked users to submit a string of characters that digisig would then search for in one of six fields.50 Four of those fields were from the various sigillographic reference works that digisig had indexed: the description of the visual content of the seal, the name of the person or corporate entity that the cataloguer associated with the seal, the place the cataloguer assigned to the seal, or the identifier for the seal in a particular reference work. digisig itself assigned each seal a unique, searchable record identifier. Finally, researchers could search for a document reference. Sigillographic reference works normally include the document references, but those references can change. For example, as already mentioned, Birch catalogued seals in the British Museum, but later in the twentieth century, those collections were moved to the British Library. To ensure that users have the information they need to locate the seals on the records in question, 46  For comparison, at the time of writing, sigilla reported that it recorded 21,140 seal impressions: http://www.sigilla.org/ 47  Harvey, “Personal Seals,” 118.

48  For a view of the earliest version of the interface, see http://web.archive.org/ web/20151014063824/http://DIGISIG.org/. 49  Many of these images were taken from reference works, but in the case of the Portable Antiquities Scheme, which operated its own online resource, the photographs were called directly from their image repository and displayed on digisig.

50  For a more detailed account of how digisig worked and how it compared to Harvey’s project, see: McEwan, “The Past, Present and Future of Sigillography.”

New Approaches to Old Questions

43

I updated the references to the seals in Birch’s catalogue to the current British Library references. Thus, digisig enabled researchers to search for seals in a number of ways, but the range of search tools that digisig provided was still limited in comparison with what TNA provided for Harvey’s catalogue. For example, there was no capacity in digisig to search by date. However, digisig did demonstrate that it was possible to search multiple sigillographic reference works in concert. digisig’s search results contained links to several types of pages. If researchers selected a repository reference from their search results, for example, they were directed to a page with information about the item or document. If they selected a digisig number, they were passed to a page with information about that particular seal. If they selected a description of the visual content, a name, a place or the reference work identifier, they were passed to a page with information from a sigillographic reference work. All these pages contained links to the other types of pages, so a user could then follow those links to browse and explore the full range of information available in digisig. A researcher might begin by selecting a name from their search results, and discover from the seal description page that dozens of impressions of that seal survived.51 The researcher could follow the links associated with those seal impressions to the various item information pages, and then follow further links from the item page to the online catalogues of the repositories where those items were kept. Alternatively, a researcher might begin with a seal description, and then discover that alternative descriptions were available. The system enabled researchers who had a firm grasp of how seal catalogues worked to move swiftly from seal descriptions to item records and vice versa, and from seal description to seal description, but it was not an intuitive information-seeking process for users who were less familiar with seals. At the time of writing, digisig is being rebuilt. The new version uses the opensource Python-based web framework Django and a PostgrSQL database. The redesign aims to make navigation of the site easier for novices and to increase the number of search fields. While these are incremental improvements, the redesign will also offer a novel feature based on a discovery that was made after the digisig prototype was completed. digisig was designed to enable researchers to discover seals and to automate the process of searching multiple sigillographic reference works, but since many of those reference works were not available online in a machine readable format, the information from these works had to be ingested into the system. With all that information at hand, digisig can perform calculations that draw on the sum total of the information in its dataset. This capacity could assist scholars who want to study how people in particular places changed the seals they favoured over time—a topic that has long been of interest to sigillographers. In the early twentieth century, Walter de Gray Birch’s introduction to English seals included a chapter on what he terms “miscellaneous seals”: seals that he associates with people outside the aristocracy. He notes that the thirteenth century “was rich in the variety of the designs of seals” and that in the fourteenth century there was a rise in the “use of numerous devices, monograms, fanciful subjects, and figures of natu51  See the discussion above of the various impressions of the first great seal of King Henry III.

44

John McEwan

ral objects.”52 Subsequent scholars who have followed Birch’s lead have extended and deepened our understanding of the history of English seal motifs; Heslop has remarked that the “quantity” of examples of seals from the thirteenth and fourteenth centuries “allows shifts in fashion to be seen.”53 However, there are limits to what they could achieve. To extend our understanding beyond, in Harvey’s phrase, “points that leap to the eye,”54 they need access to precise and structured information about the seals so they can perform calculations. digisig can help. The history of heraldic seals offers a good test case, because this type of seal interests sigillographers and the portion of people using them changed dramatically in the later Middle Ages. Heslop has observed that in the late thirteenth and early fourteenth centuries, “the most popular seals of the period were heraldic.”55 Yet exactly how ‘popular’ did heraldic seals become in the later Middle Ages? The analogue reference works available to scholars in the late twentieth century were of little help because of their inflexibility. The most important reference work for English heraldic seals is Birch’s catalogue, which describes thousands of heraldic seals.56 The catalogue includes the date of each seal impression, but as it is a printed work and does not happen to include a table listing seal impressions by date, researchers cannot manipulate it to sort seals by date. Harvey’s digital catalogue could support such a query, and in 1991 he calculated that nine percent of its thirteenth-century seals and 32 percent of its fourteenth-century seals were heraldic.57 Harvey was able to show precisely how much the portion of seals with heraldic motifs increased in his dataset, but he did not claim he had definitively established how ‘popular’ heraldic seals were in the fourteenth century. He used only the information from his own reference work, which was limited, and he commented that his results “need testing against other substantial collections.”58 At that time, no system existed to gather information from multiple reference works. Consequently, Harvey envisaged scholars running comparative tests. However, digisig can bring together information from many reference works and perform calculations on the sum total of all the available information, which will bring scholars closer to definitive answers. In practice, there are some technical challenges to overcome before digisig can create a dataset that scholars can use to study changes over time in the sealing practices of the people of England. digisig can draw information from many sigillographic reference works, but sigillographic reference works do not always supply date information.59 Seal impressions normally survive as features of documents, and information in 52  Birch, Seals, 190.

53  Heslop, “Peasant Seals,” 214.

54  Harvey, “Personal Seals,” 119. 55  Heslop, “Peasant Seals,” 214. 56  Birch, Catalogue of Seals.

57  Harvey, “Personal Seals,” 120.

58  Harvey, “Personal Seals,” 119–20.

59  On the complex history of the relationship between seal cataloguing and the cataloguing of documents, see McEwan, “Tout and Seals,” 189–92.

New Approaches to Old Questions

45

those documents can help to establish the period when the seals were used and where they circulated.60 Repositories will normally maintain a separate catalogue entry for the document, which will report the document’s date. When seal cataloguers create sigillographic reference works, they have to decide how much information from the document catalogue entry to duplicate in the seal catalogue entry. Researchers find it convenient to have temporal information in seal descriptions, but this increases their length and complexity—which can be an issue in printed catalogues, where the number of pages increases the cost of producing the work. Ellis, in one of his catalogues for TNA, was inconsistent in his approach to assigning dates to seals.61 Birch, in his catalogue, routinely supplies the dates of the documents to which seals are appended. Greenwell and Blair sometimes provide dates.62 Consequently, digisig may provide to researchers the seal descriptions verbatim, but to search or sort seals by date, digisig has to first compare the date information available in the sigillographic reference work to the information available in the catalogue entry for the document. As digisig has links to the online catalogues of the repositories, that information can be collected from the online records for the documents. As a result, in the revised digisig researchers should be able to search by date even if the original sigillographic reference work does not supply this information. To show how this would work in practice, we can revisit Harvey’s observation, based on his sample of two thousand five hundred seals, that about 9 percent of the thirteenthcentury seals and 32 percent of the fourteenth-century seals are heraldic.63 As digisig can draw information from many reference works, it has the capacity to construct a set of seals that is tailored to the requirements of the research question. If several additional datasets are added to the information available to Harvey, then the total number of seals in the analysis can be raised to about 5580.64 In the mid-thirteenth century, the percentage of seals with the shield of arms was 5.9 percent. At the turn of the fourteenth century, this rose to 13.6 percent, more than doubling in half a century. Then the figure leapt upwards to 35.2 percent in the mid-fourteenth century, when it reached its peak. Then, at the beginning of the fifteenth century, the portion of seals with the shield of arms motif began to decline, dropping to about a quarter of the total (25.6 percent). Harvey suggested that heraldic seals formed about nine percent of the total in the thirteenth century, but dividing the data into five periods, in contrast to Harvey’s two, shows that 60  The dating of seals from the twelfth century or earlier can require a different approach: Heslop, “English Seals from the Mid-9th Century to 1100.”

61  Ellis, Monastic Seals. For an analysis of Ellis’ approach to dating, see the review by Bedos Rezak in Speculum 64. 62  Greenwell and Blair, Catalogue of Seals in the Treasury of the Dean and Chapter of Durham.

63  Harvey, “Personal Seals,” 118, 120.

64  For the purposes of this demonstration, seals from four reference works were added to the seals from Harvey’s catalogue, then limited to seals associated with documents dated prior to 1426 where the motif is known; McEwan, “Seals in Sussex”; McEwan, Seals in Medieval London; Greenwell and Blair, Catalogue of Seals in the Treasury of the Dean and Chapter of Durham; Schofield et al., Seals and Society.

46

John McEwan

there was considerable dynamism in the use of heraldic seals: a swift rise, followed by further growth in the use of heraldic seals over the course of the thirteenth century, and then a decline at the end of the fourteenth. The results not only confirm Harvey’s findings in general, but add considerable further detail. digisig thus will enable scholars to reconsider the history of sealing in England with a firmer understanding of the evidence. In the early twentieth century, at The National Archives (Public Record Office), Hilary Jenkinson argued that “no catalogue would be satisfactory which was not com­ prehensive.”65 Jenkinson had in mind the role of the catalogue as a point of access for the seals in a repository, but in the digital age, it might be added that to be satisfactory the catalogue should also be searchable online and in concert with catalogues from other repositories. The digisig project initially aimed to allow scholars to search multiple sigillographic reference works online so that they could more efficiently identify the repositories that held seals relevant to their research. Yet it is digisig’s ability to manipulate the information in those reference works and to create new datasets from their contents that is potentially transformative for sigillographic studies. digisig can combine information from different reference works to enable scholars to address questions that require statistically significant numbers of cases. Moreover, with systems such as digisig, scholars can begin to investigate the ideas, values, and beliefs not just of individuals or small groups of people, but of larger-scale communities, from cities to regions and kingdoms. Systems such as digisig that can manipulate the organization of the information already contained in sigillographic reference works expand the scope of historical knowledge. What scholars can know about the past depends not only on the sources that survive, but also on the systems that exist to search, organize, and analyze the information in those sources. Scholars should be actively involved in the development of systems such as digisig, because they extend and define the frontiers of scholarship.

Bibliography

Anon. “Miscellaneous Communications from an American Naval Officer, Travelling in Europe; Forwarded from the Mediterranean, May, 1834.” American Journal of Science and Arts 27, no. 1 (1835): 74–83. Bedos-Rezak, Brigitte. Review of Roger H. Ellis, Catalogue of Seals in the Public Record Office: Monastic Seals and René Laurent, Sigillographie. Speculum 64, no. 1 (1989): 158–62. Birch, Walter de Gray. Catalogue of Seals in the Department of Manuscripts in the British Museum. 6 vols. London: Longman, 1887–1900. —— . Seals. London: Methuen, 1907. Chalmers, T. M. PRO Catalogue of Seals Database: Reference Manual. Unpublished manuscript, 1997. Clanchy, Michael T. From Memory to Written Record, England 1066–1307. 2nd ed. Oxford: Blackwell, 1993. Coulon, Auguste. Le Service Sigillographique et les collections d’empreintes de sceaux des Archives Nationales: Notice suivie d’un catalogue du Musée Sigillographique. Paris: Champion, 1916. 65  Jenkinson, A Guide to Seals in the Public Record Office, 2nd ed., x.

New Approaches to Old Questions

47

Dalas-Garrigues, Martine. “L’inventaire informatique de sceaux: Un essai aux Archives Nationales à Paris.” Janus: Revue Archivistique 1 (1993): 69–81. Digard, Françoise, C. Abellard, and L. Bourelly. Répertoire analytique des cylindres orientaux: Publiés dans des sources bibliographiques éparses (sur ordinateur). 3 vols. Paris: CNRS, 1975. Ellis, Roger H. Catalogue of Seals in the Public Record Office: Monastic Seals. London: HMSO, 1986. —— . Catalogue of Seals in the Public Record Office: Personal Seals. 2 vols. London: HMSO, 1978–81. Greenwell, W., and Charles Henry Hunter Blair. Catalogue of Seals in the Treasury of the Dean and Chapter of Durham. Archaeologia Aeliana, 3rd ser., 7–16. Newcastle upon Tyne: Society of Antiquaries, 1911–1921. Harvey, Paul Dean Adshead. “Computer Catalogue of Seals in the Public Record Office, London.” Janus: Revue Archivistique 2 (1996): 29–36. —— . “Personal Seals in Thirteenth-Century England.” In Church and Chronicle in the Middle Ages: Essays Presented to John Taylor, edited by G. A. Loud and Ian N. Wood, 117–27. London: Hambledon, 1991. Harvey, Paul Dean Adshead, and Andrew F. McGuinness. A Guide to British Medieval Seals. London: British Library and Public Record Office, 1996. Hatton, Christopher. Sir Christopher Hatton’s Book of Seals. To Which Is Appended a Select List of the Works of Frank Merry Stenton. Edited by Lewis Christopher Loyd and Doris Mary Stenton. Oxford: Clarendon, 1950. Heslop, T. A. “English Seals from the Mid-9th Century to 1100.” Journal of the British Archaeological Association 133 (1980): 1–16. —— . “Peasant Seals.” In Medieval England, 1066–1485, edited by Edmund King, 214–15. Oxford: Phaidon, 1988. —— . “Twelfth-Century Forgeries as Evidence for Earlier Seals: The Case of St Dunstan.” In St Dunstan: His Life, Times and Cult, edited by Nigel Ramsay, Margaret Sparks, and T. W. T. Tatton-Brown, 299–310. Woodbridge: Boydell, 1992. Jenkinson, Charles Hilary. A Guide to Seals in the Public Record Office. 2nd ed. London: HMSO, 1968. —— . A Manual of Archive Administration Including the Problems of War Archive Making. Oxford: Clarendon, 1922. Libert, Marc. “The Seal Casts Collection and the Digitisation of the Sigillographic Collections of the National Archives of Belgium.” Vestnik of Saint Petersburg University (History) 62, no. 2 (2017): 255–66. https://doi.org/10.21638/11701/spbu02.2017.203. McEwan, John A. “The Challenge of the Visual: Making Medieval Seals Accessible in the Digital Age.” Journal of Documentation 71, no. 5 (2015): 999–1028. —— . Digisig: Digital Sigillography Resource. http://www.digisig.org/. —— . “Does Size Matter? Seals in England and Wales, ca.1200–1500.” In A Companion to Seals in the Middle Ages, edited by Laura Whatley, 103–28. Leiden: Brill, 2019. —— . “Formation of a Sealing Society: London in the Twelfth Century.” In Medieval Coins and Seals: Constructing Identity, Signifying Power, edited by Susan Solway, 319–30. Turnhout: Brepols, 2015. —— . “Making a Mark in Medieval London: The Social and Economic Standing of Seal-Makers.” In Seals and Their Context in the Middle Ages, edited by Phillipp R. Schofield, 77–88. Oxford: Oxbow, 2015. —— . “The Past, Present and Future of Sigillography: Towards a New Structural Standard for Seal Catalogues.” Archives and Records, no. 2 (2018): 224–243. https://doi.org/10.1080/ 23257962.2017.1353412.

48

John McEwan

—— . “Reflectance Transformation Imaging and the Future of Medieval Sigillography.” History Compass 16, no. 9 (2018): e12477. https://doi.org/10.1111/hic3.12477. —— . Seals in Medieval London, 1050–1300: A Catalogue. London: London Record Society, 2016. —— . “Seals in Medieval Wales and Its Neighbouring Counties: Trends in Motifs.” In Seals and Society in Medieval Wales and Its Border Region, edited by Elizabeth A. New and Phillipp R. Schofield, 13–34. Cardiff: University of Wales Press, 2016. ______. “Seals in Sussex.” BodoArXiv Works (2021). osf.io/preprints/bodoarxiv/7s3d9. —— . “Tout and Seals.” In Thomas Frederick Tout: Refashioning History in the 20th Century, edited by Joel T. Rosenthal and Caroline M. Barron, 185–98. London: Institute of Historical Research, 2019. McEwan, John A., and Elizabeth A. New. Seals in Context: Medieval Wales and the Welsh Marches. Aberystwyth: CAA Aberystwyth University, 2012. Melvyn, Jones. “Seal Repair, Moulds and Casts.” The Paper Conservator: Journal of the Institute of Paper Conservation 1, no. 1 (1976): 12–18. New, Elizabeth Anne. “Digital Imaging of British Medieval Seals: Report on a Six-Month Pilot Project.” Archives 26, no. 104 (2001): 11–17. —— . Seals and Sealing Practices. London: British Records Association, 2010. Pastoureau, Michel. Les Sceaux. Typologie des sources du Moyen Â� ge occidental 36. Turnhout: Brepols, 1981. Schofield, Phillipp R., Elizabeth Anne New, Susan M. Johns, and John A. McEwan. Seals and Society: Medieval Wales, the Welsh Marches and Their English Border Region. Cardiff: University of Wales Press, 2016. ______. “Seals in Medieval Wales and the Welsh Marches, 1200–1500 [Computer File]; Arts and Humanities Research Council (Ah/G010994/1): Principal Investigator, Phillipp Schofield; Co-Investigator, Sue Johns; Senior Research Officer, Elizabeth New, and Research Officer, John McEwan.” Aberystwyth, 2012. Steane, John. The Archaeology of the Medieval English Monarchy. London: Routledge, 1993. Tonnochy, A. B. Catalogue of British Seal Dies in the British Museum. London: British Museum, 1952. Tyacke, Sarah et al. The 40th Annual Report of the Keeper of Public Records on the Work of the Public Record Office and the 40th Report of the Advisory Council on Public Records 1998–99. London: Public Record Office, 1999. Villela-Petit, Inès. “Les techniques de moulage des sceaux du XVe au XIXe siècle.” Bibliothèque de l’École des chartes 152, no. 2 (1994): 511–20. Wyon, Alfred Benjamin and Allan Wyon. The Great Seals of England, from the Earliest Period to the Present Time, Arranged and Illustrated with Descriptive and Historical Notes. London: Stock, 1887.

Chapter 3

CORPUS SYNODALIUM: MEDIEVAL CANON LAW IN A DIGITAL AGE ROWAN DORIN*

Corpus Synodalium began, as so many research projects do, with a question that could not be answered. I had spent the previous five years exploring the spread of ideas and practices concerning mass expulsion in the later Middle Ages, a topic that had pulled me far from my earlier stomping grounds in the economic history of the high medieval Mediterranean. Instead, I found myself wading through Parisian sermon collections, Flemish settlement privileges, English administrative records, and—most fearsomely of all—the arcana of canon law. In 1274, in an attempt to halt the spread of professional Christian moneylending north of the Alps, the Second Council of Lyon had mandated the expulsion of foreign usurers.1 Early in my research, it occurred to me that it might be interesting to trace the dissemination of this conciliar decree in local sources, in particular the statutes that were issued at provincial councils and diocesan synods. After all, one of the primary purposes of such gatherings—as spelled out in the Fourth Lateran Council’s decree Sicut olim—was to disseminate knowledge of church law among the clergy, who might then proclaim it to the faithful.2 I thus was curious about the role that these local ecclesiastical statutes had played in normalizing the penalty of expulsion in Christian thought and practice. Tracking down these statutes proved considerably more challenging than I had expected. I discovered firsthand the limitations of the early modern conciliar collections *   I would like to thank the many, many people who have contributed their ideas, time, and expertise to the Corpus Synodalium project, especially Christine Barralis, Maria van Buiten, Charlie Donahue, Jasmin Hauck, Medina Husakovic, Kelly Katz, Sara Ann Knutson, Edward Dettmam Loss, Katie McDonough, Maureen Miller, Thawsitt Naing, Joel Pattison, May Peterson, Gavin Robinson, Clara Romani, Lauren Schlansky, Claire Womack, and the staff at Stanford’s Center for Spatial and Textual Analysis (CESTA). I would also like to express my gratitude to the Milton Fund at Harvard University, the Denning Fund for Humanities and Technologies at Stanford, the FACE Foundation’s Thomas Jefferson Fund, and the France–Stanford Center for Interdisciplinary Studies for their generous support of the project. A full list of acknowledgments can be found on the project website: http://www.corpus-synodalium.com.

1  Lyon II, c. 24, in Conciliorum oecumenicorum generaliumque decreta, ed. Alberigo and Melloni, 2/1:346–48. 2  Lateran IV, c. 6: in Conciliorum oecumenicorum generaliumque decreta, ed. Alberigo and Melloni, 2/1:170. Rowan Dorin, Department of History, Stanford University.

50

Rowan Dorin

(especially Mansi’s Sacrorum conciliorum);3 the frustrations of phantom references to long-lost manuscripts; the unpredictable hours of local diocesan archives; and (on a more positive note) the miracle-working abilities of my institution’s interlibrary loan staff. By the time I finished hunting down roughly eight hundred of these statutes, it was clear that the diffusion of the Lyonese expulsion decree was patchy at best. About thirty late medieval ecclesiastical statutes mentioned it explicitly, but most episcopal lawgivers simply ignored it, while a few others reworked the decree’s language and provisions so as to weaken its force. Whether because of ignorance, disinterest, discomfort, or resistance, the bishops of late medieval Europe had done little to disseminate knowledge of the decree’s demands. I presented these findings to my dissertation committee, only to be asked: “Were these patterns specific to this decree? Or was such sparse local reception true of conciliar decrees in general?” In retrospect, while I should have anticipated such a question, it had not occurred to me to gather comparative evidence as I scoured these texts for echoes of expulsion. I had no answer, nor was there any obvious means of coming up with one, short of fully retracing my research path. With my graduate funding exhausted and my completion deadline looming, that was an impossible task. I took some comfort in the reassuring words of one of my doctoral committee members, who noted that nobody else had ever tried to answer this question either, at least not for late medieval Christendom as a whole. Since the question was only tangentially related to the broader project, I set it aside to address more immediate (and readily resolvable) concerns. But the question continued to bother me just the same, since all of the possible answers seemed to unleash a new set of questions. If the limited dissemination of this particular decree was a unique occurrence, then what accounted for this anomaly? Strong reticence toward the penalty of expulsion? Widespread concerns about heightened punishment of usurers? Episcopal anxieties about encroaching too far on matters of secular jurisdiction? Alternatively, if a similar pattern held true for other conciliar decrees as well, what common features did they share, and did this pattern become more or less common over time? And if this pattern was a general phenomenon—i.e., if it was typical for bishops to ignore (or substantially rework) new conciliar decrees in their own legislation—then what exactly was the relationship of local episcopal statutes to the church’s general law? What were the limits of episcopal discretion? How variable was the church’s law as pronounced in individual dioceses in different parts of Latin Christendom, and how far did it deviate from what was being promulgated at the papal curia or taught in the lecture halls of Bologna? On the whole, specialists in late medieval canon law have shown far more interest in how legal doctrine was developed and debated than in how it was disseminated. There are some notable exceptions. Over the past few decades, for instance, several studies have examined regional differences in the legislation governing the formation of marriage and the enforcement thereof.4 A number of scholars have also examined the immediate reception of general church councils (especially Lateran IV) in different parts of 3  For a discussion, see Kay, “Mansi and Rouen.”

4  For examples, see Sheehan, “Marriage Theory and Practice”; Donahue, Law, Marriage, and



Medieval Canon Law in a Digital Age

51

Christendom, though these have often focused more on tallying up extant references to the conciliar decrees than on exploring the ways in which local statutes rewrote or selectively transmitted their contents.5 Such tallies are facilitated by the seductive appeal of the lists of sources (apparatus fontium) that modern editors so laboriously compile. But whether a local statute recopies verbatim the entire text of a conciliar decree, or instead adopts a few of its words for rhetorical flourish, critical conventions treat both borrowings equally, reducing such substantive variation to the same laconic acknowledgment of textual dependence. Meanwhile, scholars interested in many other fields of medieval studies often cite canonical texts in their codified form without worrying just how widely they were known, or buttress their arguments with references to local episcopal statutes without questioning whether these were representative or exceptional. Cumulatively, these approaches risk both overemphasizing the church’s universal law at the expense of its particular expressions, and treating as exemplary the local legislation that happens to be the most widely available and readily searchable. As for those—like me—who were curious about the diffusion of canonical norms into local legislation, there were high hurdles to overcome, especially if one wanted to push into the fourteenth century (let alone the fifteenth). Most provincial canons had been printed in at least one early modern collection of conciliar texts or archiepiscopal acta, and thanks to the spate of recent digitization campaigns, many of these weighty volumes could now be consulted online. In the absence of any repertories, however, tracking down these texts was a daunting challenge—to say nothing of the difficulties posed by disputed datings and conflicting textual traditions.6 For diocesan statutes, the editorial landscape was even more forbidding. Many early modern editors had simply passed over these texts, or else abandoned their early efforts in the face of their sheer number. The seventeenth-century Jesuit scholar Gabriel Cossart, for instance, had stopped editing diocesan statutes after reaching the year 1300, on the grounds that they were “huge in scale and small in harvest.”7 Few of his modern successors had fared much better. In the 1950s, a team of French scholars had produced a repertory of all medieval French synodal statutes, but the subsequent editorial labours of Odette Pontal and Joseph Avril had again focused almost entirely on thirteenth-century material.8 The exemplary editions of later medieval diocesan synods and provincial canons that C. R. Cheney and his colleagues had produced for England stopped at the Society in the Later Middle Ages; and the essays gathered by Mia Korpiola in Regional Variations in Matrimonial Law and Custom in Europe.

5  For two recent discussions of Lateran IV’s diffusion (with ample references to earlier literature), see Duggan, “Conciliar Law 1123–1215”; and Wayno, “Rethinking the Fourth Lateran Council of 1215.” For the later thirteenth century, see especially Boisset, “Les conciles provinciaux français”; and Unger, Generali concilio inhaerentes statuimus. 6  Pietro Palazzini’s six-volume Dizionario dei concili was a useful starting point, although it is far from exhaustive. Most of the corresponding volumes of the Konziliengeschichte series have yet to appear in print.

7  Sacrosancta concilia ad regiam editionem exacta, ed. Cossart and Labbé, 11/2:1467: “et vero nimia eorum esset moles, fructus exiguus.” 8  See Répertoire des statuts synodaux des diocèses de l’ancienne France, ed. Artonne, Guizard, and

52

Rowan Dorin

year 1313.9 The Italian material was a tangle, with dozens of transcriptions hidden away in unpublished tesi and many of the surviving sources still in manuscript.10 The same was true for most of the Holy Roman Empire, with the notable exception of Bohemia, for which the extant statutes had recently been edited in full.11 Elsewhere in Europe there were some welcome bright spots for the later Middle Ages, among them Jakub Sawicki’s ten-volume edition of medieval and early modern Polish ecclesiastical statutes.12 Nearly all of the Scandinavian texts were available in reliable nineteenth-century editions, even if the absence of a repertory made it difficult to track them down within the various national documentary collections in which they appeared. Most significantly, the monumental effort to produce critical editions of all diocesan statutes from late medieval Iberia, which Antonio Garcí�a y Garcí�a had begun in the 1970s and which Francisco Cantelar Rodrí�guez (and others) had then carried forward, was nearing completion.13 As this quick survey suggests, early modern and modern editorial efforts alike had largely proceeded along regional or national lines.14 Diocesan and provincial statutes rarely appeared in the same volumes, despite their frequent interdependence, and large swathes of western Europe remained essentially uncharted—and woe to anyone whose interests extended much beyond the thirteenth century. My own research had already demonstrated that with sufficient research time, travel funding, library resources, and language skills, it was not impossible to engage in a comparative study of this material across the breadth of late medieval Christendom. Still, the obstacles were significant, and I could certainly understand why no previous scholars had made the effort.15 If any future scholars were going to make the effort, surely they should at least be spared the need to reinvent the wheel. With that in mind, midway through my dissertation research I had posted online a downloadable version of my working repertory of Pontal; and Les statuts synodaux français du xiiie siècle, ed. Pontal and Avril. For a notable edition of provincial canons, see Les conciles de la province de Tours, ed. Avril.

9  Councils and Synods, ed. Powicke and Cheney. While fourteenth- and fifteenth-century England produced little in the way of diocesan statutes, its numerous provincial ones await a critical edition.

10  For a partial orientation, see Silvino da Nadro, Sinodi diocesani italiani. Catalogo bibliografico degli atti a stampa, 1879–1960, along with Tilatti, “Sinodi diocesane e concili provinciali in Italia nord-orientale.” 11  See Pražké synody a koncily, ed. Polc and Hlediková and Synody a statuta Olomoucké diecéze období středověku, ed. Krafl.

12  Concilia Poloniae, ed. Sawicki. See also the updated presentation in Krafl, Polské provinciální synody.

13  Synodicon Hispanum, ed. Garcí�a y Garcí�a. Twelve of the fourteen volumes had been published as of the project’s launch in 2016, with the outstanding volumes covering the dioceses of the Crown of Aragon.

14  Some important early collections gathered together the material from particular ecclesiastical provinces, such as Guillaume Bessin’s two-volume collection for Rouen (published 1717), or Florian Dalham’s collection for Salzburg (published 1788). 15  For an early call for such comparative work, see Helmrath, “Partikularsynoden und Synodalstatuten des späteren Mittelalters im europaï�schen Vergleich.”



Medieval Canon Law in a Digital Age

53

late medieval local ecclesiastical legislation. The first version, which I posted in December 2012, was a spreadsheet listing 715 diocesan, provincial, and legatine statutes issued between 1274 and 1400, along with information about their date, place, classification type, and whatever printed edition or manuscript source I had consulted. My initial post included an open plea for corrections and updates, and to my great delight these began to trickle in over the coming months and years. Drawing on this feedback as well as my ongoing discoveries, I continued to update the spreadsheet periodically, such that by the time I completed my dissertation in June 2015, the online repertory had grown to 850 entries spanning the period from 1192 to 1400. The repertory had significant shortcomings, some of which became evident only in hindsight. I had only scratched the surface of the fifteenth-century material, and even for the earlier period I knew that there were still more texts to be found, especially in Germany and Italy. I had also made little effort to track the editorial history of each statute, listing only a modern critical edition (where one existed) or whichever earlier version I had consulted, without much attention to variant textual traditions. As for those statutes that remained unpublished, only rarely had I attempted to trace all extant copies; the repertory thus listed only whichever manuscript I had managed to consult, which reflected more my travel constraints than the textual merits of the manuscript in question. These shortcomings mostly reflected the repertory’s origins, namely, as a private finding aid that I had compiled to keep track of my sources as I traced the local diffusion of a single conciliar decree. It therefore stood in marked contrast, for instance, to the vastly more systematic and comprehensive effort carried out by André Artonne and his team with regard to French synodal statutes. As I emphasized in the description accompanying the online version, it remained a work-in-progress. Moreover, although I planned to update the repertory periodically, I was not about to invest the effort needed to significantly expand its remit. My aims, at this point, were quite limited: I hoped that the repertory would alert other scholars to the sheer quantity of surviving material; help orient them to its distribution in time, place, and print; and spare them from redoing spadework that I had already done. Soon after I began my postdoctoral fellowship at the Harvard Society of Fellows, an open-ended grant opportunity landed in my lap, courtesy of Harvard’s William F. Milton Fund. In a burst of wild optimism (and grave miscalculation as to the time and effort that would be required), I began envisioning a project that would gather together fulltext transcriptions of all extant local ecclesiastical legislation into a digital database. Although my working repertory marked a first step in helping scholars to determine what was extant and where it could be found, many of the texts still remained hidden away in manuscript repositories or hard-to-find publications. By making such texts readily accessible in a single online collection, such a project would allow scholars to gain a much broader perspective than could be achieved using only those sources that were already comparatively easy to access—and to do so with only a few keystrokes. From a Digital Humanities perspective, the idea of gathering together a collection of transcribed texts was decidedly unadventurous. Having previously served as the student coordinator of a departmental pedagogy initiative aimed at incorporating digital projects into undergraduate History courses, and having also contributed directly to a

54

Rowan Dorin

handful of ongoing Digital Humanities projects, I was well aware of some of the exciting new tools and trends in the field. By comparison, this project smacked more of the early days of “humanities computing,” with its commitment to the creation of large-scale text corpora.16 Still, I drew inspiration from scholars who were drawing attention to the ways in which the expanding landscape of digital sources was transforming research trajectories, and the ominous implications for whatever sources were left behind.17 The field of medieval canon law was already dominated by scholarship on the so-called “learned law” of the classical period (from ca. 1140 to ca. 1250), and I worried that current tendencies in textual editing and digitization campaigns were likely to further skew scholarly attentions toward the “universal” at the expense of the “local.” Making the sources of the latter more readily accessible seemed a tangible way to address this concern. In a similar vein, I hoped that this project might draw attention to the statutes surviving from less well-trodden periods and places, as a counterweight to the long-standing (and paradigm-shaping) focus on thirteenth-century England and France. A full-text database would also help to overcome another challenge inherent in the sources themselves: the fact that they were often repetitive and derivative, and therefore (to be frank) rather boring—especially if one was flipping through thousands of pages of them in search of chance references to a particular word or theme (as I had done). To my mind, what seemed most promising about this material were the insights that could only emerge through comparison: subtle variations concealed beneath the veneer of seeming repetition; regional divergences in the reception of new church laws and norms; shifts in episcopal preoccupations across place and time. While previous scholars had considered some of these themes within relatively narrow chronological or geographical boundaries, a comprehensive digital database would presumably make it possible to explore them over the course of multiple centuries across the entirety of Latin Christendom.18 Moreover, these themes—concerning patterns, variations, borrowings, and discontinuities within large bodies of material—fell squarely within the ambit of the various text analysis techniques that scholars in the Digital Humanities had begun to explore in earnest. Assuming that the transcriptions were of sufficient quality, several features of this corpus made it particularly well-suited to these techniques. The statutes were generally localizable in time and place, having been issued on a specific date by a specific authority for a specific jurisdiction. That some statutes survived only via later compilations raised certain complications in this regard, but these cases were relatively unusual. The statutes were generally consistent in form: a series of injunctions of varying lengths, often introduced by a preamble identifying their issuing authority and the circumstances of 16  The fruits of this early period include such research staples as the cetedoc Library of Christian Latin Texts, the Patrologia Latina Database, the Acta Sanctorum Database, and many others. For the shift from Humanities Computing to Digital Humanities, see Svensson, “Humanities Computing as Digital Humanities.” 17  See especially Putnam, “The Transnational and the Text-Searchable.”

18  See especially Johanek, “Die Pariser Statuten des Bischofs Odo von Sully”; and Barralis, “Législation provinciale, législation diocésaine.”



Medieval Canon Law in a Digital Age

55

their promulgation. As for linguistic variation, aside from the Iberian peninsula (where nearly one-fifth of statutes survived in vernacular form), almost all of the extant texts were in Latin.19 Orthographic variation would pose some difficulties, but thanks to some prior experience in the field of computational philology, I knew that much of this variation followed predictable patterns and could therefore be “normalized” through an automated process. Furthermore, a Frankfurt-based research team had recently made considerable breakthroughs in the development of lemmatization tools for medieval Latin, which could help mitigate the distorting effects of word inflections.20 In short, I knew just enough about the sources to believe that it would be feasible to gather and transcribe most of them within the project’s two-year timeframe. I also knew just enough about the available tools to believe that once I had generated these transcriptions, it would be possible to answer the question that had stumped me at my dissertation defense. This was enough to convince me that the project would be a worthwhile undertaking, and I was heartened by the enthusiastic response of some senior colleagues in the field, even as they (rightly) cautioned that it was liable to distract me from the more pressing task of turning my dissertation into a book. Having solicited advice from Digital Humanities veterans and library staff on such topics as budgeting, copyright considerations, platform options, metadata structure, markup schemas, and archiving processes, I submitted the funding application and crossed my fingers. A few months later, I received the welcome news that my project had been awarded the grant, along with the equally welcome news that Stanford University—where I had just accepted a tenure-track position—was willing to match it. Now the real challenges would begin.

Building the Corpus

I began by identifying three guiding principles that I could use in making decisions about the project’s structure and workflow. The first was that the project should be designed for permanence (at least so far as any digital frameworks allowed). My past experience rummaging around the world of Digital Humanities had introduced me to the dangers of link rot, orphaned projects, obsolete file formats, and the like. As such, I was determined that the project’s outputs—whatever these might end up looking like—should be in a format that was resistant to obsolescence and could be stored in a stable institutional repository.21 The second principle was that the project should continue to be driven by specific research questions. This arose largely from my pedagogical work integrating digital projects into undergraduate courses: I had grown skeptical of anything that seemed to be driven by the appeal of building or using new tools, rather than by a clear sense of 19  Roughly 95 percent of the texts are written in Latin (though some of these feature brief passages in the vernacular). Other represented languages include Middle Dutch, Middle French, Middle High German, Old Norse, Old Swedish, and Tuscan. 20  See the Computational Historical Semantics Project (https://www.comphistsem.org/).

21  In my case, either the Harvard Dataverse (https://dataverse.harvard.edu/) or the Stanford Digital Repository (https://library.stanford.edu/research/stanford-digital-repository).

56

Rowan Dorin

the questions that they might be used to answer. Ideally, whatever I ended up creating would be flexible enough to answer all sorts of questions that I had not anticipated – but when it came to making decisions around scope and complexity, blue sky thinking would take a backseat to concrete scholarly desiderata. The third principle was that the project should not replicate (or worse, exacerbate) the existing unevenness in the availability of these sources. I did not want for the project to become only a tool for doing more efficiently what one could already do in two or three weeks of work at a well-stocked research library (even if that was certainly among the goals). A great deal of questionable digital humanities research (like a great deal of questionable quantitative social science research) relied on the analysis of digital corpora for which the selection criteria was essentially “whatever was easiest to digitize.” I took seriously the warnings of some of my future Stanford colleagues about the “tendency to privilege the ‘codeable’ and the ‘clean,’ the available and the cheap,” whereby “‘data’ will be shaped by and for the logic of digital tools.”22 Users could perhaps compensate for these tendencies if they understood the choices that were made about what to include or exclude in the digitization process, but too rarely did creators foreground these choices. So far as my own project was concerned, this meant that I could not begin by tackling all of the texts that existed in widely available and easy-to-digitize modern editions, lest I exhaust all of my funding before reaching the more obscure or recalcitrant texts, in which case my project would simply deepen the geographical and chronological asymmetries that I was aiming to overcome. Several early decisions flowed directly from these principles. In the interests of permanence and long-term interoperability, files in plain text format (.txt) would be used for the transcriptions. Meanwhile, the metadata associated with each transcription would be recorded in a separate spreadsheet file in .csv (comma-separated values) format. This was essentially a much-expanded version of the working repertory that now included up to forty fields for each record, ranging from the name of the issuing authority to the source of the transcription data. I knew that it would eventually be necessary to gather all of the project materials into a comprehensive user-friendly online framework, but setting up the transcriptions and metadata as autonomous elements from the start meant that they could also be preserved independently in the long run. This also meant that I could start producing the transcriptions immediately, while postponing the choice of an online framework until I had a better sense of the available options and the analyses to which they were suited. The latter concern was directly tied to my second guiding principle. To ensure that the project would be able to answer not only my own research questions, but also those of other scholars, I began reaching out to colleagues to solicit their ideas and wish-lists. Conveniently, the quadrennial International Congress of Medieval Canon Law was scheduled for three weeks after the initial grant funding was released, which allowed for many such conversations. Most fruitful of all were a series of trans-Atlantic workshops co-organized with a French colleague, Christine Barralis. Funded by an inaugural grant from the FACE Foundation’s Thomas Jefferson Fund (a new initiative 22  Nygren et al., “Connecting with the Past,” 63.



Medieval Canon Law in a Digital Age

57

for fostering Franco-American research collaborations) and matching support from the France–Stanford Center for Interdisciplinary Studies, we held four workshops over an eighteen-month period, which cumulatively brought together forty researchers, ranging from graduate students to senior faculty. These workshops offered an opportunity to get feedback on platform prototypes, consider new research questions and the adjustments needed to facilitate them, and reflect on our collective findings as the project developed. Regarding the third principle, in order for the project to include transcriptions of less accessible sources, I first had to establish where these might be found and how to acquire copies. This prompted a more extensive scouring of the literature, as well as fresh solicitations to colleagues across the United States and Europe for updates and corrections to the existing repertory. The tally of extant diocesan, provincial, and legatine statutes from 1200 to 1400 jumped to more than fourteen hundred (a sixty percent increase), while seven hundred statutes were identified from the fifteenth century. These findings alone marked a significant revision to the long-established consensus that saw the thirteenth century as the high point of episcopal statute-making.23 In fact, this revised tally made clear that more than twice as many diocesan statutes survive from the fourteenth century as from the thirteenth, and in much of Europe the trend continued apace in the fifteenth century.24 Meanwhile, I embarked on the necessary (though time-consuming and expensive) process of securing reproductions, which involved not only an endless stream of manuscript imaging orders and interlibrary loan requests, but also a cadre of obliging colleagues willing to snap photos in local repositories, share their own accumulated research materials, or hunt down contact information for the authors of unpublished theses dating back decades. It quickly became clear that the extant sources fell into three categories for the purposes of producing transcriptions. The first category included those that existed in machine-readable editions, generally those from the nineteenth and twentieth centuries. After being manually scanned (assuming they did not already exist in digital form), these texts were run through an OCR (Optical Character Recognition) program that was equipped with a built-in Latin dictionary, and which I gradually trained to recognize medieval Latin word-forms and recurring technical vocabulary. The resulting transcriptions were then reviewed and formatted by one of the undergraduate or graduate research assistants recruited to the project, all of whom had prior training in Latin and could thus catch most of the errors. I then reviewed each transcription myself as a final check. The second category included printed texts that current OCR technology could not easily handle, whether because of unusual characters, frequent abbreviations, or other complicating factors. Most of the major early modern conciliar collections fell into this category, as did incunables and other early printings. For most of these texts (amounting to ca. 650,000 transcribed words thus far), I relied on the services of Gavin Robinson, a 23  See, for instance, Trexler, “Diocesan Synods in Late Medieval Italy,” here drawing on similar earlier statements by C. R. Cheney and Odette Pontal.

24  Contrary to Trexler’s claims, for instance (“Diocesan Synods in Late Medieval Italy,” 295), twice as many Italian diocesan statutes survive from the fourteenth century as from the thirteenth, and the numbers for the fifteenth century are higher still.

58

Rowan Dorin

freelance UK-based transcriber who specializes in early modern English and Latin texts, followed again by my own further review. The final category included all transcriptions that would have to be made directly from manuscripts. Some of these I would transcribe myself, and several helpful colleagues also volunteered to produce transcriptions or set these as exercises for their paleography students. To tackle the rest, I hired a small team of European graduate students and postdoctoral researchers who had the necessary paleographical skills along with appropriate knowledge of local sources and writing conventions. Once each transcription was finished, it was sent to one of the research assistants to check for typos and systematize the formatting, after which I then did a final check of my own involving spot checks against the original manuscript rather than a systematic review of the entire transcription. Given the unanticipated increase in the total size of the corpus, some sort of selection process was needed to sort out which places, periods, or categories of texts should take priority. Adding transcriptions of previously unpublished texts was exciting but expensive; early accounting revealed that the average cost of adding a single unpublished text was equivalent to adding ten transcriptions extracted from modern editions. The cost of adding a text that had been outsourced for manual transcription fell roughly in the middle. I ultimately decided to focus on assembling a near-complete corpus of transcriptions of the extant statutes issued throughout Latin Christendom before 1350, which totalled roughly one thousand texts. Of these, roughly ten percent would have to be transcribed directly from manuscripts, another twenty percent would have to be manually retyped, and the rest could be scanned using OCR. This plan took advantage of the fact that the thirteenth-century sources were disproportionately well-edited (and hence quicker and cheaper to transcribe), while also pushing into the less familiar fourteenth century, thus adhering to my third guiding principle. Furthermore, privileging a chronological period rather than a geographical region would still make it possible for me to explore how the church’s universal law was diffused across the whole of Latin Christendom, thus aligning with my second guiding principle. The only significant modification to this plan arose as a result of the FACE Foundation workshops; since these added to the team several researchers who were experts in the history of the church in late medieval France, we expanded our transcription target to include all French statutes issued before 1400. Two other significant decisions had to be made at the outset of the project. The first was how to handle textual variants or competing textual traditions in the transmission of a particular set of statutes. One of the standout achievements of the Digital Humanities as practised by medievalists has been the production of dynamic editions that are freed from the constraints of the printed page, and which can thus convey more effectively the instability, or mouvance, of medieval texts.25 Given that the idea for this project had emerged out of my work on the unevenness of legal dissemination as expressed in part by episcopal reworkings of papal laws, I was very sensitive to the dangers of privi-

25  See Jänicke and Wrisley, “Visualizing Mouvance.” For an example centring on medieval legal sources, see the collaborative research project, The Community of the Realm in Scotland (https:// cotr.ac.uk/), which aims to produce a dynamic edition of the Declaration of Arbroath.



Medieval Canon Law in a Digital Age

59

leging certain texts as “authoritative” and thus obscuring their mutability. Moreover, I had already stumbled on compelling examples of scribal errors in the transmission of legal texts.26 A scribe making an early copy of the conciliar decrees of the Second Council of Lyon, for instance, had inadvertently dropped the word “not (non)” from a clause, thus dramatically transforming its meaning. The omission went unnoticed, and the manuscript made its way to Mainz, where the archbishop’s staff used its text as the starting point for a new set of statutes that they were drawing up for the diocese. A scribal slip in a single copy of the Lyonese decrees was thus reified into new law in one of the most important dioceses in Christendom.27 Beyond such examples of scribal error, there were also numerous statutes that survived in strikingly different versions for the same diocese or province, without any clear explanation for the divergences.28 Conceivably, different communities of readers would therefore have had correspondingly different understandings of the legal regime under which they lived. All of this seemed to open up intriguing new lines of research, yet I swiftly realized that it was going to be impossible to incorporate textual variants in any sort of systematic way, at least in the initial stages of the project. Given that I was extracting many of these texts from modern critical editions, I could not include material from their apparatus criticus without running afoul of copyright restrictions (to say nothing of scholarly ethics). Most of the early modern editions were such unreliable witnesses to their manuscript sources that the base manuscripts would have to be transcribed entirely anew. As for the manuscripts that were already going to be transcribed directly, here at least we were freer to adopt whatever editorial norms we saw fit, although collating manuscripts obviously would slow the project’s progress still further. (In this respect, it was comforting that most of the manuscript texts on our immediate priority list survived in only a single copy, thus rendering moot the question of collation.) It was with genuine regret, therefore, that I decided that the project’s transcriptions would simply replicate the texts as given in existing editions (despite the wide variation in their underlying editorial criteria), while resorting to a single “best witness” for the texts that survived only in manuscript. Where an edition highlighted the existence of important variants or alternate textual traditions, this would be recorded in the metadata but not incorporated into the transcriptions.29 The second immediate decision was whether to follow the widespread practice among Digital Humanities text-based projects of marking up transcriptions using a 26  For a rare discussion of this phenomenon and its implications, see Gaines Post, “Copyists’ Errors and the Problem of Papal Dispensations.”

27  For the exemplar (or copy thereof), see Vatican, Biblioteca Apostolica Vaticana, MS Pal. Lat. 832, fols. 91v–92r; the manuscript formerly belonged to the library of St. Martin’s Cathedral in Mainz. For the subsequent statutes (issued in 1274/1275), see the two extant copies: Oxford, Bodleian Library, MS Laud 401, fol. 3r; and Leipzig, Universitätsbibliothek, MS lat. 1085, fol. 34v.

28  For two examples among many, see the 1216/1217 provincial canons of Tours, in Les conciles de la province de Tours, ed. Avril, 119–25, at 124; and the 1305 diocesan statutes of Wrocław, in Concilia Poloniae, ed. Sawicki, 10:339–44. 29  We made an exception for vernacular medieval translations that survived in tandem with a Latin version, opting to include individual transcriptions of both texts within the corpus.

60

Rowan Dorin

standard encoding schema, most commonly TEI (Text Encoding Initiative). Such a schema could embed into the transcriptions information concerning paragraph divisions, foliation (for manuscripts) or page numbers (for editions), rubrication, marginal or interlinear annotations, textual variants, and much else besides. It could also mark out structural features (such as dating clauses, subscriptions), personal names, place names, legal citations, even the topics of each subsection or clause, all for the purposes of faciliating subsequent analysis. After much conversation and reflection, I decided against initial encoding. The project was not seeking to produce digital facsimile editions of texts, but rather searchable transcriptions—a fact that meant that much of the TEI-encoded information would be irrelevant to the project’s research goals. Copyright restrictions once again posed a concern for any paratextual information that we were extracting from modern critical editions. Encoding the texts from the outset would also have involved even more training for the team of research assistants, while much of the more subject-specific encoding work would likely have fallen to me, risking a significant bottleneck to the project’s progress. Finally, any TEI-encoded document would have to be run through another program in order to make it readily legible to non-specialist readers, which seemed in tension with my first guiding principle of permanence-through-simplicity. I took heart from discovering that a broader trend was afoot, in which many other text-based projects were opting “to forego rich encoding in order to quickly produce large corpora of texts.”30 I figured that if rejecting TEI meant being consigned to the outer darkness of the Digital Humanities universe, I at least would have some good company. Furthermore, thanks to some helpful advice from more experienced colleagues, I also knew that a lot of encoding could be generated automatically down the road, so long as the project adhered to consistent transcription conventions and maintained well-structured metadata. All internal numbering, for instance, was put in parentheses, while any notes, uncertainties, or formatting adjustments were indicated by a number in square brackets, with the corresponding information itself being entered in a “Notes” field in the metadata. Opting out of TEI at the outset therefore did not mean foregoing all encoding forever. With these decisions made, the transcription process began. The early stages were marked by some setbacks, in particular the high turnover among undergraduate research assistants, as their initial eagerness succumbed to the eye-numbing challenge of proofreading hundreds upon hundreds of lines of medieval Latin. A few extraordinary stalwarts emerged among the transcription team, however, with three doughty students together processing and reviewing nearly seven hundred and fifty texts. Within two years of the start of the project, we had completed the transcriptions of one thousand texts, thus coming very close to achieving our initial targets for both coverage and quantity. We decided to start expanding our efforts into the late fourteenth century, and by summer 2021 (five years after the start of the project) Corpus Synodalium contained roughly fourteen hundred texts, including nearly ninety percent of all extant diocesan and provincial legislation issued before 1400.31 30  Niles and Poston, “Re-Modeling the Edition,” 117.

31  The raw text files and their accompanying metadata are archived in the Stanford Digital



Going Online

Medieval Canon Law in a Digital Age

61

From the perspective of interoperability and long-term preservation, simplicity had its virtues. But if I wanted other scholars to use this material to answer their own research questions, leaving it in the form of .txt and .csv files was hardly an option. Few medievalists had the necessary familiarity with programming languages (particularly Python or R) to manipulate the raw data, and the project had been inspired by a desire to make this source base more accessible, not less so. That meant integrating the transcriptions with a platform that would not only make them easy to navigate, but would hopefully include some basic search and analysis tools as well. Lacking both the budget and skillset to create such a platform from scratch, I turned for advice to our department’s resident Academic Technology Specialist, Dr. Katherine McDonough, who had the double advantage of being deeply immersed in the digital humanities and being an accomplished scholar of early modern European history with plenty of experience grappling with non-English text corpora. She immediately suggested PhiloLogic, an open-source online text search, retrieval, and analysis tool.32 This had been developed out of the ARTFL Project, a long-running cooperative enterprise between the University of Chicago and the French CNRS that focused on the creation of digitized French-language texts. The most recent version (Philologic4) included a range of built-in features that lent themselves nicely to the sorts of research questions that had emerged from the discussions thus far. The search options were especially powerful, with faceted searches enabling the results to be explored and rearranged based on their underlying metadata. So far as Corpus Synodalium was concerned, this meant that one would easily be able to compare the distribution of findings across diocesan versus provincial statutes, or chart relative frequencies between different regions and periods. A “Keywords in Context (KWIC)” feature allowed for the quick comparison of search results to determine common sources or other intertextual relationships. Thanks to the “Time Series” feature, one could quickly visualize the chronological distribution of results. As shown in Figure 3.1, results could be displayed both in absolute terms (i.e., the number of search results falling within a given period) and in relative terms (i.e., the number of hits in proportion to the size of the text corpus from a given period). Most importantly of all, perhaps, Philologic4 allowed for various forms of so-called “fuzzy” searching. This would not only reduce the “noise” resulting from the wide orthographic variation among the texts, but also spared us the need to lemmatize the entire corpus in order to mitigate the problem of Latin declensions and conjugations.33 (Complex text analyses might still require such lemmatization, but Philologic4’s functionality would suffice for most run-of-the-mill searches.) Repository (https://doi.org/10.25740/zr170fy8693).

32  For more details, see the Philologic4 website (https://artfl-project.uchicago.edu/philologic4).

33  In the absence of fuzzy searching or corpus lemmatization, a search for the Latin word matrimonium would not turn up any of its inflected forms (e.g., matrimonia, matrimonio, matrimoniis, etc.).

62

Rowan Dorin

Figure 3.1a: Search results for the word matrimonium (marriage), showing absolute frequencies over twenty-five-year intervals.

Figure 3.1b: Search results for the word matrimonium (marriage), showing relative frequencies over twenty-five-year intervals.



Medieval Canon Law in a Digital Age

63

Three other features of the Philologic4 tool also proved appealing. First, unlike a host of text analysis tools that simply spit out results without much indication of what lies beneath, Philologic4 was admirably transparent: one could navigate smoothly between the visualization of results and the corresponding textual building-blocks.34 If a search result turned up a passage of interest, clicking on it would then take you to the full text of its originating source, so that one could quickly understand its textual context (figure 3.2). Similarly, if the Time Series seemed to show an unexpected spike for a particular period, one could click the corresponding bar and immediately display all of the underlying search results in order to check for outliers or other distortions. All of this offered reassuring means to check the robustness of the findings. Second, the tool was not only open-source, but was also being used to support numerous other high-profile (and long-running) projects, meaning that it was likely to enjoy ongoing institutional support at least for the immediate future. Finally (and this too was significant), our Academic Technology Specialist had worked closely with the Philologic developers in the past, and knew them to be creative, responsive, and eager to explore new uses for the tool. The next task was figuring out how to configure my accumulated materials to make them compatible with Philologic4, and equally, how to configure the tool to accommodate their peculiarities. To my great delight, one of the students in my medieval studies seminar that quarter was a computer science major with an interest in database design and a familiarity with all of the requisite programming languages. Within the span of six weeks, Thawsitt Naing managed to familiarize himself with the inner workings of both my corpus and Philologic4, and produce a prototype ready for presentation at our second trans-Atlantic workshop. Along the way, we had to confront once again the question of encoding, since Philologic4 required the inputted texts to conform to certain structuring guidelines (or schemas). Dividing up the texts into smaller chunks was fairly straightforward, since the transcription process had already involved numbering the individual articles within the statutes, and running the texts through a simple Python script could reformat the numbering system to meet Philologic’s requirements. Much more complicated was figuring out how to wrangle our elaborate metadata to make it compatible with Philologic’s schema. Conventional TEI standards were fine for expressing the date and place that a set of statutes was issued, or the classification to which it belonged (for instance,. diocesan, provincial, legatine, etc.), but some of our other information could not be readily matched to existing TEI categories.35 Fortunately, the developers behind Philologic4 had made the tool sufficiently flexible that we could mix-and-match different markup standards and create some customized elements (or “tags”) as well—even if the latter earned us some finger-wagging from more doctrinaire members of our local digital humanities community. 34  For a thoughtful discussion of Philologic’s comparative merits, see Jo Guldi, “Scholarly Infrastructure as Critical Argument.” 35  It was unclear, for instance, what TEI tag should be used to record whether a bishop had delegated the task of issuing the statutes to his vicar general.

64

Rowan Dorin

Figure 3.2: Search results for the word matrimonium (marriage); clicking any result brings up the full text with the search term highlighted.



Medieval Canon Law in a Digital Age

65

Over the following year, Thawsitt managed to add several improvements to our online tool in response to user feedback. Among these was a “normalized” version of the text corpus in which classical Latin diphthongs (ae, oe, etc.) were rendered in their standard medieval form, v’s and j’s were transformed into u’s and i’s, and so forth, all with the aim of further minimizing distortions in the search results. (Users who wished to work with the unmodified corpus could of course still do so.) We also made it easier to display the metadata associated with any given text, and expanded the range of searchable fields and facets. The most significant new feature, however, was a novel mapping tool through which one could visualize the spatial distribution of the search results. I had been interested in hunting for spatial patterns since the early days of my research into these sources, and already in my dissertation I had drawn some maps depicting the differential regional reception of the Lyonese expulsion provision. For these early efforts, I had simply plotted the statutes according to the latitude and longitude of their issuing location, but I was unsatisfied with the resulting maps. Most obviously, the statutes theoretically applied to the entire area encompassed by the associated diocese or province, rather than to single points therein. This was an adequate approach for regions in which the jurisdictions were relatively small (such as Italy or southern France), but it was misleading when applied to the sprawling ecclesiastical jurisdictions that characterized much of northern Europe. The province of Bourges vertically spanned half of France, for instance, but the metropolitan see was situated in the very northern part of the diocese. Meanwhile, the province of Mainz ran from the Alps almost to the North Sea, something that could hardly be conveyed by a single dot on the city of Mainz itself. But given the absence of any digitized boundary data for most of the ecclesiastical jurisdictions of late medieval Latin Christendom, plotting points was the best available option. The original Corpus Synodalium grant application had therefore included a proposal for creating a digital map of late medieval dioceses and provinces. While the transcriptions team set to work on creating the text corpus, I worked with another team of research assistants to assemble the mapping materials. As I knew from previous Geographic Information System (gis) training workshops, this process was straightforward enough in theory: find printed maps, scan them at sufficient resolution, and then manually extract their boundary information using the appropriate software suite. I figured that it would prove more complicated in practice, but in retrospect I grossly underestimated just how complicated it would be. Establishing the roster of dioceses and provinces was the first hurdle, given the expanding boundaries of late medieval Latin Christendom, long-running disputes over suffragan lists, the frequent reorganizations of ecclesiastical jurisdictions (especially in Italy and Iberia), and continuing uncertainty over the foundation or suppression dates of many dioceses. An even greater hurdle was finding suitable maps. To our great relief, we discovered usable digital map files for medieval British and French dioceses. High-quality printed maps also existed for the dioceses of the Holy Roman Empire and the Italian peninsula, though these did not always capture the many small shifts in the ecclesiastical geography of either region over the course of the later Middle Ages, and extracting the boundary information for each individual diocese remained a taxing process. As for

66

Rowan Dorin

Figure 3.3: Map showing frequency of references to matrimonium (marriage) in diocesan statutes; clicking any jurisdiction brings up its associated texts with the search term highlighted.



Medieval Canon Law in a Digital Age

67

the rest of Latin Christendom, we ended up relying on more than one hundred sources, ranging from major historical atlases to hand-drawn maps hidden away in studies of local church history. (Here we were very grateful for the advice and materials provided by numerous colleagues across the United States and Europe.) Iberia proved especially challenging, and other regions proved simply impossible: we ultimately abandoned our effort to include many of the titular, short-lived, or otherwise poorly-documented dioceses in the Latin East and along the Slavic frontier. Fortunately for our purposes, none of the omitted dioceses left behind extant episcopal statutes, and at any rate, it is unclear to what extent many of them ever had firm territorial boundaries in the first place. Led by a dedicated undergraduate, Clara Romani, the mapping team ultimately generated boundary data for some 720 dioceses and 90 provinces.36 Thawsitt then created a script that could feed search results directly from the Philologic4 web application to the standalone map site.37 The map displays the number of search results within each jurisdiction, using a colour gradient to indicate frequency. (Thawsitt thoughtfully included a black-and-white option to accommodate my colour-blindness.) Clicking on a jurisdiction pulls up a list of the associated results, and clicking on any of these leads back to the corresponding entry in the full-text database (figure 3.3). This therefore keeps the map from becoming a mere abstraction of the underlying text data; instead, it functions in a dynamic and transparent relationship to its source material. We immediately faced the challenge, however, of determining which year to use as a basemap. Many of the jurisdictions changed form at least once between 1200 and 1500, as new dioceses were erected and others suppressed, or as provinces acquired and lost suffragans. If we used the year 1350 as a basemap, for instance, then no search results would be displayed from dioceses that were either suppressed before that date or created afterwards. The same was true for provinces. In resolving this problem, we returned to the principle that the project’s development should be driven by specific research questions. For almost every spatially-oriented question that we had come up with thus far, it was more important to display the complete search results than to display rigorously accurate boundaries. With that in mind, we ended up creating a composite map that includes virtually every diocese or province that existed at any point between 1200 and 1500, even if at no point during this period did these jurisdictions all exist simultaneously.38 While the text corpus and online tool were essentially ready for release by October 2019 (a little more than three years after the project’s launch), another year was needed to finish up the mapping tool. What had been proposed as a two-year project 36  The boundary data and accompanying documentation can be downloaded from the Stanford Digital Repository (https://doi.org/10.25740/rh195hm5975).

37  We considered integrating the map directly into the search interface, but this proved too much of a lift in terms of the necessary coding work.

38  In order that our mapping efforts might prove useful to researchers beyond the confines of the Corpus Synodalium project, we also produced a separate set of maps corresponding to the years 1250, 1350, and 1450. While these were not integrated into the database, they are all freely downloadable online.

68

Rowan Dorin

therefore expanded to more than four. The project’s timeline was not the only thing that doubled; so too did its associated costs. Having quickly exhausted the initial Milton Fund grant and Stanford’s matching institutional contribution, I gradually cobbled together funding from a range of other sources. As noted already, grants from two organizations devoted to fostering trans-Atlantic collaboration provided the bulk of the funding for the four project workshops. Summer undergraduate research grants from several internal sources covered most of the wages for the mapping team, while a (now-discontinued) Stanford grant designed specifically to support Digital Humanities projects covered the cost of my undergraduate developer. A surprise windfall from a faculty award allowed me to commission a dozen additional manuscript transcriptions, and I repeatedly drained my annual research allowance to cover the remaining project expenses. Had it not been possible to muster up additional funding, Corpus Synodalium would look very different: no mapping tool, fewer manuscript transcriptions, and none of the features and improvements that arose as a result of the workshops. Moreover, it was only due to the support of two generous institutions that I was able to launch the project in the first place, and I fully appreciate that few other medievalists can count themselves so fortunate. That said, more than half of the project’s total funding came through competitive grants that were open to any field, rather than being restricted to the humanities, let alone the digital humanities. That I even considered applying for these grants is due mainly to a handful of colleagues and collaborators who swatted away my instinctive skepticism, and I hope that the project’s successes on this front will encourage other digitally-minded medievalists to be similarly ecumenical in seeking support.

Early Results

The project’s public launch was initially scheduled for June 2020, a date that was then postponed by a year due to pandemic-related delays. Even before its widespread release, however, news of the database had been spreading via our workshops, seminars, conference presentations, internet searches, and word-of-mouth, with the resulting research findings already confirming the project’s potential. A doctoral candidate in Hungary, for instance, used the database to reveal that the influence of Bishop William Durand’s 1292 statute compilation for his diocese of Mende was even greater than had previously been realized, extending well into central-eastern Europe. Rather than occurring directly, however, evidence from the database demonstrated that the textual borrowings took place through a chain of copying and reworking routed mainly through northern Italy.39 A postdoctoral scholar studying voluntary reclusion in late medieval Italy (which has been much less studied than comparable phenomena north of the Alps) used the database to demonstrate the exceptional nature of a cluster of Umbrian and Tuscan diocesan statutes on the topic.40 Looking broadly at the corpus as a whole, a Stanford doctoral student identified a widespread episcopal concern with the regulation of noise, 39  Szilvia Somogyi, “Az esztergomi szinodális könyvről” and her “Das spätmittelalterliche Synodalbuch der Diözese von Gran.” 40  Allegria, “Costituzioni diocesane e reclusione volontaria.”



Medieval Canon Law in a Digital Age

69

one that was limited not only to internal church spaces or cemeteries, but which gradually extended outward to encompass much of the lay landscape as well. Since these concerns go almost entirely unmentioned in the codified canon law (as represented in the Corpus iuris canonici), this finding thus reveals how the comprehensive examination of local ecclesiastical legislation can provide an expanded understanding of the normative ambitions of the late medieval church.41 As for the question that had launched this project in this first place, it was now possible to determine the local reception of almost any provision of canon law with only a few keystrokes (rather than weeks or years of research). Little effort was therefore needed to establish that the varied responses of late medieval bishops to the Lyonese expulsion decree were indeed representative of broader patterns in the local reception of the church’s universal law, rather than exceptional. Much more work remains to be done to understand the dynamics of these responses, especially the degree to which they varied according to place, period, the personality and training of the issuing bishop, or the topic and tenor of the laws themselves. It is already clear, however, that bishops exercised considerable discretion in choosing which new laws—and which elements thereof—to incorporate in their statutes. To take one example, in disseminating the corpus of canonical restrictions governing Christian usury, late medieval bishops systematically downplayed the importance of papal decretals on this subject, highlighting instead the decrees issued by general church councils. Even with regard to the latter, bishops picked and chose such elements as they deemed important or appropriate, frequently omitting particular penalties or relaxing the sanctions for clerical transgressors. As a result, in only a handful of dioceses across Latin Christendom did the accumulated corpus of local statutes come to include all of the substantive penalties on both usurers and those who facilitated their wrongdoing. In short, the church’s law as proclaimed in local synods and thus conveyed to priests and parishioners could look very different from what was being promulgated at the papal curia and taught in law schools—a finding that upends longstanding scholarly assumptions regarding the role of episcopal statutes in disseminating the universal law to local communities.42 The ability to search the corpus quickly and broadly has revealed other surprising patterns as well. As other scholars have noted, computers are generally better than humans at identifying things that appear rarely or not at all in a given set of texts, and Corpus Synodalium has accordingly proven to be a powerful tool for identifying omissions and silences in the surviving examples of local ecclesiastical legislation.43 These are especially striking with regard to the ways in which this legislation articulated its relationship to higher sources of law. Even where the drafters of thirteenth- and early fourteenth-century diocesan statutes did draw directly from the church’s universal law, they frequently downplayed this dependence, even going so far as to quote entire con41  Baker, “Regulating Noise with Church Law.”

42  For a more detailed discussion of the findings summarized in this paragraph and the following one, see Dorin, “The Bishop as Lawmaker.” 43  Witmor and Hope, “Books in Space,” 22.

70

Rowan Dorin

ciliar decrees verbatim with no acknowledgment of their source. Notwithstanding considerable disagreement among canonists as to the extent of episcopal legislative authority, bishops embraced language that asserted their own role as law-makers rather than presenting themselves as mere mouthpieces of laws issued by popes and councils. For the clergy who heard or read the resulting statutes (as well as the parishioners to whom they were announced), they appeared not as adaptations or restatements of general laws issued by popes and councils, but as expressions of episcopal will. It remains to be seen whether this pattern persisted into the later fourteenth century and beyond, but this preliminary finding already suggests the need to explore these statutes from the perspective of those who were meant to abide by them, unmediated by the strictures of medieval canonists or the apparatus fontium of modern editors.

Current Concerns and Future Challenges

As early users began to poke around in Corpus Synodalium, or better yet draw on it for their research, certain challenges emerged. Some of these could be addressed through tweaks to the online tool, as discussed already. Other challenges concerned the interpretation of search results, with the illusion of completeness being especially distorting. The project website includes a report of the rate of completed transcriptions for different regions and time periods, and an even more precise indicator is provided by the downloadable repertory, which highlights all of the known texts that have already been added to the database. But one can easily skip over this information en route to the search interface, and therefore not realize that (as of summer 2021) the database includes only fourteen percent of the extant fifteenth-century statutes. One could equally run a search for references to adultery or heresy and conclude that the bishops of southern Italy ignored these topics in their statutes, whereas this silence instead reflects the fact that the bishops of southern Italy issued few extant statutes on any topic. Similarly, one might look for references to Jews in the ecclesiastical legislation from the province of Reims and wonder at the divergent responses of the dioceses of Cambrai, Noyon, Soissons, and Tournai (all of which issued statutes mentioning Jews) compared to those of Amiens, Beauvais, Châlons-sur-Marne, Laon, Thérouanne, and Senlis (in which there are no references to Jews), without realizing that almost no thirteenth- or fourteenthcentury statutes survive from the latter dioceses. In short, the breadth and scale of the database make it dangerously easy to overlook what is absent—statutes still to be added, statutes that have been lost over time, or ecclesiastical jurisdictions that never issued statutes in the first place. When consulting a modern edition of the statutes from a particular diocese or province, for instance, a reader understands implicitly that statutes from the rest of Latin Christendom are not included. But when a user runs a search in the Corpus Synodalium database, it is much less straightforward to know what is encompassed in the search versus what is missing. The mapping tool is specifically designed to mitigate this issue: alongside displaying the frequency of search results for any given jurisdiction (whether a diocese or province), it also visually distinguishes between areas with zero results and areas with no associated texts in the database. On a more general level, prior to the creation of Corpus Synodalium



Medieval Canon Law in a Digital Age

71

the uneven distribution of modern critical editions meant that whole regions and time periods were too readily ignored in comparative or synoptic studies; by contrast, careful users can now use the repertory and full-text database in tandem to correct for such skewed perspectives. It certainly remains possible for users to draw hasty conclusions from the corpus without paying heed to its particular minefields, but perhaps this is the inescapable consequence of making these sources more readily accessible. As with any large text database, the convenience of accessing and manipulating the corpus as a whole must be weighed against the loss of granularity that is gained by examining each text individually. It seems unlikely that anyone else will ever read through all of the texts in the corpus, and anyway the primary purpose of producing online searchable transcriptions of all of these texts was precisely so that nobody else would ever have to read through all of them. Even so, the database was certainly not intended as a substitute for existing critical editions, which contained a wealth of textual and contextual information that was omitted from the online transcriptions (for the reasons discussed above). If users turned up relevant passages or patterns, my initial assumption was that they would then check these against the corresponding printed editions, or even against the manuscripts (at least where these were freely available in digital form already). It became clear, however, that many users’ research efforts would carry them no further than the online database, with search results being noted but not checked further. Often this was because of the difficulty of hunting down the underlying sources—which of course was one of the problems that had inspired the project in the first place. But the price of gathering together simple transcriptions was the stripping away of much of the information that would ordinarily frame one’s encounter with an edited text: the context of its promulgation; its relationship to earlier texts from the same jurisdiction or another one; the circumstances of its transmission; the existence of variant textual traditions; and so forth. The metadata for the transcriptions occasionally recorded some of this information, though never with the level of detail found in a modern critical edition. Then again, many of the standard editions of these texts (such as Mansi and other early conciliar collections) present them in a similarly flattened form, devoid of much contextual information and with little regard for the complexity of their textual transmission. From this perspective, many of the online transcriptions are certainly no worse than the editions from which they derive, and often they are markedly better.44 Citation practices raised another challenge. The Philologic4 interface helpfully makes use of a dynamic URL that embeds all search terms and parameters within itself; users could therefore save and replicate any search simply by copying and pasting the associated URL. Every text (and every subsection of a text) likewise possessed its own distinctive URL. When it came to publishing their research, however, many users understandably preferred to cite their results according to the corresponding printed editions or manuscripts, rather than use a complex URL (which might eventually end up as a 44  On this issue, see Davis, “Content is not Context,” especially at 95, as well as the essays gathered in “The Trials of the Digital Medievalist”, a special issue of Digital Philology edited by Aditi Nafde and Emma Gorst.

72

Rowan Dorin

dead link).45 Thawsitt therefore added a feature by which users could instantly generate a downloadable list of abbreviated bibliographical references for every text that appeared in the search results. There was a catch, since our early decision to forego rich encoding of the transcriptions meant that we had not recorded original pagination or foliation. Consequently, any users seeking to cite specific page numbers for a particular chapter or passage (rather than just the page range corresponding to the text as a whole) must locate the original source on their own and then identify the appropriate page, thus raising anew the problem of source access. Then again, if users cannot easily access the original sources for the purposes of providing exact citations, they are more likely to acknowledge their use of Corpus Synodalium and their reliance on its transcriptions as the source of their research findings. The issue of acknowledgment raises the broader question of scholarly recognition and (e)valuation. During the project’s development phase, the online database was password-protected and the transcription files were not made available for separate download. As a result, early users generally had to request permission to access the database, which made it more likely that they would reference the project in any resulting publications. The public version of the project includes no such gatekeeping features, and users may therefore feel less obligation to acknowledge it. Such public acknowledgements matter, first and foremost, because they allow me to see how the database is being used and the questions that it is helping scholars to answer. On a more pragmatic level, they offer concrete evidence of the research value of the time, effort, and resources that underpin the project. This evidence will certainly be helpful should I end up seeking additional funding to expand the project. But it also matters on a more immediate level. The nature of my current position—as an untenured junior faculty member at a research-intensive university—means the work that I have put into the project accrues institutional value only inasmuch as it can be captured in the form of original scholarly publications.46 It was presumably with this in mind that a senior medievalist colleague, upon hearing my initial project proposal, responded: “This sounds like something that ought to be tackled at a later stage of one’s career.” I freely confess that over the past five years I have periodically wondered whether I should have deferred to his judgment. For now, at least, the jury is still out, but I am heartened by the research results that Corpus Synodalium has generated thus far. Finally, there is the matter of the project’s future development and long-term sustainability. Expanding the database into the fifteenth century would allow for a host of new research questions to be explored, such as the impact of the Observant reform movement on the contents of local statutes or the effects of late medieval ecclesiological controversies on these statutes’ language and substance. Moreover, many dioceses preserved their early legislation only in the form of fifteenth-century compilations, such 45  See Fisher, “Authority, Interoperability, and Digital Medieval Scholarship.”

46  This constraint long has stymied other efforts to develop digital corpora of medieval texts, along with other forms of outward-facing digital infrastructure; see Widner, “Toward Text-Mining the Middle Ages”, and Guldi, “Scholarly Infrastructure as Critical Argument.”



Medieval Canon Law in a Digital Age

73

that an investigation of this later material is necessary for understanding what came before. For now, however, the database will remain focused on pre-1400 material, with transcriptions of the remaining texts from this earlier period to be added as time and funding allow. At some point over the coming years, it also may become necessary to rework the online tool to be compatible with new releases of Philologic, especially once Philologic4 is no longer being actively maintained or supported. At that point I will no longer be able to fall back on the continuing generosity and availability of the project’s original developer, though his successor will at least be able to rely on the ample and careful documentation that he compiled throughout the process. Looking further into the future, Corpus Synodalium will eventually succumb to the fate of nearly all digital humanities projects, doomed to dysfunctionality, clunkiness, or neglect. Even then, however, the raw materials from which the online database was constructed—the transcriptions, repertory, and mapping data—will persist in stable institutional repositories, awaiting both the questions of a new generation of medievalists and the answers that new digital methods will provide.

Bibliography Allegria, Simone. “Costituzioni diocesane e reclusione volontaria in Italia centrale tra Due e Trecento. Qualche riflessione.” Quaderni di storia religiosa medievale 24 (2021): 143–65. Artonne, André, Louis Guizard, and Odette Pontal, eds. Répertoire des statuts synodaux des diocèses de l’ancienne France du xiiie à la fin du xviiie siècle. 2nd ed. Paris: CNRS, 1969. Avril, Joseph, ed. Les conciles de la province de Tours / Concilia provinciae Turonensis (saec. xiii–xv). Paris: CNRS, 1987. Baker, Lane. “Regulating Noise with Church Law (1200–1400).” Paper presented at the California Medieval History Seminar (Huntington Library, San Marino, CA), February 2020. Barralis, Christine. “Législation provinciale, législation diocésaine dans la province de Reims aux xive et xve siècles.” Travaux de l’Académie nationale de Reims 178 (2008): 353–64. Boisset, Louis. “Les conciles provinciaux français et la réception des décrets du IIe Concile de Lyon (1274).” Revue d’histoire de l’Église de France 69 (1983): 29–59. The Community of the Realm in Scotland. https://cotr.ac.uk/. The Computational Historical Semantics Project. https://www.comphistsem.org/. Concilia Poloniae. Ed. Jakub Sawicki. 10 vols. Warsaw: Nakł. Tow. Naukowego Warszawskiego, 1948–1963. Conciliorum oecumenicorum generaliumque decreta. Ed. Guiseppe Alberigo and Alberto Melloni. 3 vols. in 2. Turnhout: Brepols, 2006–2013. Councils and Synods, with other Documents Relating to the English Church. Vol. ii: a.d. 1205– 1313. Ed. F. M. Powicke and C. R. Cheney. Oxford: Clarendon, 1964. da Nadro, Silvino. Sinodi diocesani italiani. Catalogo bibliografico degli atti a stampa, 1879– 1960. Milan: Centro Studi Cappuccini Lombardi, 1962. Davis, Matthew Evan. “Content is not Context: Radical Transparency and the Acknowledgement of Informational Palimpsests in Online Display.” In Meeting the Medieval in a Digital World, edited by Matthew Evan Davis, Tamsyn Mahoney-Steel, and Ece Turnator, 93–122. Leeds: Arc Humanities Press, 2018.

74

Rowan Dorin

Donahue, Charles Jr. Law, Marriage, and Society in the Later Middle Ages: Arguments about Marriage in Five Courts. Cambridge: Cambridge University Press, 2007. Dorin, Rowan. “The Bishop as Lawmaker in Late Medieval Europe,” Past and Present 253 (March 2021): 45–82. https://doi.org/10.1093/pastj/gtaa045. ______. “Corpus Synodalium (March 2021): Archiving Dossier Narrative.” BodoArXiv Works, March 24, 2021. doi:10.34055/osf.io/qx2ve. ______. Corpus Synodalium: Repertory and Corpus of Local Ecclesiastical Legislation in Late Medieval Europe, 1200–1500. Stanford Digital Repository, 2021. https://purl.stanford. edu/zr170fy8693. ______. ed. Corpus Synodalium: Local Ecclesiastical Legislation in Medieval Europe, June 30, 2021. http://www.corpus-synodalium.com. Duggan, Anne J. “Conciliar Law 1123–1215: The Legislation of the Four Lateran Councils.” In The History of Medieval Canon Law in the Classical Period, 1140–1234: From Gratian to the Decretals of Pope Gregory IX, edited by Wilfried Hartmann and Kenneth Pennington, 318–66. Washington, DC: Catholic University of America Press, 2008. Estill, Laura, Diane K. Jackacki, and Michael Ullyo, eds. Early Modern Studies after the Digital Turn. Tempe: Arizona Center for Medieval and Renaissance Studies, 2016. Fisher, Matthew. “Authority, Interoperability, and Digital Medieval Scholarship.” In “E-Medieval: Teaching, Research, and the Net.” Special issue, Literature Compass 9, no. 12 (December 2012): 955–64. Guldi, Jo. “Scholarly Infrastructure as Critical Argument: Nine principles in a preliminary survey of the bibliographic and critical values expressed by scholarly web-portals for visualizing data.” Digital Humanities Quarterly 14, no. 3 (2020): http://digitalhumanities. org:8081/dhq/vol/14/3/000463/000463.html. Helmrath, Johannes. “Partikularsynoden und Synodalstatuten des späteren Mittelalters im europaï�schen Vergleich: Vorüberlegungen zu einem möglichen Projekt.” Annuarium Historiae Conciliorum 34 (2002): 57–99. Jänicke, Stefan and David Joseph Wrisley, “Visualizing Mouvance: Toward a Visual Analysis of Variant Medieval Text Traditions.” Digital Scholarship in the Humanities 32, Supplement 2 (December 2017): 106–23. Johanek, Peter. “Die Pariser Statuten des Bischofs Odo von Sully und die Anfänge der kirchlichen Statutengesetzgebung in Deutschland.” In Proceedings of the Seventh International Congress of Medieval Canon Law. Cambridge, 23–27 July 1984, edited by Peter Linehan, 327–47. Vatican City: Biblioteca Apostolica Vaticana, 1988. Kay, Richard. “Mansi and Rouen: A Critique of the Conciliar Collections.” The Catholic Historical Review 52 (1966): 155–85. Korpiola, Mia, ed. Regional Variations in Matrimonial Law and Custom in Europe, 1150–1600. Leiden: Brill, 2011. Krafl, Pavel. Polské provinciální synody 13.–15. století. Prague: Historický ústav, 2016. Nafde, Aditi and Emma Gorst, eds. “The Trials of the Digital Medievalist.” Special issue, Digital Philology 4, no. 2 (Fall 2015): 147–308. Niles, Rebecca and Michael Poston. “Re-Modeling the Edition: Creating the Corpus of Folger Digital Texts.” In Early Modern Studies after the Digital Turn, edited by Estill et al., 116–44. Tempe: Arizona Center for Medieval and Renaissance Studies, 2016. Nygren, Thomas, Zephyr Frank, Nicholas Bauch, and Erik Steiner. “Connecting with the Past: Opportunities and Challenges in Digital History.” In Research Methods for Creating and Curating Data in the Digital Humanities, edited by Matt Hayler and Gabriele Griffin, 62–86. Edinburgh: Edinburgh University Press, 2016.



Medieval Canon Law in a Digital Age

75

Palazzini, Pietro, ed. Dizionario dei concili. 6 vols. Rome: Istituto Giovanni XXIII nella Pontificia Università lateranense, 1963–1967. Philologic4. https://artfl-project.uchicago.edu/philologic4. Pontal, Odette and Joseph Avril, eds. Les statuts synodaux français du xiiie siècle. 6 vols. Paris: CTHS, 1971–2011. Post, Gaines. “Copyists’ Errors and the Problem of Papal Dispensations ‘contra statutum generale Ecclesiae’ or ‘contra statum generalem ecclesiae’ according to the Decretists and Decretalists, ca. 1150–1234.” Studia Gratiana 9 (1966): 357–405. Pražké synody a koncily (předhusitské doby). Ed. Jaroslav Polc and Zdeňka Hlediková. Prague: Univerzita Karlova, 2002. Putnam, Lara. “The Transnational and the Text-Searchable: Digitized Sources and the Shadows They Cast.” American Historical Review 121 (2016): 377–402. Sacrosancta concilia ad regiam editionem exacta. Ed. Gabriel Cossart and Philippe Labbé. 16 vols. Paris: n. publ., 1671–1672. Sheehan, Michael M. “Marriage Theory and Practice in the Conciliar Legislation and Diocesan Statutes of Medieval England.” Mediaeval Studies 40 (1978): 408–60. Somogyi, Szilvia. “Az esztergomi szinodális könyvről: új források, redakciók és szövegtöbbletek nyomában. (Előtanulmány egy disszertációfejezethez).” Magyar Könyvszemle 135 (2019): 1–17. ______. “Das spätmittelalterliche Synodalbuch der Diözese von Gran.” Annales Historiae Conciliorum 51 (2021, forthcoming). Svensson, Patrik. “Humanities Computing as Digital Humanities.” Digital Humanities Quarterly 3, no. 3 (2009): unpag. http://digitalhumanities.org/dhq/vol/3/3/000065/000065. html. Synodicon Hispanum. Ed. Antonio Garcí�a y Garcí�a et al. 14 vols. Madrid: Biblioteca de autores cristianos, 1981–2020. Synody a statuta Olomoucké diecéze období středověku. Ed. Pavel Krafl. Prague: Historický ústav, 2003. Tilatti, Andrea. “Sinodi diocesane e concili provinciali in Italia nord-orientale fra Due e Trecento. Qualche riflessione.” Mélanges de l’Ecole française de Rome. Moyen Âge 112 (2000): 273–304. Trexler, Richard C. “Diocesan Synods in Late Medieval Italy.” In Vescovi e diocesi in Italia dal xiv alla metà del xvi secolo. Atti del vii Convegno di storia della Chiesa in Italia (Brescia, 21–25 settembre 1987), edited by Giuseppina de Sandre Gasparini et al., 1:295–335. 2 vols. Rome: Herder, 1990. Unger, Stefanie. Generali concilio inhaerentes statuimus: Die Rezeption des Vierten Lateranum (1215) und des Zweiten Lugdunense (1274) in den Statuten der Erzbischöfe von Köln und Mainz bis zum Jahr 1310. Mainz: Gesellschaft für mittelrheinische Kirchengeschichte, 2004. Wayno, Jeffrey. “Rethinking the Fourth Lateran Council of 1215.” Speculum 93, no. 3 (2018): 611–37. Widner, Michael. “Toward Text-Mining the Middle Ages: Digital Scriptoria and Networks of Labor.” In The Routledge Research Companion to Digital Medieval Literature, edited by Jennifer E. Boyle and Helen Burgess, 131–44. Abingdon: Routledge, 2018. Witmor, Michael and Jonathan Hope. “Books in Space: Adjacency, EEBO-TCP, and Early Modern Dramatists.” In Early Modern Studies after the Digital Turn, edited by Estill et al., 9–34. Tempe: Arizona Center for Medieval and Renaissance Studies, 2016.

Chapter 4

TEACHING CONSTANTINOPLE AS A (PIXELATED) PALIMPSEST J. W. TORGERSON

The study of

the Middle Ages remains unapologetically materialist. It certainly seems that medievalists, more than any other historical field, will profess that the unique tactility of mesmerizing objects and the impossible dynamics of indescribable spaces drew them to their period of study. In typological fulfillment of Thomas’ doubt, we fully believe our texts must be touched to be truly perceived. And yet among humanists it is also medievalists who have been the “early adopters of the digital, and continue to play an important role in the development of a broader field, which came to be called digital humanities.”1 We embrace our manuscripts via codices and in pixelated UHD. This curious but productive tension is, surprisingly, often least manifest where it might seem most relevant: the university classroom. For many medievalists standard teaching practices are still very literally by the book as most classrooms are no more digital-born than crisp lecture handouts, authentic audio recordings, and uncluttered PowerPoint presentations (Zoomification notwithstanding). Our historical, field-wide adoption of new research technai has pedagogical progeny as yet unborn. But why? Medievalists’ long familiarity with computer-based practices means that the field is self-aware of how habits of study determine habits of thought. Conditioned perhaps by Roberto Busa’s mid-century text analyses, the field enthusiastically engages in the current debates over machine learning and distance reading.2 In just a few short years medievalists have transitioned from trekking across Europe and ordering microfiche to accepting that we will read most of our documents digitized and online. We embrace the fact that our research practices must and will adapt, and we also know that this is a good thing. The technologies into which we invest ourselves fashion what and how we think: they structure our communications, govern our ability to organize knowledge, and determine how we re-produce or re-present past worlds. We know this not only from ourselves, for our propensity to innovation even inheres in the materials we study. Both Eusebius and Accursius re-conceived the graphic possibilities of the technology of the codex with the fourth-century Chronicon, and then 1  Birnbaum, Bonde, and Kestemont, “The Digital Middle Ages,” S2.

2  See the entire journal issue in which the Birnbaum, Bonde, and Kestemont article appears. On these specific topics see therein: De Gussem, “Bernard of Clairvaux and Nicholas of Montiéramey”; Romanov, “Algorithmic Analysis of Medieval Arabic Biographical Collections”; and Cruse, “A Quantitative Analysis of Toponyms.” Jesse W. Torgerson, College of Letters, Wesleyan University.

78

J. W. Torgerson

the thirteenth-century Glossa ordinaria, while the monks of St. Martin of Tours and St. John in Stoudios reconceived the very appearance of letters with their eighth- and ninthcentury standardizations and disseminations of “miniscule” scripts. Studying these materials has taught us that ideas are grounded in the limits of the forms of their representation, and that changes to those forms can present anew a past that has grown too familiar. We know that in insisting our research methods and our publishing outputs remain persistently open to the influences of developing technologies we keep open the possibility of continued insights. And yet. And yet, we have not habituated our most important re-presentations of the past into this dynamism: our classrooms are technologically stagnant. It is time we invested our pedagogies with the robes of creativity and the rings of innovation with which we approach our own research. It is time we re-programmed how we teach new minds to inhabit the medieval. The next computing revolution in study of the Middle Ages is upon us: it is time to bring the digital turn to pedagogy. I believe we can take this turn simply by claiming the technologies we have been given for our own ends, by bringing the technological structures of our communication and education into better alignment with the way we actually want to think and work and share. The tools are in front of us and if we are the ones to pick them up, we can decide how they are used. The following narrative is an impetus to do just that. I describe my mediocre, still-in-process online teaching encyclopedia project in a way that I hope will inspire the scholar willing to consider such a stance, the scholar willing to reconsider whether the technologies we have been given, with which we were taught, are the best tools for what we hope the next generations of scholars in our fields will do. Over the last five years I have employed digital technologies to turn students’ work in three iterations of a course in my core subject—the medieval Roman Empire of Byzantium—into a collaborative teaching resource. By approaching instruction as collaboration, I was able to develop a process for students to use their own learning to create material that would teach others. Embracing the digital gave me the means to curate this material into an interactive, topographically-organized, pixelated collection of medieval Constantinopolitan paraphernalia—what I have come to call the Constantinople as Palimpsest project.3 When I initiated the project five years ago, I modestly expected and planned that I would develop a teaching resource for myself, which I would perhaps share with select others. What I didn’t expect to create was a resource which had equal potential for public outreach, which facilitated new interdisciplinary connections, which prompted me to re-conceive the very structures of my knowledge of the past.

How Did the Project Come about (and What Is It)?

The project began as an exploration into how students in a survey course might learn the material realities of the capital of the medieval Roman Empire through their own 3  Constantinople as Palimpsest, https://arcg.is/01GXyj. For previous write-ups, see Jesse W. Torgerson, “Constantinople as Palimpsest,” “Constantinople as Palimpsest: Now Live,” and, “Constantinople as Palimpsest: Now With Words!” on the Wesleyan University Traveler’s Lab.



Teaching Constantinople as a (Pixelated) Palimpsest

79

research rather than by reproducing truncated versions of my lectures and our class discussions as papers and exams. In Spring 2015 I designed a rudimentary course unit that required students to perform library-based research on items found in Constantinople, from the textually-attested statues in the Baths of Zeuxippus to the still-standing Golden Gate. Once their research was complete, students then uploaded their miniature research projects onto ArcGIS online (https://www.arcgis.com/home/index.html). Over the course of multiple, gradually improving iterations of this course project I was able, with the help of students and colleagues, to curate and organize these research projects as a cumulative collaborative database via StoryMaps (https://storymaps.arcgis.com/), an associated component of the Esri mapping package. Constantinople as Palimpsest, the title we eventually gave to this project, is a simile analogizing our online map with both the medieval process of creating a palimpsest— scraping off the ink of a previous text and over-writing another—and the modern process of using ultraviolet or infrared light to recover those earlier texts without removing the latter. As a metaphor, palimpsest “lends itself especially well to the interpretation of architectural monuments and landscape sites” because it encourages us to always be mindful of changes in appearance and usage over time.4 Analogous to a palimpsest, our project preserves the long process by which any built environment breaks down or buries earlier structures: not only as Istanbul did to Constantinople, but as late Byzantine Constantinople did to earlier Byzantine Constantinople, and as Constantinople did to Byzantion. The project works to not only study Constantinople as a palimpsest but is also a literal and metaphorical palimpsest itself. As described below, the mapping software we used—Esri’s ArcGIS online—envisions items and geo-referenced images as layers added onto a base map, just as new hands can gloss or overwrite a base text. Furthermore, even in its earlies stages our project presented multiple overlapping and contradictory stories and eras within a single object. Like a palimpsest, successive student contributors invested new realities into a single space shared between times.5 The StoryMaps platform allowed us to use the digital technology of ArcGIS layers to present multiple past realities at once, with separate views of the city placed one on top of the other. The base is a digital projection of the contemporary world. Onto this base the first layer added is a schematized, geo-referenced map of medieval Constantinople including the outlines of walls, harbours, hypothesized major roads, as well as Hagia Sophia and the Hippodrome for reference. The map was drawn by Alp Eren as part of his work in Wesleyan University’s Traveler’s Lab.6 In their work for my semester-long Byzantine history course students used ArcGIS online to add interactive content as digital layers added on top of this first layer. These additional layers were drawn polygons, 4  Palimpsests: Buildings, Sites, Time, edited by Aksamija, Maines, and Wagoner.

5  Neither the alpha (Spring 2017) nor the beta (Spring 2020) versions of our place-based encyclopedia are able to take time into account: they visually display every item as though equally present at the same time in the millennium-long history of the Byzantine Empire. While we have embedded this chronological information in the text of each entry, enabling time-sensitive visualizations remains a goal for future versions of the project. 6  Traveler’s Lab: https://travelerslab.research.wesleyan.edu.

80

J. W. Torgerson

lines, or pins which ArcGIS online calls map notes. These map notes contain text and images which, when added to a map as layers appear as items on the map: since they are geo-referenced, or digitally linked by the latitude and longitude at which they are placed, base map and layers are thus exactly coordinated with each other. In other words (to return to the pal­­ imp­sest metaphor), if one were to scrape off these layers, as well as Eren’s over-laid historical map, the exact corresponding location in modern Istanbul would appear. Figure 4.1 illustrates the concept with Figure 4.1: Constantinople as Palimpsest concept, with map for the topic the map for Monumental Architecture, Monumental Architecture. one of eight topical maps in the project. The ArcGIS online platform that hosts this map allows it to remain dynamic. Both the first layer—which functions above as the base map—and added layers of map notes with their associated text and image content are all interactive. Users can navigate the map by dragging or zooming in and out and access the text content of the overlaid map notes by selecting any of the green lines, points, or polygons. Thus, when any one is selected, a window opens to display an encyclopedic write-up by an undergraduate student with the historical images and reconstructions they selected, all linked to the appropriate place in Constantinople. The potential of this project as an attractive repository for sharing student work is obvious: students are drawn to the invitation to create something for their coursework that has lasting, public value. However, compiling and archiving this content as a palimpsestic-accumulation of ArcGIS layers also means all of the text content added to the project (the map notes) is now indexed by place. That new categorization has pedagogical value in displaying known information in a more accessible manner. In addition, it also has the potential to provide new scholarly insights by offering an entirely new way of connecting and associating our bits of historical knowledge. As I will describe in later sections of this article, the project’s framework not only offers a new means of sharing knowledge between specialists and from specialists to the public, but also by sorting known information in a new way contains the potential for new knowledge: not through discovery but through new connections and associations. Before turning to that potential, however, I first will describe how this project came into being. Slowly

I never sat down and designed Constantinople as Palimpsest. It came to be through an unplanned, heuristic process of experimentation with project-based teaching, beginning in 2015 as a shared folder on ArcGIS.com called “COL 128 – Constantinople.” Students



Teaching Constantinople as a (Pixelated) Palimpsest

81

Figure 4.2: Constantinople as Palimpsest, ca. 2015.

in that course uploaded their course work (micro-research projects on sites and objects) to the folder to display the work to the rest of the class. I had students put these research projects into the pixelated topography of an online map—the layers and notes described above—because I wanted a part of their research to be identifying where in Constantinople each place or item was. I required this for no more profound reason than I was trained that provenance matters. I wanted to make sure my students also learned to value that idea by being required to plot out on a map where they believed an item should be located and what scholarship backed up that assertion. The screenshot above gives a sense of what this initial 2015 version of the project looked like. This is not to criticize student work (I myself tried to draw a freehand hippodrome and did no better), but to demonstrate the rudimentary and unprofessional way this project began. I came to understand that the bar to start a “digital humanities project” is literally whatever your abilities are, right now. It is easy, in conceiving of digital projects, for the aesthetics to become the starting point and so completely overwhelm any idea. Instead, I encourage my colleagues to worry about what we are already used to concerning ourselves with: the content, quality, and organization of the information. The final aesthetics of digital design cannot be anticipated but, in my experience, are something that is incorporated when the time is right. What is essential is to decide what information will be included, and how that information will be organized. Thus it was that in the summer after this initial haphazard and exploratory venture into project-based learning, it occurred to me that beneath these poor drawings I already had the makings of an interactive digital topography of Byzantine Constantinople. The shared class folder was not merely proof of work done (which is all that I initially thought I would achieve) but was also a real collection of historical information. If I spent the time to edit the students’ work and consolidate it onto a single map I could teach the next instantiation of my course from this material. Then, in the process of curating this accretion of historical knowledge I realized I was working with a palimpsest. And so,

82

J. W. Torgerson

with the addition of a name, a rudimentary course unit turned into a Digital Humanities Project. I decided to focus the next offering of my Byzantine history survey (in Spring 2017) on the city of Constantinople, and to make the entire goal of the course to turn the shared ArcGIS.com folder into a collaborative, interactive online map of the city. With Help, and the Right Timing

It is only in retrospect that I can state the Constantinople as Palimpsest project evolved in response to the desires and ideas of students enrolled in three classes of between ten and eighteen Wesleyan undergraduate students as they generated content over the course of the Spring 2015, Spring 2017, and Spring 2020 semesters. To be clear, I did not turn the student coursework into this resource on my own, and I have no interest in giving the impression I did. The project is a true team effort. After the initial exploratory offering (Spring 2015), the subsequent versions of the course (Spring 2017, and Spring 2020) each had two groups of students who assisted with the project. Even the Spring 2015 offering was only possible through generous gifts of time, expertise, and training from my Wesleyan University colleagues Dr. Jason Simms and Prof. Kim Diver. Without my student gis intern, Marjahn Finlayson, assisting students in the course with technical difficulties in using ArcGIS.com the course project would have been impossible. When my colleague Prof. Lisa Dierker won (and shared!) a Davis Foundation grant that year to support project-based learning, I was able to hire the student research assistants to help me organize and catalog the work the students had produced. Turning the initial course into an ongoing project with a life of its own was invigorating, exhausting, and the result of many willing hands and minds. For subsequent semesters I was determined to transform the idea from a time-consuming experiment in project-based learning into a sustainable ongoing project around which I would design future courses. To do so I built my pedagogy around institutional resources I thought would be relatively permanent, and which I thus could take for granted. First, I had a Course Assistant (Connor Cobb in 2017, and then Nathan Krieger in 2020). These were history majors who earned course credit for attending the class, serving as an initial research resource to help students study and write up their projects, and at the end of the course working with me to index and edit all of the research submitted as map notes. I also had a gis assistant for each year (Nadja Shannon-Dabek in 2017, and then Alp Eren in 2020) each trained in ArcGIS by Prof. Kim Diver. These gis assistants worked with students on uploading their projects as layers to ArcGIS.com, and then for the most recent beta version of the project, Mr. Eren re-drew the historical base layer we had used previously, which Konstantinos (Kostas) Plakidas had published to Wikimedia commons from a synthesis of Raymond Janin’s Constantinople Byzantine.7 Mr. Eren’s new drawing meant that the image of Constantinople’s shoreline and schematized roads was no longer an image only fairly-accurately super-imposed on a map, but is its own geo-encoded shapefile, which can thus be shared for similar uses. 7  https://commons.wikimedia.org/wiki/File:Byzantine_Constantinople-en.png; Janin, Constantinople—Byzantine.



Teaching Constantinople as a (Pixelated) Palimpsest

83

Additionally, I had the unique resource of the Traveler’s Lab, a collaborative multicampus research network run by myself, Gary Shaw of Wesleyan University, and Adam Franklin-Lyons of Emerson College. Each year students from various majors collaborate with us to work—depending on resources and interest, either voluntarily, for credit, or as paid Research Assistants—on ongoing research projects using a combination of methods (both analogue and computer-based). Through the Traveler’s Lab some of these same students (Connor Cobb, Nathan Krieger, Alp Eren) as well as others (Caroline Diemer, Jesse Simpson, Jonah Skolnik) contributed additional work beyond the semesters when I was actually teaching the course. These lab students helped me curate, organize, and standardize the research done by the students in the class, to prepare resources for student research projects, and to test run various course materials. I must emphasize the significance of these individuals’ contributions to Constantinople as Palimpsest. In theory, the project might have taken place without them, but in reality, it never would have done so. If I myself had logged the thousands of hours of training and support these students and colleagues collectively invested in the project, it would have come at the direct expense of the sort of research outputs recognized by the structures of academia as legitimate for advancement: as we all know, an exploratory project such as Constantinople as Palimpsest does not fit into traditional categories for tenure and promotion. It is worth considering how often we, as scholars, are entirely responsible for inhibiting innovation by insisting that the forms of scholarship that happen to be in place, in which we published, remain necessary.8 In my case, the risks of investing in this project were possible to sustain for two reasons. First, given the assistance I received (and only because of that assistance) I was able to restrict the amount of time I spent on the project to essentially what I would have spent teaching a traditional version of a historical survey. The time spent researching, writing lectures, and grading traditional exams or papers was roughly equivalent to that spent providing students with their own miniature research projects and assessing, uploading, and curating those projects on the website. Second, I was correctly advised to begin the experiment of this project-based course at the very outset of my time in a tenure-track position. Some studies indicate that when students are given the opportunity to be properly self-reflective they will identify that they learned more in a projectbased learning environment.9 Conversely, other research and widely-shared anecdotal evidence indicates that without extensive coaching of student expectations, projectbased learning can be perceived negatively in evaluations of teaching.10 Since I engaged in experimental project-based learning from my very first year in a long-term contract I was able to sustain critical student feedback and confusion about the course by improving the class each time I taught it, throughout the duration of my contract. I found three lessons here. In designing my long-term project, it was essential to identify the specific resources at my own institution and to create the project out of 8  Kleinberg, “4. The Analog Ceiling,” in his Haunting History, 115–33.

9  Clark, “Project Based Learning”; Mou, “Students’ Evaluation of Their Experiences.”

10  Shih and Tsai, “Students’ Perception of a Flipped Classroom Approach”; Beckett, “Teacher and Student Evaluations of Project-Based Instruction,” 52.

84

J. W. Torgerson

those resources rather than trying to replicate what others had done elsewhere. Second, it was essential to think strategically about timing, both professionally and personally. My idea for a project needed to be developed slowly and heuristically and through accretion and experimentation. I needed to plan time for the ideas to sit and percolate. Finally, the previous point is why, in my experience, many brilliant and completely viable ideas never see the light of a classroom for want of job security and institutional support. Brilliant ideas can wait to be realized as successful projects, but they cannot wait forever. By Rethinking the Nature of Student Coursework

Once I decided to use student coursework to create something of lasting pedagogical value, I was immediately confronted with the challenge of getting students to produce quality work. Students will produce work that reflects not only their knowledge and abilities, but their own expectations for a grade. I had to get students’ personal standards and expectations for their work to meet my expectations and standards for a public-facing resource. Thus, after the project’s initial run in Spring 2015 I set out to design assignments in ways that student submissions would be of high-enough quality that subsequent students could both learn content from them and imiate them as models. I had no idea how to do this. Approximately half of the assignments that students submitted in Spring 2015 were unusable as learning materials for the next version. Here I was once again assisted by institutional investment in pedagogy. I needed help thinking through a new pedagogical approach on the conceptual level, and this aid was provided by my institution through the persons of Paula Blue and Dan Mercier at Wesleyan’s Center for Pedagogical Innovation. Thanks to discussions with Paula, Dan, and Lisa Dierker (mentioned above), three changes made the difference. First, in the 2015 instantiation of the course, I was interested in having students approach historical study differently, considering, among other things, past place rather than past time or event. We read and discussed selections from Walter Benjamin and Michel de Certeau which gave the students models for how to conceive of the spectral necropolitics of the city, and of the creative tactical resistance of urban everyday life.11 We read Poe’s “The Man of the Crowd” and theorized the way in which our historical gaze would, for both good and ill, resemble that of a nineteenth-century flâneur.12 We asked: could we do better? Fantastic discussions and exercises notwithstanding, this effort to apply reflections on the nature of “city” developed in modern Paris and New York did not help my students produce more focused and accurate research on Constantinople. Therefore, in subsequent offerings I scaled back our work in criticism and instead limited theory to articles produced by Byzantinists on Constantinople itself.13 11  Benjamin, The Arcades Project; de Certeau, The Practice of Everyday Life; Bavidge, Theorists of the City.

12  Poe, “The Man of the Crowd,” in his Tales (1845), 219–28. Shaya, “The Flâneur, the Badaud, and the Making of a Mass Public in France.”

13  Accessible, discussion-provoking chapters include Cyril Mango, “The Disappearance and Revival of Cities,” in his Byzantium: The Empire of New Rome, 60–87; Judith Herrin, “Constantinople,



Teaching Constantinople as a (Pixelated) Palimpsest

85

Second, in the initial course offering I allowed students free rein over the topics they were interested in researching. This led to some delightful, singular experiences—such as when we collectively debated how to “map” a processional route—but in general it incited students to stretch themselves too far. In the subsequent offerings, students identified thematic interests and I created small groups from these themes. This allowed me and my course assistants to issue dossiers of materials and bibliography to groups rather than to each student individually. Students were also able to take on slightly larger projects by collaborating for a week or two. Finally, students naturally increased their sophistication as the weeks progressed and their attention focused on single topics such as defensive structures or economic exchange, for example. Their later research projects were thus informed by previous work and previous work was then revised in light of later learning. All of this substantially improved the quality of what students submitted. Third, I developed increasingly exacting rubrics for grading, balanced by an invitation to revise as many times as desired. When the course was taught in 2017 we developed a ten-page guide on creating and uploading items to the project platform on ArcGIS online. As student submissions came in, we honed the grading rubric until it communicated exactly what I wanted from research submissions. The eventual guidelines identified all of the areas where students tended to be less exacting. I assigned percentages to each of these areas so that students’ submissions were harshly docked for neglecting essential items: students dropped an entire letter grade for minor infractions such as forgetting to put their name on the submission, forgetting to provide a working link to an image, or not formatting their text correctly. With this rubric, students’ initial submissions often received grades as low as twenty-five to fifty percent. I allowed students to re-submit as many times as they wanted, however, challenging them to get every one of their submissions up to at least ninety-five percent. This allowed our interests to align. Students responded well to the real opportunity to earn full marks in the course, and I enjoyed seeing students master the material as they continued to revise and learn the importance of form and citation in scholarship. The project as a whole was only possible because publication to the internet no longer depended on my re-writing, editing, and adding content to the vast majority of student submissions. The project maintained its pedagogical value beyond the mere confines of a single course exercise due to the crucial adjustments we made along the way.

Achieving the Goal: Pedagogical Value

The initial goal of the Constantinople as Palimpsest project, as the narrative above explains, was to present students of my Byzantine history survey with a textbook whose interactivity facilitated learning about the city of Constantinople in a manner such that it took shape in their imaginations. Although the project will continue to grow, it has already achieved this initial goal. I will use Constantinople as Palimpsest for my future university classes on Byzantium, I invite others to do so as well, and hope they will conthe Largest City in Christendom,” in her Byzantium: The Surprising Life of a Medieval Empire, 12–21; and Robert Ousterhout, “Constantinople and the Construction of a Medieval Urban Identity.”

86

J. W. Torgerson

tribute assignments as additions to the content. Furthermore, I hope to introduce secondary school teachers to the resource for units touching on medieval history. To this end, in this section I will lay out how our project may be read and studied. The three key aspects to the online map are the city base layer, the thematic “tabs,” and the pop-up windows where much of the student-generated material resides. Once a user has chosen a thematic tab, the historical content is revealed by exploring the topography of the city: scrolling and clicking on any pin, line, or polygon in the map (green) activates a “popup” dialogue box. Each dialogue box contains a brief entry on the item (whether object, monument, site, region, or route) which includes a short definition of the item, its date, some commentary, further bibliography, and image citations. The content in these boxes represents a student’s self-designed research project, generated as their coursework for a single week of the semester. Each item’s pop-up box consists entirely of the student research with only corrective editing. Thus, these are fully attributed to those students as authors. These are the surviving traces of Constantinople, as “unearthed” by students. Each item is categorized according to our unique collaboratively-designed system. Items are noted by type, by the nature of their survival into the present, by name, and by date of appearance in the city. We identified four item “type” categories so that users will know what they have chosen since it is impossible to provide scale drawings for each item. The categories are the following: • • • • •

Region—an area: neighbourhoods, regions, etc. Route—a way: roads, processional routes, etc. Site—built structure: forums, harbours, palaces, walls, cisterns, etc. Monument—immovable item: obelisks, columns, mosaics, etc. Object—moveable item: statues, lamps, hairpins, etc.

We also created three coded categories to designate the status of each item’s survival into the present. These are: • IP—In Place: intact/visible; object never been moved • DP—Displaced: survives but has been moved from its medieval location • TA—Textually Attested: completely gone, but texts attest its existence.

Finally, students identified a date—whether a specific year or a century—when the item in question appeared in Constantinople. The Serpent Column of Delphi, or the Obelisk of Theodosios, for instance, are much older than the city but are given dates according to when they arrived there. An example here may be in order. The label for Constantine’s column is the following: Monument IP_Column of Constantine_330. The codes indicate that this is a “monument” (according to our categories); it is still “in place” today; it is commonly known as the “Column of Constantine”; and, it is believed to have been constructed in ad 330. Below this label the pop-up window contains a description and images of the site or item where possible or relevant. The item is briefly defined and dated, a fuller commentary follows, and a bibliography of relevant images and sources is appended. After the first two iterations of the course, the students had generated over three hundred distinct items. We therefore needed to redesign how our materials were pre-



Teaching Constantinople as a (Pixelated) Palimpsest

87

sented, since a single map populated with hundreds of nondescript green points was no longer legible to our users. The knowledge we accumulated needed to be curated. At this point in the project I realized this was no longer simply a way of housing student projects but rather a new means of knowledge curation that needed a logic different than that of the encyclopedia, which had governed my organizational strategy until that time. My Course Assistants and I talked through our material. The goal was to make it possible for students to explore the current historical image of Constantinople as it lives in the minds of Byzantinists. The traditional item-by-item heading-based encyclopedia is not a helpful introductory tool since even digital versions such as Wikipedia require the reader to already know something—often quite a bit—about what they are looking for. Furthermore, even when advanced scholars use traditional encyclopedias, they tread the same trails again and again, looking up entries on subjects or items they wish to know rather than reading ecumenically for discovery. A map, by contrast, is an easier way to orient oneself (quite literally) to an unfamiliar field of study since they are a dynamic means to re-conceptualize information already known via other formats. But our dynamic map needed to give first-time readers some orientation. My Course Assistants, gis Assistants, Lab Students, and I had numerous discussions—both theoretical and practical—over this issue. The StoryMaps platform (provided by Esri and fully compatible with their ArcGIS online software) made it possible for us to present multiple curated maps for readers or users to explore. We finally decided upon the now-obvious idea of creating multiple maps whose contents would be defined by overlapping themes. Adopting an approach using overlapping themes was an important step: we wanted a compromise between our contemporary need to have some help in understanding, and the past reality which we were trying to enable users to re-imagine. We did not want to lose the sense of a real, lived city which could only truly be known by living through the uninhibited cacophony of quotidian life. We tried to make our themes reflect what students had chosen to work on, rather than import established scholarly topics. We settled upon Monumental Architecture, Water Infrastructure, Exchange Economy, Religious Life, Private Life, and Administrative Regions as ways to bring our readers gently into the life of the City. It became clear to me that what would perform this function for my students would do the same for both the general adult public and students at the secondary level whose teachers might introduce a unit on a particular medieval empire. Constantinople as Palimpsest was suddenly not just a course project, but a way to present what interdisciplinary scholars of Byzantine Studies have uncovered about the medieval life of the city of Constantinople (or Byzantium) in an interactive, open, public-facing format that can serve a range of readers equally well. Persons with no knowledge of Byzantium at all can “walk” the imagined medieval city by working through our curated, manipulable maps. Experts will explore what they already know in a new form. Using location as the organizing construct for artefacts (rather than initial letter, discipline, or historical period) opens up a field in which new associations and ideas can germinate. As we continue to populate the map, historians, scholars of literature, art historians, and archaeologists will be reminded of neglected or underappreciated material, and start to make new connections like those described in the next section.

88

J. W. Torgerson

Beyond Expectations: from Teaching Experiment to Historical Source What I have found with this project is that re-integrating abstracted texts or artefacts with their provenance changes the vantage point for scholars at all levels of expertise— from students to endowed chairs—as well as when and how we recall what we have learned. I know this because of how Constantinople as Palimpsest has changed my own mental landscape. I have always been an avid fanatic of imagined worlds depicted on maps—whether fantastical or historical. But even after a decade studying Byzantium, even after spending time in the streets of Istanbul, I was still not able to hold medieval Constantinople in my imagination. Even the truly incredible three-dimensional design work of the “Byzantium 1200” project did not quite suffice.14 Instead, it was through designing a platform in which my students deposited their research on Constantinople that I was finally able to visualize, in my own mind, the historical world I spend my days studying. The realization of an imagined landscape has made my engagement with the computer-based possibilities of organizing, storing, visualizing, and sharing our knowledge of the past worthwhile for me both as a teacher and scholar. In historical studies, the associations made by our imaginations often determine the limits of our ideas and insights. When we first began the project, I gave the students who worked on Constantinople as Palimpsest my abstract ideas about the items and places in Constantinople as I had come to know about them through the course of my own education from the work of Raymond Janin, Wolfgang Müller-Wiener, Averil Cameron, Judith Herrin, and others.15 Week after week they brought this knowledge back to me sometimes with content I did not know, but always in forms I did not recognize. Not only were the students re-placing what I knew about Constantinople into its topographical context, they were adding to that knowledge through new bibliographies and new readings. This multi-layered reorientation made me quite literally re-visualize everything I knew about Constantinople and in the process created an entirely new network of knowledge that we now shared. I stopped remembering the hostel cum hospital of the Ξενὼ�ν τοῦ Κρά� λη (Xenon of the Kral) as an entry in the Oxford Dictionary of Byzantium that I had once found while double-checking what I knew about Stefan II Milutin’s patronage but re-learned it, right along with my students, as a philanthropic attachment to the monastery of St. John in Petra (more on this association below).16 Moreover, the last meeting of each week was devoted to rapid-fire presentations, where each student presented that week’s micro research project to the group for peer evaluation and collective learning. These presentations soon went from terrifying spotchecks to miniature workshops, as students became comfortable with the platform and their research methods. I began to look forward to these events with expectations simi14  https://www.byzantium1200.com/; https://www.youtube.com/user/Byzantium1200; https://twitter.com/Byzantium1200.

15  Janin, Constantinople—Byzantine; Müller-Wiener, Bildlexikon zur Topographie Istanbuls; Cameron and Herrin, Constantinople in the Early Eighth Century.

16  Talbot, “Xenon of the Kral,” entry in vol. 3 of The Oxford Dictionary of Byzantium (available online behind paywall since 2005).



Teaching Constantinople as a (Pixelated) Palimpsest

89

lar to what I bring to a promising conference panel: I expected to learn something new (even about things I thought I knew well), and especially to make connections I had never before imagined. As these experiences accumulated, I realized what made this was possible and the implications for the project’s future. Before I discuss this value as a tool for discovery, I will illustrate the sorts of insights the project made possible by following a serendipitous series of connections made in the Spring 2017 iteration of the class.17 One week, Chris Wyckoff, a junior in the course, decided to map all of the cisterns noted in the important work by Crow and his colleagues in The Water Supply of Byzantine Constantinople. Mr. Wyckoff’s group had already studied and mapped the famous cisterns, including the underground Basilica Cistern and the open-air cisterns of Aspar, Mocios, and Aetius.18 Though I thought this effort would presumably produce useful information, I was unsure of how it would integrate with the project, and how we would use it: I had always thought about the water supply of Byzantium in terms of either feats of engineering, or of the city’s ability to sustain a population and withstand a siege. At the same time as Mr. Wyckoff was working on cisterns, another group of students (Zheng Mao, Nadja Nurjadin, and Melissa Thornton) were studying places of residence in the city. They were looking for ways to locate and learn about middling or lower-nobility residences, and more specifically anything more than the few known aristocratic nonimperial palaces, such as the palace of Anicia Juliana.19 As chance would have it, the group working on residences and the group working on the city’s water infrastructure presented back to back: the comprehensive map of cisterns followed discussion of the difficulties of finding examples and patterns of elite residence. The connection was suddenly clear: while fountains were markers of population concentration, small cisterns were signs of elite residence and clusters of cisterns indications of an elite neighbourhood. In the ensuing discussion two new hypotheses emerged concerning residential patterns in the city of Constantinople that I had never before considered. First, the standard narrative of Constantinople is that imperial residence migrates from the Great Palace to the Blachernae region.20 Yet I had never thought through what this meant for elites who wanted proximity to the imperial court. An initial answer arises from the six significant clusters of cisterns that appear immediately as they are plotted on the map. The ancient and early Byzantine elite residences likely account for the largest of these groupings: on the acropolis point; near the Hippodrome; between the Forum of Constantine and 17  It should be noted that the Spring of 2020 class did not have these experiences, one of the billions of small tragedies brought about by the COVID-19 pandemic. By the time we arrived at the project-based learning component of the course that semester, everyone was at home doing the best they could with “remote learning.” Students did exceptional work on their projects, but did their work at their own pace and I was the only person who actually studied the growing collective project, and only then outside of the context of the discussions and rejoinders that made connections in previous semesters happen so continually and naturally. 18  Crow et al., The Water Supply of Byzantine Constantinople.

19  Harrison, A Temple for Byzantium.

20  Ousterhout, “Constantinople and the Construction of a Medieval Urban Identity.”

90

J. W. Torgerson

Figure 4.3: Clustered cisterns near the Imperial Court.

Figure 4.4: Clustered cisterns in the regions of Blachernae and Petra.

Figure 4.5: Cistern clusters around the monastery of Constantine Lips.

the Neorion and Prosphorion harbours (but within the Severan walls); and, around the Forum of Theodosius, as below, in figure 4.3. A more sporadic cluster in the regions of Blachernae and Petra, as in Figure 4.4, likely indicated elite residences built near Blachernae Palace during the Palaiologan period. Though it is interesting to see this mapped out, I did not find it particularly unusual. What took me completely by surprise were the obvious clusters around the monastery of Constantine Lips (restored ca. 908), as seen in Figure 4.5. If Constantine Lips’ tenthcentury patronage is an indicator of a larger pattern—the emperor Basil I (r. 867–886) also famously restored the nearby Constantinian Church of the Holy Apostles—then the Macedonian emperors (867–1057) and the elites of their era would seem to have viewed this region as a new zone for not only patronage but residence. Furthermore, the final cluster of cisterns located around the Komnenian dynastic church of Christ Pantokrator would seem to similarly indicate elite residences built (or



Teaching Constantinople as a (Pixelated) Palimpsest

91

renovated) in connection with imperial patronage of the monasteries and churches in this Platea region, just inside the old Constantinian walls.21 Seeing cisterns as traces of resident wealth suggested, immediately, that elites in Constantinople generally moved in conjunction with the more famous trajectory of imperial palaces from the ancient acropolis point during the Early Byzantine period, to the Platea region on the coast of the Golden Horn in the Macedonian and Komnenian periods, to the Blachernae Palace region in the late Komnenian and Palaiologan periods. The work of transferring already-known information to a digital-map format generated socio-economic questions which I would not have initially looked to maps to answer. Did elite residences cluster near imperial palace regions because they were zones of patronage and influence? Can this hypothesis explain the cluster of cisterns around the Pantokrator monastery (Zeyrek Camii) during the Komnenian period,22 parallel to John Tzetzes’ later patronage around the Pammakaristos church during the Palaiologan?23 Up until this experience of being taught in my own classroom, I had thought of Constantinople’s history in terms of a slow migration of elite residence from the ancient centre of the acropolis point, northwards and westwards along the Golden Horn to the Blachernae region. Instead of a story of imperial decline (emperors could no longer maintain the massive Great Palace complex on the Easternmost point of the city), a simple map of known cisterns and a series of interdisciplinary student questions allowed me to hypothesize a more engaging narrative of shifting zones of patronage. Further research may even flip the old narrative around: perhaps instead of assuming imperial patronage drew aristocratic, we should be thinking of imperial residences as following and responding to slowly shifting zones of wider trends in elite patronage. This is not, of course, the conclusion to an extensive research project, but merely an opening question. Nevertheless, finding the sparks of ideas which motivate scholars to attend conference panels in the undergraduate classroom demonstrates just how revolutionary opening ourselves to new pedagogical technologies could be: not just for making teaching more invigorating, but for the formation of new minds just beginning their work in the field. A second, more succinct example: another student group from the same Spring 2017 class took up the topic of societal infrastructure and considered specific locales like xenodocheia, hospitals, baths, and gardens. One week one of these students, Silin Chen, chose to research the Xenon of the Kral in the Petra region, and in doing so read about the famous Vienna Dioscurides manuscript associated with it. She asked to pursue that the following week, though it was a slight tangent from the group’s focus.24 When Ms. Chen included in her report that the Vienna Dioscurides had been held for a time at the Xenon of the Kral and at that time famously rebound by John Chortasmenos, the 21  Congdon, “Imperial Commemoration and Ritual”; Kotzabassi, The Pantokrator Monastery in Constantinople. 22  Ousterhout, “Architecture, Art and Komnenian Patronage.”

23  Magdalino, “The Maritime Neighborhoods of Constantinople” and his “Neighbourhoods in Byzantine Constantinople.”

24  Kiilerich, “The Image of Anicia Juliana in the Vienna Dioscurides”; Brubaker, “The Vienna Dioskorides and Anicia Juliana.”

92

J. W. Torgerson

provenance of this most famous herbal took on a new and more urgent significance. I suddenly remembered a conference I attended in which Byzantine medicine and codicology specialists argued whether, on a theoretical level, the manuscript would or could have been used “in the field.” Would, or could its famous accretion of marginal notations have been made in the context of use in actual fields and gardens to identify herbs, or to test and annotate the documented knowledge? The terms of this debate—at the time I witnessed it—concerned the size of the codex and whether one could viably carry it about. What was not debated was its location in Constantinople, for it was assumed implicitly—even by specialists trained to account for provenance—that the manuscript belonged in some kind of palatial library, in a truly urban environment. When I summarized this debate for the students, Ms. Chen insisted—correctly!— that the debate should have begun with the evidence that the herbal was actually kept in the xenon or hospice–hospital attached to the monastery of St. John in Petra. Relying on their own knowledge of mapped locales within the city, other students pointed out that this was near the cistern of Aetius and the late Byzantine inner-mural agricultural zone between the Constantinian and Theodosian walls.25 With this provenance and topography firmly fixed in their minds, it became impossible for the class to view the Vienna Dioscurides as anything but the most practical of reference handbooks for a suburban hospital attached to a monastic intellectual centre, located in a region replete with gardens and farms. Once again, I was shocked that I had never before considered this while thinking about use, for of course it should be essential to think not only about the adjacent volumes on the bookshelf, but about where the library itself was located. I had never even entertained the idea that John Chortasmenos’ rebinding was done to preserve the manuscript for continued use in the Xenon of the Kral. Now this was a real possibility. Undergraduate students who two months prior did not know Constantine I from Justinian I provided me with valid, sophisticated, and practical insight into a topic I never even would have mentioned in a lecture format due to its obscurity and my presumption of specialist-only interest. The above anecdotes are not field-changing insights; indeed, if either of these topics were the focus of my own research, they may not even qualify as insights but rather quotidian passing ideas or wrong-headed from the start. Nevertheless, the experience of literally re-mapping things I already “knew” caused me to re-think what I actually knew about these items and also an entire network of associated facts. I mentally filed these and many other ideas away for later use, as the seeds for further thought and further research. These are the sorts of ideas and experiences for which I attend presentations and lectures by my peers, not what I expect to learn from my own introductory lecture course for undergraduate students completely new to the subject. I learned through this experience that the difference between teaching and learning is not in somehow finding a pocket of undergraduates who are already exceptional researchers. Enthusiastic and motivated students are of course important, and some prior experience doing library research helps, but the project-based learning environment especially incites and activates these characteristics. Students do good work when their peers are apparently 25  Constantinides, “Byzantine Gardens and Horticulture in the Late Byzantine Period.”



Teaching Constantinople as a (Pixelated) Palimpsest

93

invested and when their instructor is spontaneously and increasingly enthusiastic, and they then in turn become increasingly invested and curious. In my case students’ good work led not just to satisfactory assignments, but to insights. It did so because their work took field-wide knowledge and converted that knowledge into a different shape, form, and outlet. Doing so generated new possibilities for understanding, for neophytes and experts alike. Certainly, this project-based course is different from other lecture-based classes, where I have good experiences as well. Even re-formulating an old lecture can allow me to see my sources differently, and I can always hear texts and ideas anew from students’ perspectives in discussions. But outside of my courses devoted to the Constantinople as Palimpsest project, I have never had that flash of insight where two and two becomes seven: when one realizes how three seemingly different things are in fact closely related, and when put together not only explain each other better but point out something else entirely. The Constantinople as Palimpsest project and the digital methodologies upon which it is based have made it possible to think about what all of my students were working on all at once, in combination with everything I already knew. It provided me with a different way of visualizing what was already present in my mind. And this is the real significance of the project: it is a new way of visualizing historical knowledge.

Conclusions: Epistemological and Theoretical Value

Project-based thinking deployed in a historical course is not just a different way of running course assignments, but a revolution in the teaching and thus practice of history. Project-based work can overturn our presumed pedagogies by combining the process of learning accepted historical information with the process of creating new historical information. My students learned about what scholars know of Constantinople, but by connecting that knowledge to a historical provenance—a time and a place that they could actually visualize and so imagine—they were able to also generate new ideas on their own and collectively. By reflecting what I, as a scholar, already knew back to me but in a different form my students showed me what I knew in a new way. This raises the question: why do we—scholars—think about what we know in the way that we do? Could we do better? Imagine with me the technology we currently use and take for granted for the mass preservation of widely-accepted basic knowledge: the encyclopedia. Encyclopedias have changed greatly over time. Nevertheless, from the great Byzantine collection known as the Suda to the Encyclopedia Britannica to the Oxford Dictionary of Byzantium and the now-ubiquitous Wikipedia, this technology of ordering knowledge primarily serves those who already know what they are seeking. I know that if I want to learn where and how imperial power was exerted and communicated in Constantinople, I might start by looking up βασιλεύ� ς in any of these resources. But if I don’t know that, these encyclopedias will at best confuse and overwhelm, and at worst lead me entirely astray. The technology of the encyclopedia structures knowledge according to an index, whether the alphabetical listings of an encyclopedia-in-codex, or the hyperlinked searchable listings of an encyclopedia-online. To use either an alphabetically listed or open-searchable

94

J. W. Torgerson

encyclopedia well, foreknowledge is required. Encyclopedias are incapable of true paideia. The presumed technology of our reference works requires that we have already been instructed in the topics of our searches. As medievalists we know that it is possible to organize knowledge differently. In addition to the encyclopaediae we have the florilegium, the liber manualis, and the πρό� χειροι κανό� νες. But despite this variety, these are innovations within a specific technological limitation: these works are all tied to the arbitrary technology of the codex on which modern historical studies continue to be absolutely reliant. Is it in our best interest to be so dependent upon the arbitrary technologies that currently structure our teaching, our inter- and intra-field communications to limit both our public outreach and the very boundaries of our thought and research? We now can quite easily think beyond how the technology of the codex predetermines the ways we organize large and diverse collections of knowledge. If we can use computer-based forms to identify a different “compilatory technique,” if we can create different “knowledge-ordering works,” shouldn’t we?26 What if expert scholars and innocent students could deepen their knowledge with the exact same tool? I have used the term imagination through this essay intentionally. As scholars of the past, our imaginations play an essential role in every act of thinking that we do. Historians of all fields imagine the worlds we study. Any tool that fleshes out our imaginations promises to exponentially increase the number of things—and ways—we can see. Before my project-based classes on medieval Constantinople, I knew a lot about my subject: I already had taught four complete lecture courses on it. My students, however, taught me that despite how much I knew, I ultimately merely had an idiomatic narrative—an abstracted, arbitrary, and frankly haphazard framework for categorizing and ordering that knowledge. In this sense, the mass of my knowledge was worth very little because it was separated from its actual provenance. Through graduate school I religiously photocopied and filed any map of the areas I studied that I happened to find—in textbooks, reference works, articles, monographs—because I knew I needed to fix the imagined historical geography of the regions I was studying in my mind. Maps were the way to create a topography for my historical imagination to roam. Ultimately this is the extent of the innovation of Constantinople as Palimpsest: to put an encyclopedia of information into a space. The digital platform simply makes it possible to load more information into a single map than would be possible in one that is static. A simple, accessible technology already provides us a way to re-integrate our specialist knowledge into its historical topography. We just need to use what is being given to us, what we already have. In other words, we can theorize the “digital” aspects of projects like mine, identifying how we are moving from a specifically codico-logical form of knowledge into the logics of the never-ending scroll or digital plane. Yet in the meantime, we also can adopt a heuristic, exploratory approach that is perfectly suited to classroom practices. Just as Anthony Grafton and Megan Williams have described the way in which Eusebius simply made use of the possibilities of the full page of the codex for his Chronicon, we simply can decide to teach through working out, in practice, what computers can do for us.27 26  König and Woolf, Encyclopaedism from Antiquity to the Renaissance, 1.

27  Grafton and Williams, Christianity and the Transformation of the Book.



Teaching Constantinople as a (Pixelated) Palimpsest

95

Digital projects can be approached as experiments in new forms, as open and collaborative explorations that end either when curiosity dries up, or when the question develops so that the medium no longer suffices. In the latter case, projects don’t really end but evolve—just as my own Constantinople as Palimpsest project continues to take new forms. The question is how to organize, and how the organization fits to what we are trying to do, how it permits what we think we might do, and what we think we might want to do. We can ease our way into how an interactive computer-generated topography (in truth no more intimidating than plugging addresses into a driving app) gives us new possibilities for thinking about how to display what we know, and what this allows us to think, together. How long my little project lasts is less important than what it accomplishes. Constantinople as Palimpsest won’t work forever, but its name signals this intent: it is there to be written over. If it is useful, other forms of it will be made. What is important now is how a re-integration of knowledge can change the way we work and think, together and with our students. As Silke Schwandt recently put it, [Digital Humanities] is a genuinely interdisciplinary endeavour making use of two things: digitization, or technologization, and hermeneutic interpretation. New digital technology transforms how we perceive and store information. It changes the ways of (social) interaction and communication. It allows access to vast amounts of information that need new ways of organization.28

Even when we pursue interdisciplinarity, we still most often functionally separate the same real objects into artificial subfields: manuscript studies, social history, intellectual history, economics, numismatics, archaeology, history of art, theology, etc. There is specialist value in these distinctions, of course. But we also all know that none of these subfields were distinct from each other in the moment-to-moment, day-to-day lives of our historical subjects. Good intentions don’t re-integrate those divisions—real structures of thought, and practice, and knowledge do. The value of project-based learning and scholarship is that it can help us practice how to return to a closer approximation of the worlds we seek to imagine, by integrating and re-connecting what we initially divided and distinguished in order to understand. We’ve come to understand a lot. Now it is time to re-integrate: to start putting the jigsaw puzzle back together, and to see what we’ve made.

28  Schwandt, “Introduction,” 16.

96

J. W. Torgerson

Bibliography

Aksamija, Nadja, Clark Maines, and Phillip B. Wagoner, eds. Palimpsests: Buildings, Sites, Time. Turnhout: Brepols, 2017. Bavidge, Jenny. Theorists of the City: Walter Benjamin, Henri Lefebvre and Michel de Certeau. Abingdon: Taylor & Francis, 2015. Beckett, Gulbahar. “Teacher and Student Evaluations of Project-Based Instruction.” TESL Canada Journal 19, no. 2 (June 2002): 52–66. Benjamin, Walter. The Arcades Project. Translated by Howard Eiland and Kevin McLaughlin. Prepared from the German volume edited by Rolf Tiedemann. Cambridge, MA: Belknap, 2003. Birnbaum, David J., Sheila Bonde, and Mike Kestemont. “The Digital Middle Ages: An Introduction.” Speculum 92, no. S1 (October 2017): S1–S38. Brubaker, Leslie. “The Vienna Dioskorides and Anicia Juliana.” In Byzantine Garden Culture, edited by Anthony Littlewood, Henry Maguire, and Joachim Wolschke-Bulmahn, 189– 214. Washington, DC: Dumbarton Oaks Research Library and Collection, 2002. Byzantinum 1200. https://www.byzantium1200.com/. “Byzantinum 1200.” Twitter. https://twitter.com/Byzantium1200. “Byzantinum 1200.” YouTube. https://www.youtube.com/user/Byzantium1200. Cameron, Averil and Judith Herrin. Constantinople in the Early Eighth Century: The Parastaseis Syntomoi Chronikai. Leiden: Brill, 1984. Clark, Bethany A. “Project Based Learning: Assessing and Measuring Student Participation.” Research and Evaluation in Education, Technology, and Design 39 (2017): unpag. http:// digitalcommons.unl.edu/cehsgpirw/39. Congdon, Eleanor A. “Imperial Commemoration and Ritual in the Typikon of the Monastery of Christ Pantokrator.” Revue des Études Byzantines 54, no. 1 (1996): 161–99. Constantinides, Costas N. “Byzantine Gardens and Horticulture in the Late Byzantine Period, 1204–1453: The Secular Sources.” In Byzantine Garden Culture, edited by Anthony Littlewood, Henry Maguire, and Joachim Wolschke-Bulmahn, 87–104. Washington, DC: Dumbarton Oaks Research Library and Collection, 2002. Constantinople as Palimpsest: The Place-Based Teaching Encyclopedia of Byzantium. Wesleyan University Traveler’s Lab. March 5, 2021 (published January 2021). https://arcg.is/01GXyj. Crow, James, Jonathan Bardill, and Richard Bayliss. The Water Supply of Byzantine Constantinople. London: Society for the Promotion of Roman Studies, 2008. Cruse, Marcus. “A Quantitative Analysis of Toponyms in a Manuscript of Marco Polo’s Devisement du Monde (London, British Library, MS Royal 19 D 1),” Speculum 92, no. S1 (October 2017): S247–64. De Certeau, Michel. The Practice of Everyday Life. Translated by Steven F. Rendall. Berkeley: University of California Press, 2011. De Gussem, Jeroen. “Bernard of Clairvaux and Nicholas of Montiéramey: Tracing the Secretarial Trail with Computational Stylistics.” Speculum 92, no. S1 (October 2017): S190–225. Grafton, Anthony and Megan Hale Williams. Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea. Cambridge, MA: Harvard University Press, 2006. Harrison, Martin. A Temple for Byzantium: The Discovery and Excavation of Anicia Juliana’s Palace-Church in Istanbul. London: Harvey Miller, 1989. Herrin, Judith. Byzantium: The Surprising Life of a Medieval Empire. Princeton: Princeton University Press, 2007. Janin, Raymond. Constantinople—Byzantine: Développement urbain et répertoire topograph­ ique. Paris: Institut français d’études byzantines, 1964.



Teaching Constantinople as a (Pixelated) Palimpsest

97

Kiilerich, Bente. “The Image of Anicia Juliana in the Vienna Dioscurides: Flattery or Appropriation of Imperial Imagery?” Symbolae Osloenses 76, no. 1 (2001): 169–90. Kleinberg, Ethan. Haunting History: For a Deconstructive Approach to the Past. Stanford: Stanford University Press, 2017. König, Jason and Greg Woolf. Encyclopaedism from Antiquity to the Renaissance. Cambridge: Cambridge University Press, 2013. Kotzabassi, Sofia, ed. The Pantokrator Monastery in Constantinople. Berlin: De Gruyter, 2013. Littlewood, Anthony, Henry Maguire, and Joachim Wolschke-Bulmahn, eds. Byzantine Garden Culture. Washington, DC: Dumbarton Oaks Research Library and Collection, 2002. Magdalino, Paul. “Neighbourhoods in Byzantine Constantinople.” In Hinter den Mauern und auf dem offenen Land. Leben im Byzantinischen Reich, edited by Falko Daim and Jörg Drauschke, 23–30. Mainz: Römisch-Germanisches Zentralmuseum, 2016. ______. “The Maritime Neighborhoods of Constantinople: Commercial and Residential Functions, Sixth to Twelfth Centuries.” Dumbarton Oaks Papers 54 (2000): 209–26. Mango, Cyril. Byzantium: The Empire of New Rome. New York: Scribner, 1984. Mou, Ty. “Students’ Evaluation of Their Experiences with Project-Based Learning in a 3D Design Class.” The Asia-Pacific Education Researcher 29 (2020): 159–70. Müller-Wiener, Wolfgang. Bildlexikon zur Topographie Istanbuls: Byzantion, Konstantinupolis, Istanbul bis zum Beginn des 17. Jahrhunderts. Tubingen: Wasmuth, 1977. Ousterhout, Robert. “Architecture, Art and Komnenian Patronage at the Pantokrator Monastery.” In Byzantine Constantinople: Monuments, Topography and Everyday Life, edited by Nevra Necipoğlu, 133–50. Leiden: Brill, 2001. ______. “Constantinople and the Construction of a Medieval Urban Identity.” In The Byzantine World, edited by Paul Stephenson, 334–51. London: Routledge, 2012. Poe, Edgar Allan. Tales. New York: Wiley and Putnam, 1845. Romanov, Maxim. “Algorithmic Analysis of Medieval Arabic Biographical Collections.” Speculum 92, no. S1 (October 2017): S226–46. Schwandt, Silke. “Introduction.” In Digital Methods in the Humanities: Challenges, Ideas, Perspectives, edited by Silke Schwandt, 7–22. Bielefeld: Bielefeld University Press, 2021. Shaya, Gregory. “The Flâneur, the Badaud, and the Making of a Mass Public in France, circa 1860–1910.” The American Historical Review 109, no. 1 (2004): 41–77. Shih, W.-L. and C.-Y. Tsai. “Students’ Perception of a Flipped Classroom Approach to Facilitating Online Project-Based Learning in Marketing Research Courses.” Australasian Journal of Educational Technology 33, no. 5 (2017): 32–49. Talbot, Alice-Mary. “Xenon of the Kral.” In The Oxford Dictionary of Byzantium. Prepared at Dumbarton Oaks. Edited by Alexander P. Kazhdan et al. 3 vols. Oxford: Oxford University Press, 1991 (online 2005): http://www.oxfordreference.com/view/10.1093/ acref/9780195046526.001.0001/acref-9780195046526-e-5847. Torgerson, Jesse W. “Constantinople as Palimpsest.” Traveler’s Lab. https://travelerslab. research.wesleyan.edu/constantinople/. ______. “Constantinople as Palimpsest: Now Live!” Traveler’s Lab. https://travelerslab.research. wesleyan.edu/2018/01/11/constantinople-live/. ______. “Constantinople as Palimpsest: Now with Words!” Traveler’s Lab. https://travelerslab. research.wesleyan.edu/2018/07/27/constantinople-as-palimpsest-now-with-words/. Torgerson, Jesse W., Jonah Skolnik, Alp Eren, Nathan Krieger, Connor Cobb, and Courtney Sachs. “Constantinople as Palimpsest.” BodoArXiv Works, January 25, 2021. doi:10.34055/ osf.io/ehmkx.

Chapter 5

LIFE ON—AND OFF—THE CONTINUUM LISA FAGIN DAVIS

When computer scientists need an antonym for “digital,” they tend to avoid

using the dated term “analog(ue).” Instead, they contrast discrete phenomena with a continuum, “digitally” with “continuously.”1 Anything digital is, at its most basic essence, a series of 1s and 0s, indivisible, with no fractional in-between. The “real world” operates on a smooth continuum, made up of building blocks that are essentially infinitely divisible. The real-world problems we want to solve exist along that continuum. Medieval manuscripts, for example, are made up of animal skin, oak gall ink, mineral and botanical pigments, all of which are, in their physical properties, infinitely complex, infinitely divisible. Digitization and data management convert our research questions into 1s and 0s, manageable and navigable pieces of binary code. That simplification sets fixed points along the continuum, transforming it into something cardinal, countable, and calculable. But the computer can’t enter its own data or interpret its own results. That’s where the humans, and the humanists, come in. It is said that Medieval Studies invented the Digital Humanities in the 1950s in the form of the Index Thomisticus, developed with IBM by Father Roberto Busa and a team of female punch-card operators. Seventy years later, medievalists such as those who contributed to this volume continue to apply resourcefulness, creativity, and sheer willpower to the development and use of digital methodologies. We brainstorm, we collaborate, we innovate, we meet obstacles, and we work to overcome them. The essays in this volume all have interrogated stages of digital medieval resources, from the oncegroundbreaking-but-now-obsolete to the redesigned to the shiny and new. All of these projects share a common formula: an investigator applies an algorithm to a dataset in an interface. The contributors have considered how digital resources and tools can, or did, or might, facilitate their work. These labyrinthine digital journeys are full of dead ends and backtracking, shortcuts and long delays. The lessons learned along the way are always valuable, whatever the research results, and each failed experiment leads to refinements of methodology. Thirty years ago, data was trapped in its interface. Today, best practices demand that data be easily extractable from the interface. We once expected our digital tools to be built to last forever. Now, it is now an accepted maxim that the framework of any given digital resource is destined for eventual obsolescence. It is the data within that must be preserved. Interoperability, controlled vocabularies, 1  Dr. Barry Fagin, Professor of Computer Science, United States Air Force Academy (private correspondence, February 25, 2021). Lisa Fagin Davis, Medieval Academy of America.

100

Lisa Fagin Davis

open source, and open access are the new best-practice watchwords. Static silos are a thing of the digital past. It was not always so, of course. Like many scholars of my generation, my graduateschool years straddled the transition from the continuum to the digital. When I began studying and cataloguing medieval manuscript fragments at the Beinecke Rare Book and Manuscript Library as a graduate student at Yale in 1988, the only way to identify a patristic text on a fragment was to use the published concordances of collections like the Patrologia Latina or Corpus Christianorum, looking up various keywords until I got lucky. It was tedious, but often satisfying, work. By the time I completed my Ph.D. in 1993, I could sit down in front of a computer station at Sterling Memorial Library, run the cetedoc cd-rom, and have a text identification within seconds.2 I remember thinking it was almost TOO easy. In the thirty years since I loaded that first cd-rom, I have started, abandoned, and resurrected three different Digital Humanities projects. In each case, digital and financial constraints of the time made the original implementation of the project essentially impossible. Years later, human encounters along the continuum led to collaborations that allowed me to reimagine each project and bring it to fruition in the digital realm.

DigiPal, or, Two Datasets in Search of a Framework (1993–2018)

My primary specialization is Latin paleography. I spent the first decade or so of my scholarly life studying a particular script in a particular time, the proto-Gothic script written by Benedictine monks in a group of closely-knit twelfth-century Upper Austrian monasteries, among them Lambach, Admont, Kremsmünster, and Melk. My research was based on the fundamental paleographical axiom that in any given place and time, there are certain script features that are particularly useful for distinguishing scribal hands. Abbreviations and punctuation, such as interrogativus, are often useful as well. As paleographers and modern handwriting experts know, every scribe has in his or her repertoire a prototypical form of each grapheme, and it towards that prototype that she or he is always working. In other words, while the form of the letter [g] used by a particular scribe at any particular time may vary in minute detail, the ideal form of the letter, characterized by its ductus, will remain the same for at least some period of the scribe’s career. Because my doctoral thesis had focused on the work of one particular scribe in one particular time and place—Gottschalk of Lambach in the late twelfth century—I had developed a particular eye for scripts and scribes then and there. I had—almost unconsciously—developed a habit of turning to particular graphemes first when encountering a new scribe. I also noticed the work of certain scribes appearing in manuscripts produced in or made for different abbeys in the region—for example, Gottschalk’s hand in a manuscript known to have been produced in nearby Kremsmünster, or a scribe prevalent in Admont appearing in Lambach. I began to wonder about such patterns of transmission and developed a proposal for a monograph study of book production in Romanesque Upper Austria. 2  The cetedoc Library of Christian Latin Texts on CD-ROM.



Life on—and off—the Continuum

101

By this time, it was 1997. The idea of a digital manuscript repository was in its infancy, with the launch of resources such as Digital Scriptorium barely on the horizon. The only way to access the hundreds of manuscripts I wanted to study was to go to Austria and photograph them myself or study them on microfilm. Off I went to the Hill Monastic Manuscript Library in Minnesota to spend two weeks browsing microfilm of more than five hundred manuscripts and printing more than twelve hundred pages of copyflow. One of the advantages of working on manuscripts from Upper Austria is that, unlike in France and England, the medieval monastic libraries are mostly intact. Some codices have been de-accessioned over the years, and many books were dismantled in the early modern period for binding scrap, but for the most part manuscripts that were in an Austrian abbey library in the Middle Ages remain in that abbey library today, or comprise a coherent collection in a larger regional or national library. This region is a nearly-sterile laboratory for studying manuscript production and transmission among a tightly-grouped and closely-connected family of abbeys. The connections between these abbeys are documented in chronicles, necrologies, charters, and other documents. Berthold, one of the first monks at Göttweig, went on to become abbot of Garsten. One of the monks who was a novice during his abbatiate, Alram, later became abbot of Kremsmünster.3 One monk, Sigibold, was simultaneously abbot of Lambach and Melk in 1116.4 A monk trained at Lambach went on to become abbot of Göttweig, the abbey that trained St. Lambrecht’s first abbot. Monks from one abbey became the founding brothers of another and novices from one abbey eventually became abbots nearby. These are just a few of the connections forged by the bonds between the abbeys of the Passau diocese, originally based on the friendship of the three eleventh-century bishops who founded Lambach, Admont, and Gottweig. These bonds were strengthened by the transmission of monastic reform in the twelfth century. My work on Lambach and a case study I published on the local transmission of Bernard of Clairvaux’s Sermones super Canticum Canticorum suggested that these connections were also made manifest in manuscript production, as monks of the reform movement copied each other’s liturgy and texts and influenced each other artistically.5 This is the complicated network of production and transmission that I hoped to untangle. From the microfilm collection at the Hill Monastic Manuscript Library, I printed more than twelve hundred scribal samples, sheets that are still filed in a cabinet in a closet in my home office. The quality is a bit spotty: this is twenty-five-year-old blackand-white copyflow printed from microfilm of manuscripts photographed sixty years ago, and many of the images are marred by scratches or blemishes on the original film. Even so, they remain, for the most part, bright and legible even now. At the same time, I worked to establish the taxonometric criteria for each letterform. In 1998, digitization and metadata modelling were in their infancy, and as an independent scholar I didn’t have access to those burgeoning technologies. I had to develop my 3  Neumüller, “Zur Benediktinerreform des heiligen Altman,” 20.

4  “Sigiboldus abbas ... qui plures rexisse abbatias fertur, inter quas et Medeliccam tenuit”: from Vita Adalberonis episcopi Wirziburgensis, ed. Wattenbach, 136. 5  Davis, “Bernard of Clairvaux’s Sermones Super Cantica Canticorum.”

102

Lisa Fagin Davis

own classification system, and consider how I was going to organize all of that data. My close work on the Lambach scriptorium helped me to identify three salient features useful when distinguishing Romanesque hands in Upper Austria: the shape of the t-ascender in the [ct] sequence, the relative size and placement of the upper and lower bowls of the [g] grapheme, and certain details of the [et]-abbreviation, whether Tironian or ampersand. I developed a system that associated a series of numbers with each grapheme, describing its particular features, creating in the process a numeric paleographic taxonomy. A particular [ct] might be classified as “12,” implying that the two letters are not joined (the first digit) and the [t] ascender curls into a circle at the top (the second digit). An [&] whose upper bowl closes on the vertical stem and whose final stroke is clubbed is classified as “311”, with each digit representing a particular feature. For [g], the system is based on the relative size and position of the upper and lower bowl, and whether the lower bowl is open or closed. Once I had established these parameters, I spent several months classifying each of the 1200 script samples in this way, associating three numbers with each sample that described its particular [ct]-sequence, [et]-abbreviation, and [g]-form. The scribe Gottschalk of Lambach, for example, is identified by the numbers [12 311 221]. To track the data, I initially set up an. mdb database model in Microsoft Access, but when I realized that I would simply be sorting data rather than running complex queries, I abandoned the .mdb model for an Excel spreadsheet saved as a .csv file. I also recorded the numeric series for each particular scribal sample on the back of its printout, identifying them by shelf mark and their order of appearance within the manuscript (for example, “Admont 256 2” refers to the second hand to appear in Admont, Stiftsbibliothek Hs 256). By sorting the spreadsheet, script samples with the same features would find one another. The associated script samples could then be compared by eye to confirm whether they indeed were written by the same hand. The system proved to be successful. A test case of forty-two script samples from twenty-one Garsten manuscripts produced ten groups of samples with the same numeric designation. Five of these groups turned out upon examination to have been written by the same scribe, and none of the samples with different classifications were the same hand. The taxonomy clearly offered a shortcut, at the very least. But I had twelve hundred lines of numeric classifications that needed to be sorted and compared. It was simply too much data for a human brain to manage. I needed a database that would associate numeric data with a digital image, although I didn’t realize it at the time. I only knew that the hybrid system I had developed—sorting the Excel data in the digital realm, then pulling copyflow manually from my continuum-based file cabinet to visually compare script samples—was too cumbersome and time consuming to allow for an analysis of more than a thousand samples in hundreds of manuscripts from a dozen scriptoria over a hundred years. I published an article about the methodology and put the project aside.6 I turned my attention to other things, always wondering what patterns might have emerged if I had been able to apply the strength of sophisticated digital analysis to the full set of copyflow samples. 6  Davis, “Towards an Automated System of Script Classification.”



Life on—and off—the Continuum

103

What I couldn’t have known then was that by choosing .csv format over .mdb, I was ensuring that my data would last. I have carried that data forward through multiple hard-drives and software updates and operating systems. Wherever I went, the spreadsheet came with me, backed up on an external drive and eventually in the cloud, waiting patiently for me to figure out what to do with it. The file remains useful and uncorrupted, even though it was first created twenty years, several software updates, and at least six computers ago. The copyflow printouts have been waiting as well, sitting in that dusty file cabinet in a closet in my home office, faintly calling to me with that persistent quiet voice that unfinished projects seem to have. Occasionally, I have had reason to refer to the spreadsheet, when an Upper Austrian manuscript comes up for sale at auction and the scribe seems familiar, or someone writes to me with a question about a manuscript, but I just couldn’t see how to turn this mass of images and data into something really valuable and lasting. Not, that is, until DigiPal. In 2015, when I saw the DigiPal framework (now known as Archetype) demonstrated by Stewart Brookes at the International Medieval Congress in Leeds, I knew that here, finally, was the answer to my paleographical prayers: a resource that combined image annotation with sophisticated discoverability and a customizable data model.7 I cornered Professor Brookes at the wine hour on the Union patio and asked: Could DigiPal combine my images and data to answer my original research question? Yes, it could. So began a fruitful collaboration with Brookes, his co-investigator Peter Stokes, and the team of coders at Kings College Digital Laboratory in London (you’ll find that cornering potential collaborators at conference receptions is a recurring theme). The system does exactly what I need it to do—a DIY framework that can be applied to digital images using annotation and a custom paleographical database model, resulting in complex discoverability. Using DigiPal/Archetype, I could apply the paleographical taxonomy to my script samples, sort them, and compare the results all in one interface. Today, Archetype allows me to combine my two datasets—taxonomy and images—and query the data to investigate my research question: How did the Benedictine scriptoria of twelfth-century Upper Austria collaborate? Network analysis will provide a visualization of scribes, texts, and other modes of interconnection. My data and images have found their framework.8

DigitalMappa, or, Free the Data (2005–2020)

In 2005, I was invited by curator Nancy Netzer to take part in the planning of an exhibition at the McMullen Museum of Art at Boston College that would contrast secular and sacred images in medieval manuscripts from the Boston Public Library and the Museum of Fine Arts Boston.9 In particular, I was asked to take on the research and

7  “Archetype,” accessed February 28, 2021; Stewart Brookes, “Getting Cursive: Extending DigiPal’s Framework for Models of Authority,” paper delivered at the International Medieval Congress, Leeds University, 6 July 2015. 8  I have applied the same methodology to study the script and scribes of Beinecke Library MS 408, a.k.a. the “Voynich Manuscript”. See Davis, “How Many Glyphs and How Many Scribes?” and “Digital Paleography and the Voynich Manuscript.” 9  The “Secular/Sacred” exhibition, February to June 2006.

104

Lisa Fagin Davis

writing for a lengthy illuminated scroll that had lain unstudied and essentially unknown in the Special Collections stacks at the Boston Public Library for a century, a universal chronicle and genealogy of humanity written in French around the year 1480 (BPL Ms Pb Med. 32).10 The anonymous text I christened the Chronique Anonyme Universelle is a fascinating compilation of historical and literary sources. My research for the exhibition raised more questions than it answered, and so I continued to study the Chronique even after the exhibition had closed. Over the course my research, I identified twenty-eight manuscripts of the Chronique, investigated its source material, and established the transmission tradition. A National Endowment for the Humanities (neh) fellowship for independent scholars allowed me to travel through Europe in 2007 studying the related scrolls in situ. Eventually, the project resulted in a monograph combining a full-scale critical edition and facing translation, a detailed commentary on the source material, and full descriptions of each of the identified manuscripts, published by Brepols in 2014.11 In the midst of this and other projects, I attended the 2010 annual meeting of the Medieval Academy of America at Yale University and heard Martin Foys introduce his DigitalMappa project.12 Using the Cotton Map as a test case, Professor Foys was developing a powerful annotation tool that would allow him to identify and tag locations on the Map, annotations that could then be made discoverable.13 That evening, I cornered Foys at a reception on the mezzanine of the Beinecke Rare Book and Manuscript Library with a question (I told you this was a recurring theme): Could DigitalMappa be used with the universal chronicle scroll to create an annotated and discoverable digital resource? Yes, it could. So began another fruitful collaboration in which the Chronique Anonyme Universelle served as a test case for the DigitalMappa framework. We selected a scroll in private hands, a choice that simplified imaging, access, and permissions. The collector funded high-resolution imaging of the scroll and development of the model, supplementing major grants from the neh. We had a working prototype by April 2011.14 By this time, my monograph was nearing publication. I reached out to my editor to discuss linking the online digital resource with the monograph. The publisher wanted the Digital Chronique to be a password-protected, monetized resource. Lacking the digital infrastructure to support the resource online, however, we eventually agreed that the only viable option was to include the digital resource as a DVD tucked into the back of the book. Unfortunately, compressing this massive online resource into a DVD led to a significant loss of functionality and compromised its long-term viability. The monograph was published in 2014, and the DVD was obsolete within two years. The data was trapped in an outdated and unsustainable framework. As one reviewer put it, 10  “Chronique Anonyme Universelle,” Digital Commonwealth, accessed online February 28, 2021.

11  Davis, La Chronique Anonyme Universelle.

12  Foys, “Interoperability, Intermediate Data, and Scholarly Siloing.”

13  London, British Library, Cotton MS Tiberius B V/1, fol. 56v, accessed online February 28, 2021.

14  The DigitalMappa 1.0 interface was developed by Martin Foys, Shannon Bradshaw, and Tim Andres of Drew University.



Life on—and off—the Continuum

105

… The software is slow and glitchy, and users may find it simpler in the long run to use the book’s printed index, but with practice and patience, the search tool becomes manageable, and the digital images make it easy both to scrutinize details of the manuscript and to experience the Chronique in something much closer to its native state. The disk may prove especially helpful for teaching purposes at institutions that do not have scroll texts in their collections.

That is, provided the user can make the DVD run at all. It bespeaks both the promise of digital resources and the frailty of the technology in which they are embedded. (Parchment here again wins the technological day for durability.) Although the operating system of my own computer was theoretically compatible with the disk’s software, hours of effort and extra software downloads yielded only frustration, and it literally took months to find a machine on which the disk would run. As time passes and the disk’s software becomes more obsolete, that problem will only grow worse, and it is to be hoped that the publisher finds a way to put the materials online and maintain them, so that the vast amount of labor they represent and the information they provide will not be lost.15

The reviewer wasn’t wrong. The decision to burn the data and its interface onto an unsustainable and closed format welded them together and doomed the resource to obsolescence. Enter the Schoenberg Institute for Manuscript Studies at the University of Pennsylvania, which in 2016 began an neh-funded collaboration with Foys to develop DigitalMappa 2.0, a project that was to be developed with the commitment to open access and open data for which the Schoenberg Institute is renowned. Once again, the Digital Chronique was to serve as one of several case studies. Some of the data from the first iteration had been lost and had to be recreated, but in spite of that delay, the new Digital Chronique launched on the DigitalMappa 2.0 platform in 2018.16 This online, open access, interoperable resource includes expanded functionality such as keyword searching, colour coding, and cross linking. Thanks to the collaborative generosity of colleagues at the Schoenberg Institute and the University of Wisconsin, a project that seemed to have hit insurmountable roadblocks in both the continuum (monetization of resources) and the digital (data inextricability and platform obsolescence) is now available online as a sustainable, open, and fully-functional resource.17

Fragmentarium, or, A Digital Solution to a Continuous Problem (1989–2021)

I have been many things over the course of my career as a medievalist: freelance manuscript cataloguer, adjunct professor of library science, administrator, blogger, paleographer, codicologist, and fragmentologist. “Fragmentology”—the discipline devoted to the study of manuscript fragments—is an invented word, a Frankenstein-ed combination of a Latin root and a Greek suffix. It is a word I heard for the first time in 2013 at a meeting in Geneva, convened to discuss the status and future of the international study of medi15  Julia Marvin’s review of Davis, La Chronique Anonyme Universelle, 1181. 16  “Digital Chronique 2.0,” accessed online February 28, 2021.

17  With thanks to William Noel, Dot Porter, Martin Foys, Kira Zimmerman, Max Gray, and Alex Ukropen.

106

Lisa Fagin Davis

eval manuscript fragments. At the time, each of the participants was asked to introduce themselves and say something about their background and scholarship. I introduced myself as a paleographer and codicologist, but by the end of the meeting I identified myself, without hesitation, as a fragmentologist. As it turned out, I had been a fragmentologist for twenty-four years, but just hadn’t had a name for it. Flashback to 1989, my second year of graduate school, the year I began working for the Beinecke Library’s Curator of Pre-1600 Manuscripts, Robert G. Babcock, as his graduate-student assistant. In addition to standard assistant responsibilities such as photocopying and tracking down books and articles, I was given what would turn out to be a life-changing assignment. Dr. Babcock sat me down in front of two enormous boxes of manuscript fragments and put me to work on their conservation and cataloguing. This project would occupy me for several years, and set us both off on a great adventure that would result in an exhibition, several collaborative publications, a few articles, my dissertation, and my first monograph.18 Unlike the single leaves I often blog or Tweet about, these fragments, now known collectively as Beinecke MS 481 and 482, were not the victims of twentieth-century greed-induced biblioclasm but rather the byproducts of late-medieval recycling and reuse.19 Parchment being a valuable resource, it was quite common in the early-modern period for out-of-date or damaged medieval codices to be dismantled for use as binding scrap—pastedowns, flyleaves, binding stays, or wraparounds—in late-medieval and early-modern bindings. Over the course of the Beinecke project, I learned how to interpret the material evidence to distinguish different layers of use: sewing holes from the original manuscript, scars left by the binder’s knife, annotations written by readers of the bound book, the tape residue from a modern dealer. I was trained on the job in the basics of parchment conservation: carefully removing the fragments from the acidic mattes into which they had been scotch-taped, making basic repairs in Yale’s conservation lab using Japanese tissue and a water-based paste, and housing the leaves in custom mylar sheaths stored in archival folders and boxes. And, of course, I catalogued them, working with a team of students, including Philip Rusche and Nancy Seybold. That work later became the fourth volume of the Beinecke’s published catalogue.20 Over the course of the project, seventeen leaves from a twelfth-century antiphonal—Beinecke MS 481.51—caught my eye.21 I found myself drawn to the elaborate and graceful initials, the unheightened neumes, and the system of marginal tonary-letters, as well as the collection’s codicological features and the intellectual and liturgical challenge posed by the task of putting the leaves in the correct order. 18  Davis, The Gottschalk Antiphonary.

19  “Beinecke Rare Book and Manuscript Library MS 481,” Yale University, accessed online February 28, 2021; “Beinecke Rare Book and Manuscript Library MS 482,” Yale University, accessed online February 28, 2021. 20  Babcock, Davis, and Rusche, Catalogue of Medieval and Renaissance Manuscripts in the Beinecke Rare Book and Manuscript Library, Vol. IV.

21  “Yale University, Beinecke Rare Book & Manuscript Library MS 481.51,” accessed online February 28, 2021.



Life on—and off—the Continuum

107

The story of how Babcock, Rusche, Seybold, and I traced this manuscript and dozens of other Yale fragments to their medieval origin at the Austrian Benediktinerstift Lambach has been told elsewhere.22 Suffice it to say that we identified the artist and scribe of the antiphonal as Gottschalk of Lambach—hence, we christened the fragmentary manuscript the Gottschalk Antiphonal. In addition to the seventeen Gottschalk Antiphonal leaves at Yale, there are two at Harvard’s Houghton Library, one that toured the midwestern United States in an aluminum trailer in the 1950s before settling down at the St. Louis Public Library, and a few still in Austria (in the abbey of St. Paul-imLavanttal, at a resort hotel in Badgastein, and in Lambach itself, although the incunable flyleaves observed there as recently as 1998 have since vanished and are known only in black-and-white photographs). In 2016, an offset of another leaf of the Gottschalk Antiphonal was found in an incunable belonging to the Beinecke Library, discovered and imaged by Dr. Elizabeth Hebbard. The mirror-image remnant was left behind when the actual leaf was peeled off from the wooden board, where it had been used to secure the leather turn-ins on the back cover. Ironically, the volume had been at the Beinecke for decades by the time I wrote my thesis there, but it was only during a recent survey of the bindings by Dr. Hebbard that the offset was photographed and identified. Inverted and rotated, it is perfectly legible and easily identifiable as a now-lost leaf of the Gottschalk Antiphonal.23 In my dissertation and the later monograph, I reconstructed the extant portion of the manuscript and conducted a detailed study of its liturgy, decoration, notation, script, and history.24 The Beinecke collections inspired my interest in fragmented manuscripts. Recently, I have focused that attention on the biblioclastic detritus left by Ohio bookdealer Otto F. Ege, who dismembered and scattered several hundred manuscripts and printed books in the first half of the twentieth century. In 2007, I began collecting images and data for the scattered leaves of one manuscript in particular, a late thirteenth-century French manuscript known as the Beauvais Missal.25 The manuscript was dismembered by Ege and his business partner Phillip Duschnes in 1942 and is now scattered across the world. With no other affordable and accessible framework yet available, I set up an Omeka.net site in 2013 where I could post Beauvais Missal images and their associated data.26 As of this writing, I have identified 113 leaves, and I continue to update the site regularly. Omeka works well as a place to post metadata and images and arrange them in exhibit spaces as tiled thumbnails, but what I really wanted was to create digital facsimiles of the Beauvais Missal and other dismembered manuscripts, digital surrogates that could be browsed page by page, leaf by leaf, opening by opening, with discoverable metadata. In 2014, I heard Benjamin Albritton give a paper at the International Congress on Medieval Studies in Kalamazoo about the potential of shared-canvas viewers to do just 22  Babcock, Reconstructing a Medieval Library.

23  “Gottschalk Antiphonal (Fragment),” Fragmentarium, accessed online February 28, 2021.

24  Davis, “Epiphany at Lambach” and Davis, The Gottschalk Antiphonary. 25  Davis, “The Beauvais Missal.”

26  Davis, “Reconstructing the Beauvais Missal.”

108

Lisa Fagin Davis

that.27 I cornered Albritton over wine and cheese in the Fetzer lounge that evening to ask if the model could be applied to the scattered leaves of the Beauvais Missal and other dismembered medieval manuscripts. Yes, he said, it could. Dr. Albritton’s case study was reliant on the newly-developed International image Interoperability Framework, or IIIF. The International Image Interoperability Framework is a way of presenting digital images in an online environment that allows them to be shared via a persistent URL instead of by downloading and uploading into a silo. In other words, if an online image is IIIF-compliant, it can be manifested in a workspace known as a “shared canvas” simply by pointing to the IIIF URL. The image is drawn into the shared canvas when called for rather than being physically stored there. This interoperability has the advantage of enabling a user to apply their own metadata and annotations and sequence the images without transforming the actual image file. An image can be stored in one place while being used in multiple workspaces. The model is completely openaccess and avoids siloing, and is thus in keeping with digital best practices. It was clear from Dr. Albritton’s case study that the model would do exactly what I wanted. IIIF is an open-source-code application. The proliferation of such projects has ushered in an era of parallel development, as multiple developers customize the code for their own use. Several fragmentology projects evolved in the mid-2010s that took advantage of the new standards of open-source, -access, and -data. Dr. Albritton’s case study used IIIF and the Mirador shared-canvas viewer to digitally reconstruct a manuscript that had been dismembered by Otto Ege in the 1930s.28 At the same time, Debra Cashion at St. Louis University was developing the Broken Books project, combining IIIF images with a fragment-centric data model to reconstruct dismembered manuscripts.29 My reconstruction of the Beauvais Missal would serve as one of two case studies for the Broken Books interface. Broken Books launched in the fall of 2015 and ran until 2019. The project served as a very successful case study in the potential of combining sharedcanvas viewers with a custom data model to digitally reconstruct dismembered manuscripts.30 A third IIIF-based project grew out of the 2013 meeting in Geneva, culminating in the 2016 launch of the online resource Fragmentarium.31 Fragmentarium is a fragment-centric workspace that takes advantage of IIIF functionality to easily allow users to upload, catalogue, and arrange leaves to create digital reconstructions. After the cessation of the Broken Books project, I migrated the data and images of the Beauvais Missal to the Fragmentarium platform and used the site’s IIIF functionality to sequence the known leaves into a virtual reconstruction that combines discoverable metadata with image scrolling.32 In 2020, I used Fragmentarium to create a digital recon27  Albritton and Whearty, “Scattered Leaves.”

28  Albritton’s proof-of-concept project is now defunct, but screencaps are available at Warner, “IIIF Introduction and Opportunities at Cornell,” lecture at Cornell University on February 5, 2015, accessed February 28, 2021. 29  Cashion, Broken Books, https://brokenbooks.omeka.net/, accessed February 28, 2021.

30  Cashion, “Broken Books.”

31  Fragmentarium, https://fragmentarium.ms, accessed February 28, 2021.

32  “Beauvais Missal (Virtual Reconstruction),” Fragmentarium, accessed online February 28, 2021.



Life on—and off—the Continuum

109

struction of the Gottschalk Antiphonal, a far cry from the scissors and paste and black-andwhite photocopies of the 1990s.33 Hundreds of thousands of manuscript fragments are scattered across the world, remnants of thousands of dismembered manuscripts. Sharedcanvas interoperability provides a digital solution to a continuous problem.

Conclusions: Life on—and off—the Continuum

The experience of putting down and picking up these three projects taught me important lessons about how we as medievalists can develop and use digital tools to effectively facilitate our investigations of the medieval past. Data must be extricable if it is to survive; resources should be open access if possible; and interoperability points the way to a sustainable digital future. The most important components of my own identity as a Digital Medievalist, however, are not found in the digital realm but exist along the continuum: the medieval objects with which we engage, and the colleagues on whom we depend. The human Here-and-Now and the object Then-and-There are chronological and geographic points along a continuum that stretches infinitely in all dimensions, a distance that can seem insurmountable. The digital lays a co-ordinate grid over the continuum and allows us to navigate it discretely and precisely, helping us find our way from Here and Now back to the medieval Then and There. The global pandemic of 2020–2021 broke the connection between the continuum and the digital, limiting us to the virtual grid and separating us from our objects and our collaborators along the continuum. For more than a year, the only way to engage was with digitized objects and virtual colleagues. As we slowly emerge from the horrors of the pandemic, we are reforging the balance between the digital and the continuum. Both are necessary components of our work. It is increasingly difficult to engage with the continuous medieval without any form of digital mediation, now that the term “digital” can mean anything from reading PDFs to mastering TEI encoding. Even as we reclaim our space along the continuum, we are all, in one way or another, digital medievalists.

Bibliography Albritton, Benjamin and Bridget Whearty. “Scattered Leaves: New Approaches to Digital Manuscript Studies.” Paper presented at the 49th International Congress on Medieval Studies, Western Michigan University, Kalamazoo, May 8, 2014. Archetype. https://archetype.ink/. Babcock, Robert G. Reconstructing a Medieval Library: Fragments from Lambach. New Haven: Beinecke Rare Book and Manuscript Library, 1993. ______, Lisa Fagin Davis, and Philip Rusche. Catalogue of Medieval and Renaissance Manuscripts in the Beinecke Rare Book and Manuscript Library, Vol. IV. Binghamton: Center for Medieval and Early Renaissance Studies, 2004. 33  “Gottschalk Antiphonal (Virtual Reconstruction),” Fragmentarium, accessed online February 28, 2021.

110

Lisa Fagin Davis

“Beauvais Missal (Virtual Reconstruction),” Fragmentarium. https://fragmentarium.ms/ overview/F–4ihz. “Beinecke Rare Book & Manuscript Library MS 481.” Beinecke Rare Book and Manuscript Library, Yale University. https://collections.library.yale.edu/catalog?search_field=all_ fields&q=481.*. “Beinecke Rare Book & Manuscript Library MS 481.51.” Beinecke Rare Book and Manuscript Library, Yale University. https://collections.library.yale.edu/catalog/2003631. “Beinecke Rare Book & Manuscript Library MS 482.” Beinecke Rare Book & Manuscript Library, Yale University. https://collections.library.yale.edu/catalog?search_field=all_ fields&q=ms+482.*. Brookes, Stewart. “Getting Cursive: Extending DigiPal’s Framework for Models of Authority.” Paper presented at the International Medieval Congress, Leeds University, 6 July 2015. Cashion, Debra Taylor. Broken Books. https://brokenbooks.omeka.net/. ______. “Broken Books.” Manuscript Studies 1, no. 2 (2016): 342–52. The cetedoc Library of Christian Latin Texts on CD-ROM. Turnhout: Brepols, 1991–. “Chronique Anonyme Universelle.” Digital Commonwealth. https://www.digitalcommonwealth.org/search/commonwealth:np193j76j. “Cotton MS Tiberius B V/1, f. 56v.” London, British Library. http://www.bl.uk/manuscripts/ FullDisplay.aspx?ref=Cotton_MS_Tiberius_B_V/1. Davis, Lisa Fagin. “The Beauvais Missal: Otto Ege’s Scattered Leaves and Digital Surrogacy,” Florilegium 33 (2016): 143–66. ______. “Bernard of Clairvaux’s Sermones Super Cantica Canticorum in Twelfth-Century Austria.” In Manuscripts and Monastic Culture: Reform and Renewal in Twelfth-Century Germany, edited by Alison Beach, 285–310. Turnhout: Brepols, 2007. ______. La Chronique Anonyme Universelle: Reading and Writing History in Fifteenth-Century France. London: Harvey Miller, 2014. ______. “Digital Paleography and the Voynich Manuscript.” In Digital Palaeography, edited by Stewart Brookes, Malte Rehbein, and Peter Stokes. New York: Routledge, forthcoming. ______. “Epiphany at Lambach: The Evidence of the Gottschalk Antiphonary.” PhD dissertation, Yale University, 1993. ______. The Gottschalk Antiphonary: Music and Liturgy in Twelfth-Century Lambach. Cambridge: Cambridge University Press, 2000. ______. “How Many Glyphs and How Many Scribes? Digital Paleography and the Voynich Manuscript,” Manuscript Studies 5, no. 1 (2020): 164–80. ______. “Towards an Automated System of Script Classification.” Manuscripta 42, no. 3 (November 1998): 193–201. “Digital Chronique 2.0.” Schoenberg Institute for Manuscript Studies, University of Pennsylvania. https://sims2.digitalmappa.org/34. Foys, Martin. “Interoperability, Intermediate Data, and Scholarly Siloing in Current Digital Manuscript Resources,” Paper presented at the annual meeting of the Medieval Academy of America, Yale University, New Haven, March 20, 2010. Fragmentarium. https://fragmentarium.ms. “Gottschalk Antiphonal (Fragment), New Haven, CT, Beinecke Rare Book and Manuscript Library, Zi +1525.” Fragmentarium. https://fragmentarium.ms/overview/F-8fu0. “Gottschalk Antiphonal (Virtual Reconstruction).” Fragmentarium. https://fragmentarium. ms/overview/F-75ud. Marvin, Julia. Review of Lisa Fagin Davis, La Chronique Anonyme Universelle: Reading and Writing History in Fifteenth-Century France.” Speculum 92 (2017): 1180–81.



Life on—and off—the Continuum

111

Neumüller, Willibrord. “Zur Benediktinerreform des heiligen Altman.” In Der heilige Altmann Bischof von Passau: Sein Leben und sein Werk, edited by S. K. Landersdorfer, 16–22. Göttweig: Abtei Göttweig, 1965. “Secular/Sacred: 11th–16th Century Works from the Boston Public Library and the Museum of Fine Arts, Boston.” McMullen Museum of Art (exhibition), February to June 2006. https://www.bc.edu/sites/artmuseum/exhibitions/secular-sacred/. Vita Adalberonis episcopi Wirziburgensis. Ed. Wilhelm Wattenbach, 127–38. MGH SS 12. Hannover: Hahn, 1856. Warner, Simon. “IIIF Introduction and Opportunities at Cornell.” Lecture delivered at Cornell University, February 5, 2015. https://www.slideshare.net/simeonwarner/2015-01-cornellvrwgiiif.

Appendix

PERMANENT LINKS TO THE CATALOGUED ASSETS OF PROFILED PROJECTS Below you will find links to the catalogued digital assets used to build and maintain the projects profiled in this volume’s essays. By cataloguing materials in their most durable file formats and according to protocols put forward in the Digital Documentation Process (digitalhumanitiesddp.com), each project initiator has ensured that his or her work will be findable and citable for future scholars, both during and after a project’s active phase.

Dorin, Rowan. “Corpus Synodalium (March 2021): Archiving Dossier Narrative.” BodoArXiv Works, March 24, 2021. doi:10.34055/osf.io/qx2ve. ______. ed. Corpus Synodalium: Local Ecclesiastical Legislation in Medieval Europe, June 30, 2021. http://www.corpus-synodalium.com. Everhart, Deborah and Martin Irvine. “The Labyrinth Medieval Studies Website.” Digital Georgetown. http://hdl.handle.net/10822/1061738. McEwan, John. “Digital Sigillography Resource, Version 1.” BodoArXiv Works, July 15, 2021. doi:10.34055/osf.io/nrh8x. Torgerson, Jesse W., Jonah Skolnik, Alp Eren, Nathan Krieger, Connor Cobb, and Courtney Sachs. “Constantinople as Palimpsest.” BodoArXiv Works, January 25, 2021. doi:10.34055/ osf.io/ehmkx.

INDEX

Accursius, 8, 77 Admont, monastery of, 100, 102 Adobe Cold Fusion (web server), 27–28, 29, 30 adultery, 70 Albritton, Benjamin, 107–8 Alram, Abbot of Kremsmünster, 101 American Standard Code for Information Interchange (ASCII), 32 Amiens, diocese of, 70 Anicia Juliana, palace of, 89 ArcGIS (geographic information system), 79–80, 82, 87 Archives Nationales (Paris), 37 Archiving Dossier Narrative, 6, 30 Ariadne and Theseus, myth of, 13–14 ARTFL Project, 61 Arthur, mythical king, 22, 29 Artonne, André, 53 ASCII see American Standard Code for Information Interchange Avril, Joseph, 51 Babcock, Robert G., 105 Badgastein, 107 Barralis, Christine, 56 Basil I, Emperor, 90 Basilica Cistern, 89 Bass, Randy, 17 Baths of Zeuxippus, 79 Beauvais, diocese of, 70 Beauvais Missal, 107–8 Beinecke Rare Book and Manuscript Library, 100, 105 Benjamin, Walter, 84 Beowulf, 22 Bernard of Clairvaux, 101 Berners-Lee, Tim, 28 Berthold of Göttweig, 101 Bible, 3, 20 biblioclasm, 106, 107, 108 bibliographies (online), 12, 20, 24, 72 Birch, Walter de Gray, 40, 41, 42–43, 45

bishops see canon law—episcopal Blachernae Palace (Constantinople), 89, 90, 91 Blair, Charles Henry Hunter, 45 Blue, Paula, 84 Bodleian Library (Oxford), 20 Bologna, 50 Boston (Massachusetts): Boston College, 103 Boston Public Library, 103–4 Bourges, diocese of, 65 British Library (London), 35–36, 39, 42 British Museum (London), 34, 39, 42 Broken Books project, 108 Brookes, Stewart, 103 Brown, George, 22 Brown University, 4 Bryn Mawr Medieval Review, 24 Busa, Roberto, 77, 99 Byzantine Studies, 84, 87, 88 See also Oxford Dictionary of Byzantium Byzantion, 79 Byzantium, 78, 79, 85, 89, 93 Byzantium 1200 project, 88 Cambrai, diocese of, 70 Cameron, Averil, 88 canon law, 7, 49 codification of, 50–51, 69 databases of, 53–55 dissemination of, 50–51, 59, 69 editions of, 49–50, 51–52 English, 51–52, 54 episcopal, 51–57, 68–70, 72–73 French, 51, 53, 54, 58 German, 52, 53 Iberian, 52 Italian, 52, 53 particularity of, 51, 54, 69 Polish, 52 Scandinavian, 52 scholarship on, 54, 56, 68–69 See also Corpus Synodalium

114

index

Cantelar Rodriguez, Francisco, 52 CARA see Committee on Centers and Regional Associations Cascading Style Sheets (CSS) (style sheet language), 42 Cashion, Debra, 108 CD-ROM, 4 Centre nationale de la recherche scientifique (CNRS), 61 CERN see Conseil européen pour la recherche nucléaire CETEDOC (Latin text database), 100 Certeau, Michel de, 84 Châlons-sur-Marne, diocese of, 70 Chalmers, T. M., 37 Charles, Nicholas, 34 charters, 4 Chartres Cathedral, 14 Chaucer On-Line Bibliography, 12 Chen Silin, 91–92 Cheney, C.R. (Christopher Robert), 51 Chortasmenos, John, 91, 92 Christ Pantokrator, monastery of (Constantinople), 90, 91 Chronicle of Higher Education, 22 Chronique Anonyme Universelle, 104–5 Cistern of Aetius, 89, 92 Cistern of Aspar, 89 Cistern of Mocios, 89 cisterns, 89–91 Clanchy, Michael, 35 CMS see content management systems CNI see Coalition for Networked Information CNRS see Centre nationale de la recherche scientifique Coalition for Networked Information (CNI), 20 Cobb, Connor, 82, 83 codex, 3, 77, 94 Column of Constantine, 86 Comma-separated values (.csv) (file format), 56, 61, 102 Committee on Centers and Regional Associations (CARA), 11 Conseil européen pour la recherche nucléaire (CERN), 16 Conner, Patrick, 15

Constantine I, Emperor, 89, 90, 92 Constantine Lips, 90 Constantinople churches in, 90–91 mapping of, 79–80, 80–82, 89–92 monasteries in, 88, 90–92 as palimpsest, 79–80 patronage in, 90–91 residences in, 89–91 scholarship on, 8, 82, 84, 87 walls of, 91, 92 water supply of, 89–91 xenodochia in, 88, 91–92 See also Basilica Cistern; Baths of Zeuxippus; Blachernae Palace; Byzantion; Christ Pantokrator; Cistern of Aetius; Cistern of Aspar; Cistern of Mocios; Column of Constantine; Forum of Constantine; Golden Gate; Golden Horn; Great Palace of Constantinople; Hagia Sophia; Hippodrome; Holy Apostles; Lips Monastery; Neorion harbor; Obelisk of Theodosios; Pammakaristos Church; Platea; Prosphorion harbor; St. John in Petra; Serpent column of Delphi; Xenon of the Kral Constantinople as Palimpsest as classroom tool, 7–8, 78–80, 80–82, 84–86, 88, 89–93 as interactive textbook, 85 as scholarly resource, 78, 80, 82, 87, 88, 94 distinguished from encyclopedia, 86, 93–94 epistemological value of, 93–95 mapping in, 8, 79–80, 81–82, 85, 86–87, 89–90, 91, 94 origins of, 80–85 pedagogical value of, 84–86, 87, 88–89, 92–93 predecessors of, 78ff staffing for, 82, 87 structure of, 86–87 student contributions to, 8, 80–81, 82–83, 84–85, 86, 88, 91 themes in, 79–80, 87 transformative nature of, 88–93



content management systems (CMS), 2, 29, 107 Corpus Christianorum, 100 Corpus Synodalium citation in, 71–72 collaboration in, 56–57, 58 funding for, 53, 55, 68 guiding principles of, 55–56, 59–60 launch of, 68 mapping in, 65–68, 70 origins of, 53–55 platform for, 61–63, 65, 73 publicity for, 68, 72 research uses of, 68–69, 72 scholarly recognition of, 72 searching in, 61–62, 63, 64, 67, 69, 70–71 shortcomings of, 70–72 sources for, 55–60 sustainability of, 72–73 transcriptions for, 56–58, 59–60, 70 value of, 54, 72 Cossart, Gabriel, 51 Cotton Map, 104 Crow, James, 89 CSS see Cascading Style Sheets Cultural Frictions (hybrid conference), 24–25 Curia Irish Text Archive, 20

Dalas-Garigues, Martine, 37 Dante Alighieri, 17, 22 Dartmouth Dante Database, 12, 17, 20, 22 Dartmouth Dante Project see Dartmouth Dante Database data: accessibility of, 38–39, 56, 99, 109 analysis, 4, 7, 99 indexing, 27–28, 102 mining, 3 sharing, 15–16, 99 storage of, 4, 95, 99, 102–3, 104 visualization of, 15 See also comma-separated values; databases; metadata; plain text; programming, computer; Text Encoding Initiative databases, 1, 5, 12–13, 16, 17, 20, 25, 26, 27, 29, 36–37, 60, 71, 100, 101, 102, 108–9

index

115

management systems for, 37–38, 42, 43, 61, 79, 102, 103, 104–5, 107 See also ARTFL Project; Bodleian Library; CETEDOC; Chaucer On-Line Bibliography; Corpus Synodalium; Curia Irish Text Archive; Dartmouth Dante Database; DIGISIG; Digital Scriptorium; Electronic Archives for Teaching the American Literatures; Fragmentarium; MALIN History Text Archive; Omeka; Online Catasto; Oxford English Dictionary; Oxford Text Archive; PhiloLogic; Seals in Medieval Wales; Society for Early English and Old Norse Electronic Texts; StoryMaps; Studies in the Age of Chaucer Bibliography; University of Virginia Electronic Text Center Davis Foundation, 82 Davis, Lisa Fagin, 6, 8, 99–111 DeGioia, John, 18n14 Diemer, Caroline, 83 Dierker, Lisa, 82, 84 Digard, Françoise, 36–37 DigiPal (paleographic image annotation software), 103 DIGISIG (sigillography database), 7 as comparative tool, 34 challenges to, 44–45 goals of, 33–34, 46 growth of, 41–42 launch of, 42 origins of, 38–39 redesign of, 43–44 searching in, 42–43, 45 sources for, 39–41 digital repositories, 1, 12–13, 20, 37–38, 77 Digital Documentation Process, 6, 9, 112 DigitalMappa (text and image annotation platform), 104–5 digital scholarship aesthetics in, 14, 81 access to, 18 authorship in, 15–16 and canon law, 53–73 collaboration in, 7–8, 11–12, 15–16, 17, 23–25, 56, 58, 63, 78, 82–84, 86–87, 88–93, 94–95, 100, 103, 108–9 definition of, 99, 109

116

index

discomfort with, 2, 7, 9, 20–21 early history of, 6–7, 11–28, 36–37, 100 editions in, 58–59, 60, 104–5 impermanence of, 5, 26, 38, 55, 73, 95, 99, 104–5 leading role of medievalists in, 3–4, 6–7, 12, 16, 77, 99, 109 and manuscript studies, 100–9 mapping in, 4, 5, 19, 65–68, 78, 79–80, 84, 86–87, 89–90, 91, 94 and pedagogy, 1, 7–8, 20–21, 23, 25, 29, 53, 55, 78–95 perceived unreliability of, 5, 104–5 perception of as “science”, 3 process-oriented nature of, 6, 94–95 research methods in, 1, 4, 9, 36–37, 61, 99–100 rooted in older forms of learning, 3–4, 5, 8, 9 role of narrative in, 5 resources for, 12–13, 16, 18–19, 20, 22–23 scholarly recognition of, 9, 72, 83 and sigillography, 7, 33–48 sustainability of, 2, 5, 6, 9, 29–31, 55–56, 72–73, 99–100, 103, 109 text analysis in, 7, 53–55, 58–59, 63, 71 tools of, 3–4 transformative potential of, 5, 6, 14–16, 36–37, 42, 46, 54, 68–69, 73, 77, 88–93, 95 as user driven, 4, 7, 8 values of, 2, 17–18, 28–29, 55–56, 83–84, 94–95, 99–100, 109 See also databases Digital Scriptorium, 101 digital video disk (DVD), 104–5 diocesan statutes see canon law, episcopal Dioscorides, 91–92 discussion fora (online), 12, 13, 14–15, 16, 17, 20 Diver, Kim, 82 Django (web framework), 43 Dorin, Rowan, 7, 49–75 Doubleday, John, 34 Dugdale, William, 34 Duncan, Edwin, 12–13 Durand, William, 68

Durham Cathedral, 35 Duschnes, Phillip, 107 DVD see digital video disk

EDSITEment directory (NEH), 25 Ege, Otto F., 107, 108 Electronic Archives for Teaching the American Literatures (website), 17 Ellis, Roger H., 36, 40, 42, 45 Emerson College, 83 Encyclopedia Britannica, 22, 93 encyclopedias, 87, 93–94 England canon law in, 51–52, 54; seal use in, 33–38, 43–44 Environmental Systems Research Institute (Esri), 79 Eren, Alp, 79, 80, 82, 83 Esri see Environmental Systems Research Institute Eusebius of Caesarea, 8, 77, 94 Everhart, Deborah, 6–7, 8, 11–32 Excel see Microsoft Excel expulsion, 49–50, 69

FACE Foundation, 56–57, 58 File Transfer Protocol (FTP), 12 Finlayson, Marjahn, 82 flâneur, 84 florilegia, 94 Forum of Constantine (Constantinople), 89 Forum of Theodosios (Constantinople), 90 Foys, Martin, 104 Fragmentarium (manuscript fragment database), 108–9 fragmentology, 105–9 France-Stanford Center for Interdisciplinary Studies, 57 Franklin-Lyons, Adam, 83 FTP see File Transfer Protocol

García y García, Antonio, 52 Garsten, monastery of, 101, 102 Geneva, 105 Geographic Information System (GIS), 65, 82 Georgetown University (Washington, DC), 12, 17–18, 21, 26, 30 Georgetown University Guide to World Wide Web Development, 21



See also Cultural Frictions GIS see Geographic Information System Glossa ordinaria, 78 glossing, 3, 78 Golden Gate (Istanbul), 79 Golden Horn, 91 Google (search engine), 1 Gopher (internet protocol), 14, 16, 18, 24, 25 Gottschalk Antiphonal, 107, 109 Gottschalk of Lambach, 100, 102, 107 Göttweig, monastery of, 101 Grafton, Anthony, 94 graphemes, 100, 102 Great Palace of Constantinople, 90 Greenwell, William, 45

Hagia Sophia, Cathedral of (Constantinople), 79 Harvard Society of Fellows, 53 Harvard University, 107 See also Harvard Society of Fellows; Houghton Library Harvey, Paul, 33, 35, 37–39, 40, 42, 44, 45–46 HathiTrust, 1 Hatton, Christopher, 34 Hebbard, Elizabeth, 107 Henry III, King of England, 35–36 heresy, 70 Herlihy, David, 3–4 Herrin, Judith, 88 Heslop, T.A., 44 Hill Monastic Manuscript Library (Minnesota), 101 Hippodrome (Constantinople), 79, 89 Holy Apostles, church of the (Constantinople), 90 Houghton Library (Harvard University), 107 HTML see hypertext markup language hyperlinks see hypertext hypertext, 3, 5, 15–16, 18, 22 hypertext markup language (HTML), 18, 21, 30, 32, 42

IBM see International Business Machines Corporation

index

117

IIIF see International Image Interoperability Framework Index Thomisticus, 99 Informix Corporation, 37–38 International Business Machines Corporation (IBM), 99 International Congress of Medieval Canon Law, 56 International Congress on Medieval Studies (Kalamazoo), 11, 12, 16, 24, 107 International Image Interoperability Framework (IIIF), 108–109 International Medieval Congress (Leeds), 103 Internet Archive, 1, 30 internet protocols, 14 Interscripta, 14–15, 22 Irvine, Martin, 6–7, 8, 11–32 Istanbul, 79, 80 Janin, Raymond, 82, 88 JavaScript (programming language), 42 Jenkinson, Hilary, 34, 46 Judaism, 70 Justinian I, Emperor, 92

Kalamazoo see International Congress on Medieval Studies Kings College Digital Laboratory, 103 Klapisch-Zuber, Christiane, 4 Kremsmünster, monastery of, 100, 101 Krieger, Nathan, 82, 83

labyrinths, 13, 14 The Labyrinth (website) advantages of, 16–17, 20 aesthetics of, 14 archiving of, 29–31, 112 audiences for, 23–25 as cataloguing site, 27–28 code for, 18–19 collaboration in, 8, 11–12, 15–16, 17, 23–25, 26, 29, 32 contents of, 16–17, 20, 23–25, 26–27, 28 contributors to, 20–21, 26, 28 funding for, 17–18, 26, 29 growth of, 19–20, 22–23, 28–29, 31 as heuristic system, 17, 27–28

118

index

influence on other scholarly websites, 17, 20–21, 31–32 launch of, 16, 19–20, 31 as media-rich resource, 19, 22 narrowing of scope of, 29–30 navigation of, 13, 14, 18, 27 non-commercial nature of, 17–18, 28–29; origins of, 11–19 pedagogical role of, 20–21, 23, 25, 29 pioneering nature of, 6–7, 11, 12, 16 presentations on, 12, 14–16, 20–21, 31 press coverage of, 13–14, 22 publications about, 24–25 recognition of, 17n13, 22, 25 resource creation for, 20–21, 26 restructuring of, 27–28 scaling of, 25–26 Labyrinth Guide to Manuscript Repositories, 24 Lambach, monastery of, 100, 101, 102, 107 Laon, diocese of, 70 Lateran Council, Fourth, 49, 50 Lawton, Robert, 18n14 Library of Congress Classification (LCC), 27 Lips Monastery (Constantinople), 90 listservs see discussion fora (online) Litchfield, Robert Burt, 4 London Metropolitan Archives, 35 Lyon, Second Council of (1274), 49, 59, 69 Mainz, 65 ecclesiastical province, 65 MALIN History Text Archive, 20 Mansi, Giovanni Domenico, 50, 71 manuscripts, 1, 3, 19, 24, 91–92, 99, 100–102, 103–4, 107 digitized, 1, 19, 22, 101, 104–5, 107–9 See also fragmentology mapping, 4, 5, 19, 65–68, 78, 79–80, 84, 86–87, 89–90, 91, 94 See also ArcGIS; StoryMaps marriage, 50 Massachusetts Institute of Technology (MIT), 22 McDonough, Katherine, 61 McEwan, John, 7, 33–48 McGuinness, Andrew, 33 McMullen Museum of Art (Boston College), 103

Medieval Academy of America, 11, 21, 27, 104 Medieval Studies, 5, 6, 8, 18, 20, 23, 24, 27, 31, 77, 99 Medievalist’s E–Resources Directory, 12–13 Melk, monastery of, 100, 101 Mende (diocese), 68 Mercier, Dan, 84 metadata, 27–28, 56, 63, 71, 101 metaphor, 13–14, 15–16 Microsoft Access (MSAccess) (database management system), 38, 102 Microsoft Excel (spreadsheet software), 102 Microsoft SQL Server (database management system), 38 Microsoft Windows (operating system), 38 Microsoft Word (word processing software), 1 Milton see William F. Milton Fund Mirador (image viewer), 108 Moelho, Anthony, 4 monasteries, 78, 88, 92, 100, 101 See also Admont; Garsten; Göttweig; Kremsmünster; Lambach; Melk; St. John in Petra; St. John in Stoudios; St. Lambrecht; St. Martin of Tours; St. Paul–im–Lavanttal moneylending, 49 Mosaic (web browser), 19 Müller-Wiener, Wolfgang, 88 Museum of Fine Arts (Boston), 103 MySQL (database management system), 42 Naing, Thawsitt, 63, 65, 67, 72 National Archives, The (Kew), 34, 35, 37–39, 46 National Archives of Scotland (Edinburgh), 39 National Endowment for the Humanities (NEH), 17, 19, 22, 25, 104 See also EDSITEment directory National Library of Wales (Aberystwyth), 35, 39 necropolitics, 84 Neorion harbor (Constantinople), 90 Netzer, Nancy, 103 Norton, W. W. (publisher), 22 Noyon, diocese of, 70 Nurjadin, Nadja, 89



Obelisk of Theodosios (Constantinople), 86 OCR see Optical Character Recognition Omeka (content management system), 107 Online Catasto (website), 4 open-source code, 108 Optical Character Recognition (OCR), 57, 58 OTA see Oxford Text Archive Ovidius Naso, P. (Ovid), 13 Oxford Dictionary of Byzantium, 88, 93 Oxford English Dictionary, 20 Oxford Text Archive (OTA), 12–13, 20 palaeography, 100–109 palimpsest, 79 Pammakaristos Church (Constantinople), 91 Passau, diocese of, 101 Pastoureau, Michel, 37 Patrologia Latina, 100 PDF see Portable Document Format Petra (quarter of Constantinople), 90, 91 See also St. John in Petra PhiloLogic (database search and reporting system), 61, 67, 73 Photogrammetry, 35 PHP (web scripting language), 42 plain text (.txt) (data format), 56, 61 Plakidas, Konstantinos, 82 Platea, 91 Poe, Edgar Allan, 84 Pontal, Odette, 51 Portable Antiquities Scheme, 39 Portable Document Format (PDF), 109 PostgreSQL (database management system), 43 preservation, digital scholarly, 2, 5, 6, 29–30 See also Digital Documentation Process; digital scholarship, sustainability of; Archiving Dossier Narrative print technology, 3 PRO see Public Record Office programming, computer, 42, 43, 60, 61, 63 Prosphorion harbor (Constantinople), 90 Public Record Office (Kew), 34 See also The National Archives Python (programming language), 43, 61, 63 R (programming language), 61 reclusion, 68

index

119

Reflectance Transformation Imaging (RTI), 35 Reims, ecclesiastical province of, 70 Robinson, Gavin, 57–58 Romani, Clara, 67 rune stones, 22 Rusche, Philip, 107

St. John in Petra, monastery of, 88, 92 St. John in Stoudios, monastery of, 78 St. Lambrecht, monastery of, 101 St. Louis, Missouri: Public Library, 107 St. Louis University, 108 St. Martin of Tours, monastery of, 78 St. Paul-im-Lavanttal, monastery of, 107 Sawicki, Jakub, 52 Schipper, William, 15, 24, 26 Schoenberg Institute for Manuscript Studies (University of Pennsylvania), 105 Schwandt, Silke, 95 Schwartz, Richard, 18n14 scrolls, 3 seals, medieval casts of, 34, 36, 37 catalogs of, 34–35, 36, 37, 38–40, 42–43, 44, 46 changing use of, 43–44 classification of, 37, 45 collections of, 35–36, 37, 39 dating of, 45 description of, 34, 40–41, 43, 45 design of, 41, 43–44 digital analysis of, 33, 35, 36–38, 42–43, 44 English, 33–34, 35–36, 37–38, 43–44 fragility of, 36 French, 37, 40 heraldic, 44, 45–46 identification of, 33, 35–36 impressions of, 7, 33, 34, 35, 36, 40 indexing of, 7 matrices, 7, 33, 35, 40 motifs in, 41 photographic images of, 34–35 reconstruction of, 36 royal, 35–36, 40 uses of, 33, 35 Welsh, 38–39, 40–41 See also sigillography

120

index

Seals in Medieval Wales (SIMEW), 37, 38–39, 40–41 Senlis, diocese of, 70 Sermones super Canticum Canticorum, 101 Serpent column of Delphi (Constantinople), 86 Seybold, Nancy, 107 Shannon-Dabek, Nadja, 82 Shaw, Gary, 83 Shoaf, R. Allen, 26 Sicut olim (conciliar decree), 49–50 Sigibold, Abbot of Lambach and Melk, 101 SIGILLA, 40 sigillography, 7, 33–46 digital tools for, 35, 36–43, 44, 46 early history of, 34 reference works for, 34–35 See also DIGISIG; Seals in Medieval Wales; SIGILLA Simms, Jason, 82 Simpson, Jesse, 83 Skolnik, Jonah, 83 Smartware II (office suite), 37–38 Society for Early English and Old Norse Electronic Texts (SEENET), 12, 20 Soissons, diocese of, 70 Stanford University, 11, 22, 55 See also France-Stanford Center for Interdisciplinary Studies Stefan Milutin, King, 88 Sterling Memorial Library (Yale University), 100 Stokes, Peter, 103 StoryMaps (mapping content management platform), 79, 87 Studies in the Age of Chaucer Bibliography, 20 Suda, 93 TEI see Text Encoding Initiative Telnet (internet protocol), 12, 16, 18, 24 Text Encoding Initiative (TEI) (XML file format), 60, 63, 109 Theodosius I, Emperor, 90, 92 Thérouanne, diocese of, 70 Thornton, Melissa, 89 TNA see National Archives Torgerson, Jesse, 7–8, 77–97 Traveler’s Lab (Wesleyan University), 79, 83

Tournai, diocese of, 70 Tzetzes, John, 91

Uniform Resource Locator (URL), 71–72 University of Chicago, 61 University of Pennsylvania, 105 University of Virginia Electronic Text Center, 20 URL see Uniform Resource Locator usury, 7, 49–50, 69

Washington Post, 22 The Water Supply of Byzantine Constantinople, 89 web browsers, 19 web servers, 27–30 Web Works, 21 Wesleyan University, 79, 82 Center for Pedagogical Innovation, 84 See also Traveler’s Lab Westminster Abbey, 35, 39 White, Stephen, 4 Wikimedia Commons, 82 Wikipedia, 86, 93 William F. Milton Fund, 53, 68 Williams, Megan, 94 WordPress (content management system), 29 word processing, 1 See also Microsoft Word World Wide Web (W3), 11, 14, 15, 16 access to, 18–19 as collaborative tool, 16–17, 23, 24–25 frameworks for, 43 growth of, 22, 28, 31 navigation of, 18, 71–72 non-hierarchical nature of, 25–26 site development in, 20–21, 31 See also American Standard Code for Information Interchange; Django; Gopher; hypertext markup language; Mosaic; PHP; Telnet; Uniform Resource Locator; web browsers; web servers; World Wide Web Virtual Library World Wide Web Virtual Library, 22 Wyckoff, Chris, 89

xenodochia, 88, 90–91 Xenon of the Kral (Constantinople), 88, 91–92



Yale University, 100, 104, 107 See also Beinecke Rare Book and Manuscript Library YouTube, 2

Zeyrek Camii see Christ Pantokrator Zheng Mao, 89 Zotero, 1

index

121