The main entrance to the Van Pelt-Dietrich Library Center is now open. Van Pelt Library and the Fisher Fine Arts Library are currently open to Penn Card holders, Penn affiliates, and certain visitors. See our Service Alerts for details.

  • Symposium

Hooking Up

12th Annual Lawrence J. Schoenberg Symposium on Manuscript Studies in the Digital Age

This year’s symposium explores the connections between historic and current approaches to data linkage in regard to manuscripts and manuscript research.

This event has already occurred

November 21-23, 2019
Kislak Center, 6th Floor Van Pelt-Dietrich Library
Open to the Public
A chromatic scale diagram from a late 15th century copy of Boethius' De institutione musica (LJS 47, fol. 41v), merged with a detail of the linked data model of the Mapping Manuscript Migrations Project

In partnership with the Rare Book Department of the Free Library of Philadelphia, the Schoenberg Institute for Manuscript Studies (SIMS) at the University of Pennsylvania is pleased to announce the 12th Annual Lawrence J. Schoenberg Symposium on Manuscript Studies in the Digital Age.

The concept of linked open data is the holy grail of the digital humanities. Yet the problem of how to link information across platforms has existed since civilization began. As knowledge and learning expanded in premodern society, the problems associated with collecting, combining, and disseminating information inspired new approaches to and technologies for the material text. In the internet age, we continue to grapple with the same problems and issues. While technologies have changed, the questions remain the same.

This year’s symposium explores the connections between historic and current approaches to data linkage in regard to manuscripts and manuscript research. Hooking Up addresses the topic from a variety of angles and considers how the manuscript book operates as a vehicle for information retrieval and dissemination from the technology of the page and the textual apparatus of a book, to the library, and finally, the internet. We will also consider such questions as how medieval practices of memory shaped information retrieval and gathering, how did the technology of the manuscripts book—in all its many forms—facilitate or hinder information processing, how can medieval solutions inform modern technologies, and how do modern technologies illuminate medieval practices? The program will also feature sessions highlighting projects that are advancing linked data technologies for manuscript researchers, including the T-AP Digging Into Data Challenge project Mapping Manuscript Migrations.

The program will begin Thursday evening, November 21, 5:00 pm, at the Rare Book Department of the Free Library of Philadelphia, Parkway Central Library, with a keynote address by Professor Mary Carruthers, New York University, and All Souls College, Oxford University. The symposium will continue November 22nd-23rd at the Kislak Center of Special Collections, Rare Books, and Manuscripts at the University of Pennsylvania.

Accordion List

  • Benjamin L. Albritton, Stanford Libraries
  • Toby Burrows, e-Research Centre, Oxford University,
  • Matthew Driscoll, Arnamagnæan Institute, University of Copenhagen
  • Christoph Flüeler, University of Fribourg
  • Katarzyna Anna Kapitan, Museum of National HistoryFrederiksborg Castle & Arnamagnæan Institute, University of Copenhagen
  • Mikko Koho, Semantic Computing Research Group, Aalto University
  • Jehnna Lewis, University of Pennsylvania
  • Megan C. McNamee, Warburg Institute
  • Aylin Malcolm, University of Pennsylvania
  • David R. Nelson, University of Pennsylvania
  • Sally Ragep, Institute of Islamic Studies, McGill University
  • Lynn Ransom, Schoenberg Institute for Manuscript Studies
  • Helmut Reimitz, Princeton University
  • Linda Safran, Pontifical Institute of Mediaeval Studies, Toronto
  • Emily Steiner, University of Pennsylvania
  • Elly Truitt, Bryn Mawr College
  • Kelly Tuttle, University of Pennsylvania Libraries
  • Nancy Um, Binghamton University
  • Athanasios Velios, Ligatus, University of the Arts London
  • Hanno Wijsman, Institut de recherche et d'histoire des textes
  • Jeffrey Witt, Loyola University Maryland
  • Elizabeth Yale, University of Iowa
  • Kıvılcım Yavuz, Kenneth Spencer Research Library, University of Kansas

Event Series



Accordion Column of Lists

Mary Carruthers, New York University and All Souls College, Oxford University

In this lecture I will explore the close relationships in medieval creative practices among geometric shapes and the human ability to craft original works. Focused on materials crafted in the twelfth and thirteenth centuries from the rich data base of literary texts for the rhetorical and logical arts of Invention, and then disseminated widely during the following centuries, my talk investigates the fundamental cognitive insight of medieval diagram makers, that not only envision what we already know, but invite us to discover surprising logical relationships that can provoke our thinking in new

Linda Safran, Pontifical Institute of Mediaeval Studies, Toronto

A close examination of Byzantine manuscripts reveals numerous features relevant to contemporary discussions about linked data. If the main text on a page is a platform for knowledge, the paratexts that surround (or invade) it constitute additional platforms. These include scholia, glosses, and diagrams—a bounty of data on a single page, often communicated in multicolored inks. There are also interventions across pages: marginal bookmarks that link disparate passages, pages of diagrams separate from their relevant texts, originally independent notes bound into a volume. Such complex page and book layouts could facilitate or hinder the processing of information, in the Middle Ages no less than today.

Aylin Malcolm, University of Pennsylvania

Newly acquired from Conception Abbey in Missouri, UPenn MS Codex 1881 contains several of the most widely disseminated astronomical texts from late medieval Europe, including a glossed, annotated, and illustrated copy of Johannes de Sacrobosco’s De sphaera mundi. This short text on the fundamentals of Ptolemaic astronomy, popular among students and amateurs alike, is here presented alongside an unusually rich variety of diagrams, ranging from marginal sketches to skillfully designed volvelles. Many of these diagrams contain large amounts of text, fostering careful study of the main text, and enabling us to interpret them as encyclopedic rather than merely schematic. Taking this fifteenth-century copy of De sphaera as my anchor for this paper, I explore the contexts and uses of diagrams in late medieval astronomical manuscripts from several connected perspectives. I consider the reading practices encouraged by diagrams that may supplement, amplify, or even contradict the text, as well as how these relationships may vary among copies of De sphaera. Focusing on diagrams also opens up questions pertaining to interactions between manuscripts and early printed books. The diagrams that typically accompanied De sphaera provided impetus for early innovations in colour printing, but the more complex among these diagrams continued to be hand-copied into the sixteenth century, since manuscript diagrams could be more flexible in their designs and uses. Dating to approximately fifteen years after the first print editions of De sphaera, MS Codex 1881 provides a particularly useful case study of the uneven transition between book production methods. Finally, I address the affordances and challenges of producing digital versions of medieval scientific diagrams, including the interactivity and information linking offered by online editions.

Megan C. McNamee, The Warburg Institute

Oxford, Bodleian Library, MS. Rawlinson D. 939, is an accordion-fold almanac made in England after 1389. It belongs to a class of familiar yet little-studied manuscripts known as vade mecum (Faltbuch or livre plicatif), which do not adhere to the traditional codex format, rather they comprise a small number of parchment leaves folded into compact packets according to a variety of possible patterns. Although this kind of book may have been common, few survive intact; fewer still maintain their original structure. Their flexible form made them especially vulnerable to wear and, as portable objects—some retain cords by which they might hang from a belt and protective cases—they were exposed to all the dangers of a vagabond life. Such was the fate of MS. Rawl. D. 939. The long parchment sheets that compose the manuscript, once joined, are now separated. This paper reunites them and examines the ways that the anonymous maker(s) knit a farrago of calendrical, chronological, devotional, medical, economic, and prognostic material all loosely related to time into a manuscript that stretches an astonishing two and a half meters when unfolded and collapses to the size of an iPhone. Following Gilles Deleuze, who demonstrated the conceptual potential of the fold (le pli), I show how MS. Rawl. D. 939 implicates, explicates, and complicates it contents. In this late-medieval context, the fold is, above all, an instrument of analogy (analogia), linking seemingly disparate elements and reifying connections that might be dismissed as miscellaneous or incidental in a codex.

Kelly Tuttle, University of Pennsylvania Libraries

This paper will look at the textual afterlife of a popular, but rather arbitrarily organized commentary, al-Ghayth al-musajjam, by Khalīl ibn Aybak al-Ṣafadī (d. 1363). The commentary is arranged according to the associations that came to the author’s mind as he was composing, which means that as a text it is good for browsing, but bad for using. Later abridgers edited al-Ṣafadī’s commentary to streamline it and make it more user-friendly, turning the source-text, a 12th century poem called Lāmīyat al-ʿAjam, into the focus of the work, rather than leaving it as it had been in the Ghayth, namely a scaffolding upon which to hang the basics of what an educated person in the 14th century should know. While doing this, however, later editors also cut out the wandering, inquisitive aspect of the work, rendering it perhaps more user-friendly but not particularly engaging. In other words, they unlinked most of the data. Rather than being turned loose in a rabbit warren of interconnected ideas, the reader of these edited commentaries is taught only what the source poem means on a surface level and how the grammar works. In the Ghayth, the focus is not on the source-text, but on the literary culture of the time and how it can be drawn out of the source-text. Later readers of the abridgements, though, seem to have picked up on this oversight and relinked earlier data found in the Ghayth. In later copies of the abridgements, there appear, as marginalia, some of the lessons and observations found in the Ghayth. This linking, unlinking, and relinking of the data provided by al-Ṣafadī, taken out and put back in by later readers, begins to provide more insight into the shifting commentary practices during the late Mamlūk and early Ottoman eras.

Emily Steiner, University of Pennsylvania

The thirteenth and fourteenth centuries witnessed the rise of the natural encyclopedia, both in Latin and in the vernaculars. This paper explores the creative ways in which medieval encyclopedias link information.

Helmut Reimitz, Princeton University

About a quarter century ago I started to work on historiographical compendia from the early medieval, respectively the Carolingian, period. It was great fun to explore how well-researched and well-known texts became much less familiar in these compendia. What I found particularly exciting however, was to explore the social logic of some of these compendia and how this social logic also affected how the texts were rewritten, re-interpreted, and passed on to later generations who were quite capable of navigating between different social logics in their study, use and production of history books. At the time I thought that many late Antique and early medieval scholars would soon work along the same lines and we would soon have a much better sense for the spectrum of possibilities regarding the rewriting of older texts and the writing of new ones. That did not happen, and in the course of my talk it will become clear why but also why I think that new digital tools might help us to finally move on. I will start with a brief introduction into the study of history books with some examples from the Carolingian and post-Carolingian period. The examples shall help me to illustrate the potential of exploring the social logic of history books not only for the study of single texts, or histories but also for the history of the book and the libraries to which they belonged. As someone who has not very much experience or training in working with digital tools, I would like to end with some ideas for a future collaborative project whose employment of digital tools might help us to arrive at a quantitative and qualitative level of research that would allow us to situate the production of historiographical compendia within the larger cultural developments and changes in the transitional period between the two medieval ‘Renaissances’ in the 9th and in the 12th century.

Elly R. Truitt, Bryn Mawr College

My talk explores the importance of two different information technologies, the codex and the clock, in making manifest expressions of Christian universality, and in fostering new kinds of historical narration. I suggest that we understand the rapid proliferation and rapidly intensifying complexity of mechanical clocks in the fourteenth and fifteenth centuries as the response to new technological possibilities for calculating, delineating, and demonstrating universal time. Both the codex and the clock allow us to consider the terms of Latin Christian universality, particularly as that concept relates to understanding time and defining community. The codex and the mechanical clock fulfilled similar roles in Latin Christianity in northwestern Eurasia over the course of late antiquity and the medieval centuries. Beginning in the third and fourth centuries of the Common Era, the codex became the dominant form of recording and storing the Christian scriptures; moreover, early Christian authors like Origen, Eusebius, and Jerome used the format of the codex to organize and present information about the past in a radically different fashion from the scrolls of antiquity. The form of the codex and the ideological convictions of early Christians also gave rise to a new kind of historiography—the universal chronicle. Early medieval kingdoms of Francia and its neighbors used the universal chronicle as a way to establish ties a longer, encompassing genealogy of Christian rulers and ancient empires. The annals and chronologies of the ninth and tenth centuries were critical to the changes to universal historical chronicles, especially the turn toward the vernacular in the thirteenth centuries with the rise of historiated Bibles and the Histoire ancienne jusqu’à César. Yet when the mechanical escapement appeared in far northwestern Latin Christendom at the turn of the fourteenth century, it was soon mobilized to depict universality in a Christian context, and astronomical mechanical clocks—huge machines with multiple complications, carillons, and automata—appeared as a new incarnation of universal Christian time and community. Monumental clocks became the medium to portray multiple temporal narratives and timelines (sacred, secular, cosmic, daily) within one unified whole.

Elizabeth Yale, University of Iowa

In the symposium description, the organizers of “Hooking Up” ask us to consider how “the technology of the manuscript book” has served as a vehicle for linking data and organizing knowledge. In effect, they ask us to imagine a usable past for information science’s digital present. In this light, I was intrigued by the relative absence of people in the organizers’ questions: they focus more on technologies and practices. This is an understandable consequence of how little it is often possible to know about individuals who made, used, and kept manuscripts. We are also often interested in questions transcending any one individual. However, when creating a usable past for how we manage, communicate, and preserve information, it is important to keep our eyes on the people, where we can. In my paper, I explore women’s roles in scientific and household record-keeping and communication in early modern Britain. Through this work, women made knowledge, as well as homes. If this surprises us, this may be in part because we have a habit (even when well-meaning, even when acting in a scholarly habitus) of effacing people from paper-keeping.

Accordion Column of Lists

Jehnna Lewis and David R. Nelson, University of Pennsylvania

In 1696, Francis Daniel Pastorius, the founder of Germantown in Pennsylvania, began work on an English-language commonplace manuscript, which he called his “Bee-Hive.” The manuscript, Ms. Codex 726 in the University of Pennsylvania’s Kislak Center, represents a significant account of early Philadelphia. The complexities of Pastorius’s beehive metaphor are interwoven throughout the entire work, as he refines and redefines the ways in which the physical construction of a beehive informs the construction of his own “Paper-Hive.” Due to his comprehensive documentation of his source material, the manuscript has previously received scholarly attention primarily as a bibliography of printed materials available in early America, especially in the Quaker context. However, for Pastorius, documenting these materials was only the first step in the construction of the manuscript. In his hive metaphor, these books serve as flowers, the quotations he collects, pollen. His commonplacing reveals the work of the hive. The quotations are interwoven conceptually into discrete honeycombs, each of which is explicitly linked to other semantically relevant combs. Starting with fine European paper and quickly settling for what early America could provide, Pastorius built his hive out of folios that he arranged and rearranged as the internal logic of the manuscript expanded. The largest section of the manuscript is comprised of so-called “Hive-Dross” (propolis), thousands of entries demarcated by single words or short phrases. Pastorius thus organizes this material into two large sections, one organized alphabetically by concept and one organized with arbitrary numerical headers. Yet this part of the manuscript also draws connections to its other sections, which are organized by diverse structural principles. In our talk, we will introduce a dynamic digitization project that presents Pastorius’s nexuses of entries, particularly in the “Hive-Dross”, as Linked Data. We will discuss not only how this project shows a static understanding of Pastorius’s cross-references between the entries, but also can be used to reveal Pastorius’s diachronic development of the Hive and his own network of information, within and beyond the manuscript. We will also discuss the particular challenges this project faces with regard to linking and interfacing diverse information structures both on paper and on the web.

Matthew Driscoll, Arnamagnæan Institute, University of Copenhagen

One of the more astonishing bibliographical undertakings of the early modern age was the project by Hernando Colón (1488-1539), son of the famous navigator Christopher Columbus, to build a universal library, one which would contain “every book on every subject and in every language, from within Christendom and without”. In the end his library in Seville comprised over 15000 volumes, the largest collection of printed books in the world at the time. To manage it all he designed a revolutionary system for storing, sorting and distilling these materials, an information-crunching engine of great power. In addition to keeping a register of purchases (including those which had been lost on route to Seville), Colón conceived four types of inventories: an abecedarium, i.e. a list of authors in alphabetical order; a book of sciences, that is, subjects; a book of materials, what we now call keywords; and finally a book of epitomes, which contained detailed summaries of the contents of each of the books in the collection, all carefully cross-referenced. The first three of these survive in the Biblioteca Colombina in Seville; the fourth, El libro de los epítomes, was presumed missing until recognized recently in the Arnamagnæan Collection in Copenhagen. In my paper, I will present Hernando’s “bibliographical tools” as an early example of linked data.

Sally Ragep, Institute of Islamic Studies, McGill University

The Islamic Scientific Manuscripts Initiative (ISMI) originated when we ran out of enough notecards to capture the metadata of tens of thousands of manuscript witnesses of Islamic astronomy, not to mention their multifaceted content and interconnections. Over the past twenty years, we’ve been able to store a lot of that information in increasingly sophisticated databases that we can retrieve in manifold ways, some that once seemed inconceivable. But we have also learned the limitations of digital humanities in dealing with the premodern world of scholarship, which itself was remarkably interrelated, interconnected, and intertextual, so much so that it often defies even our sophisticated technologies. To illustrate these points, I will use the example of a single Islamic astronomical textbook that circulated for over 8 centuries and gave rise to some 65 derivative texts. Numbering in the thousands, these extant manuscript witnesses are a gateway to the scientific culture of the Islamic world from the early thirteenth to early twentieth century. However, the amount of data embedded within these manuscripts presents serious obstacles from both technical and research perspectives. I look forward to stimulating exchanges with participants who face similar challenges in their own areas of interest.

Katarzyna Anna Kapitan, Museum of National History, Frederiksborg Castle & Arnamagnæan Institute, University of Copenhagen

Drawing on the cataloging experience for, the online catalog of Nordic manuscripts curated collaboratively by three institutions in Iceland and Denmark (the Arnamagnæan Institute in Copenhagen, the Árni Magnússon Institute for Icelandic Studies in Reykjavík, and the National Library of Iceland), and the Stories for all time catalog (, a thematic catalog of manuscripts preserving Icelandic legendary sagas in 23 repositories in Europe and North America, this paper addresses two aspects of digital cataloging. The first one is the function and purpose of a digital catalog in relationship to the recent developments in the fields of material philology and history of the book. The second one is the importance of linked data in digital cataloging of manuscripts and quantitative analysis of manuscript descriptions. This paper asks questions which may be relevant not only to cataloging Nordic manuscripts but also to other digital catalogs, such as: What purpose a digital catalog is supposed to serve and how and by whom it is supposed to be developed and used? Is it a starting point for philological and codicological research or is it a constantly updated record of the state of the scholarship on a given manuscript?  Is the collected data supposed to be analyzed manually, or it is supposed to be a basis for computer-assisted quantitative macro analysis of manuscript descriptions in order to answer new research questions? 

Kıvılcım Yavuz, Kenneth Spencer Research Library, University of Kansas

Recent decades have seen, particularly in Europe and North America, significant efforts to make information available on the digital domain with regard to ancient, medieval and early modern manuscripts as well as early printed books. Many archives and libraries have provided digital facsimiles of their manuscript holdings as well as making their manuscript catalogs digitally available. Recent decades have also seen an increasing interest in the materiality of manuscripts in scholarly circles, which grew from different theoretical approaches to medieval texts, such as transmission history, intertextuality and new/material/artifactual philology. Scholars began paying attention to not only the texts contained in manuscripts but also the mechanisms of the production, dissemination and reception of texts as material objects. 

Taking into account the opportunities and challenges presented by digital tools on the one hand and the changing demands of manuscript scholarship on the other, this paper considers contemporary manuscript cataloging both in principle and in practice. By surveying the progression of cataloging from simple booklists in early medieval manuscripts to library catalogs in early modern manuscripts, modern printed catalogs and contemporary digital catalogs, the paper reflects on to what extent concepts relating to cataloging manuscripts has changed (or remained the same). Focusing on the shift to the digital domain, the paper asks: To what extent we have access to manuscripts in the digital age? Do digital tools make manuscript data more discoverable? Do the metadata match the artifacts? What do linked data look like in the context of manuscript cataloging? To what extent digital catalogs work together with digital facsimiles? 

This session will report on the work carried out by the Mapping Manuscript Migrations (MMM) Project between 2017 and 2019. Funded by the Trans-Atlantic Platform under its Digging into Data Challenge, the MMM Project has aggregated data from three major manuscript-related datasets into a Linked Data environment. It provides a user interface for exploring research questions related to manuscript histories and provenance. The user interface will be publicly launched at the Symposium. Representatives from the MMM partner institutions will talk about and demonstrate different aspects of this complex project.

  • Toby Burrows (Oxford e-Research Centre, University of Oxford) will introduce the project, discuss its aims and outcomes, and assess its contribution to the world of Linked Data and medieval manuscript research.
  • Mikko Koho (Semantic Computing Research Group, Aalto University) will present the project’s data model, its pipeline for transforming the different datasets, and the processes used in reconciling the names of persons, places, works, and manuscripts.
  • Lynn Ransom (Schoenberg Institute for Manuscript Studies)will examine the way in which research questions have been developed and refined in order to shape the design of the user interface, and will reflect on the need to encourage transparency between the interface and the data.
  • Hanno Wijsman (Institut de recherche et d’histoire des textes, Paris) will discuss the data transformation and reconciliation processes from the point of view of a major contributor of data, in the form of the Bibale database.

Jeffrey C. Witt, Loyola University Maryland

The medieval scholastic commentary tradition is a vast network of citations, references, and commentaries. But this is often very hard to see. The traditional practice of quoting and citing text passages of interest (whether online or in print) with "dead-end" references to a text string continually fragments the corpus (and the modern scholarly discussion of that corpus), silo-ing information, and making it harder to see connections.

In order to connect the medieval scholastic commentary tradition (and its modern discussion), we need to think more precisely about the targets of our citations and references. Are we citing the singular string of words that appear in our editions or in our articles? or are we referencing a common textual idea shared by the scholarly community? The former isolates and silos one discussion from the next. The latter creates the possibility of a common networked discussion.

The ability to cite a common textual idea is provocative; however, in practice, it requires the ability to actually refer to an idea rather than a text string. In this presentation, I will examine how the Scholastic Commentaries and Texts Archive ( leverages ongoing editorial work and the RDF (Resource Description Framework) data model to "reify" the text, creating identifiers for textual ideas at granular levels. With these target-able ideas in place, I will show how our practice of citation and commentary can change for the better. Through the targeting of textual ideas, we will be able to draw inferences from the aggregation of distributed citations, and we will be able to create more transparent encounters with these text ideas in distributed environments.

Athanasios Velios, University of the Arts London

Significant material evidence about the provenance of a book can be observed on its binding components. These observations include the types of materials and techniques used during both the production of the book and often during modifications of its structure (e.g. cover replaced or extra set of endleaves added). They also include evidence of condition which may throw light on aspects of the book's history such as storage environment and frequency of use.

These observations of material evidence lead to conclusions about provenance based on previous examples of similar evidence having been linked to established provenance details (e.g. book origin from a bookbinding workshop based on known decoration patterns). Because our reasoning on provenance is entirely based on past experience, the accuracy of our assessments depends on the available sample of past observations. By linking observations about material evidence across libraries and collections, we are able to build a more representative sample which will improve the quality of our conclusions.

This paper describes a method for encoding bookbinding descriptions based on the Conceptual Reference Model (CRM) of the Documentation Committee (CIDOC) of ICOM and the concepts formalised in the Language of Bindings thesaurus (LoB) which is encoded using the Simple Knowledge Organisation System (SKOS). This paper also describes the reasoning behind the organisation of LoB based on the CRM. It includes examples of encoded bookbinding descriptions from manuscript collections and it proposes ways of linking descriptions of physical evidence with provenance events and bibliographical information.

Top image: A chromatic scale diagram from a late 15th century copy of Boethius' De institutione musica (LJS 47, fol. 41v), merged with a detail of the linked data model of the Mapping Manuscript Migrations Project.