Visualizing Collation in the Hunterian Manuscript Collection

By Sam Gilchrist on July 1, 2022 • ( 1 )

Guest blog by Dot Porter, University of Glasgow Library Research Fellow.

My name is Dot Porter and I’m the Curator, Digital Research Services at the University of Pennsylvania Libraries. In 2020 I was awarded a Visiting Research Fellowship at the University of Glasgow Special Collections, which would allow me to come to Glasgow and spend a month examining manuscripts from the Hunterian Collection. I was so excited! Unfortunately Covid-19 happened, and I had to wait until 2022 to make the trip. I was finally able to do it, and I spent all of April in the reading room, paging through an amazing collection of medieval and Renaissance (and some later) manuscripts. In this blog post I want to say a bit about what I did during my visit, and why it’s important.

First, the focus of my research is the physical construction of codex manuscripts. Codex manuscripts – or “book shaped books,” as I like to call them – are complex physical objects, consisting of leaves physically attached to other leaves through the center of a sewn quire. However, traditional photographic digitization flattens them out. Manuscripts are presented as a series of flat images, divorced from the book, usually in a facing-page view that mimics how books look in real life.

*Figure 1: UPenn Ms. Codex 1056, Book of hours, use of Rouen, 15th century, presented as a page opening in the Colenda image viewer (https://colenda.library.upenn.edu/catalog/81431-p3b853k0m*)

In reality, books are fully moveable objects. Most leaves are connected to each other through the structure; others are single; some have been added and others removed or moved around. Traditional methods for describing manuscripts and presenting them digitally either ignore these facts or present this information as formulas or as narrative description, both of which can be confusing especially when a manuscript’s organization is complex. I believe that visualizations are a much better way to present this data, and the month I spent as a University of Glasgow Library’s Visiting Research Fellow put this idea to the test.

In my day job I think a lot about what digitization does to how we think about medieval manuscripts. Although this involves a lot of theory, I’m primarily interested in taking theory and making it practical for our collections at scale. Since 2013 I’ve been co-directing the VisColl project with Dr Alberto Campagnolo. Through VisColl, or Visualizing Collation (https://viscoll.org) we have been developing a data model and software to describe and visualize the gathering structures of codex manuscripts. In September 2021 we launched VCEditor, a publicly available tool which enables catalogers and scholars to build models and visualize them (https://vceditor.library.upenn.edu). VCEditor is built on the VisCodex system developed at the University of Toronto’s Old Books, New Science Lab, but updated and with new functionality added. VCEditor provides an interface for building the physical collation of a codex manuscript using a diagram of quires as a visual format, and then users can add additional information – physical details, notes on textual and illustrative content, or any other aspect of the manuscript – overlaying this information on the diagram to show at a glance how content and physical organization overlap. In addition to gathering structured data (in XML format)—that one can subsequently export into any collation formula—VCEditor permits users to generate automatic diagrammatic representations of the gatherings and to visualize digital facsimiles that re-join into virtual bifolia the photographs of the leaves of books, thus making overt the physical makeup of books.

Figure 2: Photo of the outer edge of MS Hunter 15, which includes parchment and paper interleaved to build the quires. One could use VCEditor to show the pattern of manuscripts that are made with both parchment and paper, such as this one.

My aim for my fellowship at the University of Glasgow was to build collation models of manuscripts in the Hunterian Manuscript Collection, a collection of over 600 manuscripts collected by William Hunter FRS, a Scottish anatomist and physician, in the 18th century. Six-hundred manuscripts is too many to examine in one month, so I set the scope for the project:

Only western manuscripts
Only manuscripts written before 1700

This took me down to 293 manuscripts, less than half of the total but still too many to physically examine in the span of a month. So I spent many hours before my visit (mostly over my winter break) going through the printed catalog, A catalogue of the manuscripts in the Library of the Hunterian Museum in the University of Glasgow, published in 1908 (aka Young & Aitken) and building collation models using the formulas and descriptions printed in it as a guide. Once I built basic models, I went back and added textual contents and illustrations to the model. As I went I made notes of issues that I noticed – things that would necessitate an in-person examination during my visit.

*Figure 3: Collation Model for MS Hunter 231, Quires 5 and 6, showing structure, contents, and illuminations*

Collation formulas and descriptions can be valuable and provide enough information to create a model, but sometimes it’s clear that something isn’t quite right. A finished model may have more or fewer leaves than the catalog record says it should have, or the folio or page numbering provided in the contents list in the catalog might not agree with the model (for example, a text that “begins on quire 8, leaf 1 (f. 81)” only in the model it’s folio 79). Here is a screenshot of a section of notes in my spreadsheet, showing the kinds of notes I left for myself when deciding which manuscripts I would definitely need to examine during my visit.

*Figure 4: My notes about issues I was seeing after creating models using information from Young & Aitken*

I ended up with 119 manuscripts on my list to examine in person, and during the month I looked at every single one of them, and created finished collation models for all of them, too. If you want to take an initial look, there’s a Google Spreadsheet organized by shelfmark here (and for reference, a PDF of Young & Aitken here).

In addition to the collation models, a second thing that came out of the fellowship was that I was able to do a thorough testing of VCEditor on a large project. I got a sense of the types of physical information that is easy to describe in VCEditor, and how the model can be stretched to describe complex and unexpected information.

I’m spending this summer on a pilot project at Penn, to create collation models for our manuscripts, and the Glasgow fellowship was an excellent warm-up for this project.

But of course the most pleasurable part of the fellowship was getting to spend hours in the reading room with the University of Glasgow’s amazing medieval manuscripts!

There are so many amazing ones, but here are four in particular that struck me.

MS Hunter 75, which I quickly dubbed “The Bell Manuscript” as I turned each page and came across more and more Bells! The text of the manuscript is Pietro de’ Crescenzi’s Ruralia commoda, an agricultural treatise based largely on classical and medieval sources and the author’s experience as a landowner. But why all the bells? The word “Belton” appears in scrolls at the openings of Books IV, VI, VII, and X. Young and Aitkin describes the situation and notes the punning around a name, but doesn’t say anything about who this “Beltun” or “Belton” might be. Perhaps the person who commissioned the manuscript? I don’t know but in any case, the bells sure are striking.

Full view of illuminated page of MS Hunter 75

MS Hunter 231 is a new favourite for me. I have an interest in fandom studies applied to medieval manuscripts (particularly my work around manuscripts in Star Wars and Books of hours as transformative works) and MS Hunter 231 is a neat addition to this circle of interests. This manuscript is a 14th century compilation of devotional and philosophical works, including Saint Anselm, Saint Augustine, and – notable – many works by Seneca. The person who commissioned the manuscript, Rogerus, who has been identified as Roger of Waltham, Canon of St Paul’s and Chamberlain of Edward II, had himself painted into many of the illustrations, and Julie Gardham pointed out to me that the large miniature of Aristotle, Plato, and Seneca, all look an awful lot like Rogerus too! Self-insert is a major part of modern fandom; Rogerus was truly a man ahead of his time.

I can’t write this post without mentioning MS Hunter 229, aka the Hunterian Psalter, a 12 century English manuscript. This extraordinary manuscript is undoubtedly the most treasured manuscript I’ve ever seen in person. The illuminations and text are perfect, as is the parchment. It’s a gorgeous manuscript.

Detail of ‘Capricorn’ illustration, MS Hunter 229.

One last one that caught my eye is MS Hunter 169, which I dubbed “The cute diary.” This is a bit later than the others, mid-17th century (1616-1640), and it’s Medical Receipts and Memoranda written by Henry Fowler, M.A., Rector of Minchampton’s. It contains notes, thoughts, recipes, and drawings from one man over a period of years, and I think that’s delightful.

View of MS Hunter 169, showing writing with small drawings.

I thoroughly enjoyed my Visiting Research Fellowship at the University of Glasgow Libraries. Having the time to focus was helpful, and I think the experience has moved my research with VisColl and practical work with VCEditor forward. I also made some new friends and colleagues, and I’m already looking forward to going back again!