Mapping pre-1600 European manuscripts in the U.S. and Canada

Pre-1600 European manuscripts in the United States and Canada (detail)

Today marks the beginning of the 7th annual Schoenberg Symposium on Manuscript Studies in the Digital Age here in Philadelphia. This year the symposium theme is "Collecting Histories" and features a line up of speakers discussing the ways in which provenance and the history of collecting informs our wider knowledge about manuscript culture. As readers of this blog know, I'm very much interested in the historical movement of books and manuscripts and I'm excited to speak during the conference on the ways in which the Schoenberg Database of Manuscripts (SDBM) can be used to track manuscripts over time.

For this post though I want to highlight the fantastic work done by a team of scholars whose work very much informs the SDBM project. Over the past two decades, Lisa Fagin Davis and Melissa Conway have worked to create a new directory for all institutions in the U.S. and Canada which hold European manuscripts dating to before 1600. They have published their own excellent description of the origins and methodology of the project but in short their work began as a way to update the censuses of American manuscripts created by Seymour de Ricci from 1935-40 and supplemented by Faye and Bond in 1962. Their census includes entries for 937 entities:  historical owners of manuscripts derived from previous censuses, the former names of institutions now renamed, as well as current holders. Running to 126 pages in a freely available PDF sponsored by the Bibliographical Society of America, the census is an incredibly helpful resource and I wanted to find a way to make the data contained within it browseable in a different way than just on the printed page.

Example of a listing from the Fagin Davis & Conway Census (p.37)

I extracted the text from the PDF census and chopped it up into relevant delimited fields like "Name" "Address" "Holdings" etc.  and then mapped the results using CartoDB. I had to make a few decisions about display along the way, especially when it came to how to determine the size of each manuscript owning dot on the map. Most institutions provided Fagin Davis and Conway with numbers for how many manuscript codices they held as well as how many leaves, documents, and scrolls were in their collection (though others reported only an aggregate number). Most institutions with full-fledged manuscript books had a fairly well-informed count of exactly how many they had but the numbers for leaves and documents often were estimated in larger round figures. As a result, the default map view gives all locations in the census with dots on their locations by number of total manuscripts held (leaves, codices, scrolls, documents, etc.). Using the "visible layers" dropdown you can turn off and on just those locations currently holding manuscripts or just those recorded in earlier censuses which no longer hold manuscripts or both together. Of course sizing the dots by total manuscript holdings will be necessarily a bit misleading as a university with 2 codices and 37 leaves appears to have total holdings of 39 manuscripts, so there is also an option in the "visible layers" menu to view only holdings of codices.

[View map in separate window] 

Unsurprisingly one can see the concentration of pre-1600 European manuscript holdings along the east coast. In a league table of manuscript holders New York, Washington, and Philadelphia(!) come out on top by volume but in terms of individual institutions the Huntington and Folger with their extensive holdings of pre-1600 documents come out on top.

Top-15 current owners of pre-1600 manuscripts by "total" count in the Fagin-Davis/Conway census

Given the fuzziness of this catch-all "total" manuscript number it's helpful to also get a sense of institutions by number of codices held:

Top 15 current owners of pre-1600 manuscript codices in the Fagin Davis/Conway census

One of the advantages of using the Fagin Davis and Conway survey is that it lists private collections, and in the cases where these were dispersed or relocated, notes their current location. I don't think it would be terribly controversial to say that most collections of medieval manuscripts in the U.S. and Canada rest on substantial gifts from individual collectors or families. The remarkable extent of these private collections can be seen in part below:

Collections of pre-1600 manuscripts now identified as being relocated in the Fagin Davis/Conway census

Top 15 now-relocated collections of pre-1600 manuscript codices in the census

It's edifying to see the late Larry Schoernberg at the top of the list of codices, especially today during the conference celebrating his legacy. His manuscripts are now here at Penn but a decade ago when they were in Longboat Key, Florida they made that small community the largest holder of pre-1600 manuscript codices in the south. Others on that list will be familiar to many, including George Plimpton whose manuscripts are now largely at Columbia University and Thomas Marston whose collection is at the Beinecke, and Ricketts, whose collection is now mostly at the Lilly library.



Fagin Davis and Conway have already written on the implications of their survey for the history of manuscript movement but I was curious to see how manuscript holdings related to holdings of early printed books.  Since I had data on incunabula holdings for the U.S. and Canada handy I thought I would overlay the two sets of information together.

[View map in separate window] 

 Holdings of pre-1600 manuscripts (orange) and pre-1500 printed books (green)

As one would expect, the majority of institutions with pre-1600 manuscripts also hold pre-1500 printed books with the major institutions of the east and west coasts remaining prominent. Though it is de rigueur now to speak of early printed books and manuscripts as complementary genres rather than necessarily competing ones, I know from experience in special collections libraries that print and manuscript remain separate in many places and that historically collectors of one genre might not necessarily be interested in the other (Larry Schoenberg for instance was a manuscript collector and for the most part did not own incunabula). I was surprised then to see how well the two sets of data overlap - that is, there were many fewer outliers than I expected. There were a few though such as the Mount Angel Abbey Library in Benedict Oregon which reported 25 codices and 63 pre-1600 leaves but is not listed in the GW as possessing any incunabula (though they may very well have them!). Or a smattering of public collections which have small collections of either incunabula or manuscripts but not both. The Lima Public Library in Lima, Ohio for instance reports holdings of 68 pre-1600 manuscript leaves but no early printed books. Though I'm still working on wrangling the data between the two sets, I'm eager to see what other small collections or big discrepancies pop up.

Finally, if you're interested in learning more about medieval manuscript collections in the U.S. ckeck out the program of this weekend's conference as well as Lisa Fagin Davis' blog detailing her virtual travels to manuscript repositories around the country over the past year.