Skip to content

More Mapping Experiments

November 11, 2014

In this post, I want to present some experiments with mapping manuscripts containing texts related to Judith that circulated among Anglo-Saxons. Several months ago, I had worked up a table of the raw data for mapping, but then set it aside. Recently, I took up the data again, revised it, and decided to start experimenting with different mapping tools to see which ones worked, what issues I encounter, and to think through what and how exactly I want to map the data. In general, I want to map manuscripts at their supposed origin points (more on that below), as well as their movements in time and space, to be able to visualize clusters of key points overall and at particular points in time. For example, if I were to walk into the library at Worcester in the year 1000, what books containing texts related to Judith might I find on the shelves, or in the scriptorium? It’s this sort of question, as well as larger questions about visualizing manuscript data, that I have in mind.

So here are a few of my visualizations. All of the following maps are based on a much more robust set of raw data, which can be accessed via a Google spreadsheet here. For the dates, places of origin, and places of provenance, I have relied on both N. R. Ker’s Catalogue of Manuscripts Containing Anglo-Saxon (Oxford, 1957) and the more recent bibliographical reference by Helmut Gneuss and Michael Lapidge, Anglo-Saxon Manuscripts: A Bibliographical Handlist of Manuscripts and Manuscript Fragments Written or Owned in England up to 1100 (Toronto, 2014)–references to which I include in the data table. Also notable is the trove of descriptions and suggestions in The Production and Use of English Manuscripts 1060-1220, although Gneuss and Lapidge do account for this project.

After playing with different tools, I decided to focus on CartoDB, because it is very user friendly, it had several options for visualizing that I felt worked well (enough) for what I want to do, it liked my data, and it led to interesting results. It’s not the perfect tool for what I eventually want to map, but it is a good place for my data to reside for now.

So here is one map using it.

Click to open dynamic map.

This map depicts two sets of data points: 1) the origins of manuscripts containing texts within the project corpus (which should be interpreted as somewhat fuzzy, based on suggested provenances); and 2) the modern-day libraries in which those manuscripts are held. The two sets of data are by default depicted together, but the “Visible Layers” option allows viewers to choose which data to present. Of course, more interactivity would be ideal, but that just is not possible with the (free) version of CartoDB I used for this. Unfortunately, there is no way to link the two data points–to visualize lines drawn from origin points to modern-day library locations for each manuscript–but perhaps that’s possible with another tool, or in the future.

Another issue I encountered was the problem with “fuzzy” locations. For example, several of these manuscripts are marked by Gneuss and Lapidge with “?”–naturally indicating that this is a suggestion, but not definite. In the case of the provenance of other manuscripts, new scholarship could certainly shift assessments. So, for some manuscripts, “England” or “France” is indicated, or other regions (“SE England,” “S England,” “NE France,” etc.), but the only way to map those is to provide a generic data point somewhere in the center of England or France–which doesn’t depict accuracy as much as it does an aggregate. Even locally, there is the issue between Christ Church or St. Augustine’s in Canterbury–if we know a manuscript was at Canterbury, how do we determine which of the two specific locations? How do we distinguish? How do we show data that we’re fairly sure about rather than data that is only suggestive? The map, in other words, should be viewed with caution, and the raw data should be consulted for more details.

Here is another way to present this data, focused only on the probable or suggested origins of these manuscripts, with clusters indicating the more saturated locations:


Click to open dynamic map.

The clusters do a nice job of quantifying the hot points, although, again, fuzzy points should be kept in mind. For example, some of the clusters in the center of England aren’t for specific libraries (you might be wondering what major manuscript centers you’ve missed out on there!), but they’re generic data points for general origins like “England” or “Mercia.”

Finally, none of these maps allow for visualizing time along with the data. There are tools that will allow that, but I haven’t yet found the right tool. For now, these help to conceptualize where I want to go. Hopefully working with the data in an actual mapping tool has helped me sort out some of my future goals and the difficulties of visualizing complex, multi-dimensional data.


From → Uncategorized

One Comment

Trackbacks & Pingbacks

  1. Source Study in a Digital Age | Brandon W. Hawk

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: