Last week’s post was about Piers Plowman copying as a temporal phenomenon. This week’s is about Piers as a spatial phenomenon.
I’m going to post two different images of the same map created in two different programs. Both are live links to the complete graph that can be explored online as well.
This map is built on GoogleMaps, but the classic maps variety, not the new maps. Currently it is probably easier to build in Google Earth. In either place, though, regions are easily drawn onto Google’s standard basemap and then saved as KML files.
What you’re seeing in both maps is a representation of the dialect regions identified in A.V.C. Schmidt’s Parallel Text Edition. It’s important to remember that there is a difference between being able to identify the scribal dialect and being able to identify where the MS came from. What this map is telling us is that the scribes of the different MSS come from these various regions. Given the circumstances of copying, there may be MSS in which the dialect is more the dialect of the exemplar than of the copy text (thus not even indicating where the immediate scribe hails from). Moreover, there may well be instances in which the dialect of the scribe obscures where the MS itself may have come from, just as there may be instances where scribe and MS come from the same region.
There is one more limitation to this map that I want to point out before I explain the components. Any MSS that don’t have an identifiable dialect or region are not included on the map. That is, if there isn’t a place on the map to put them, they don’t appear in the data. There are eight full MSS that do not appear in on this map. Moreover, regions are drawn with whatever specificity they are given in Schmidt’s edition. If he can pinpoint something to Southwest Worcestershire, that’s how big the region is. If, however, it is only pinpointed to vaguely “the East Midlands,” the region appears bigger. Smaller regions indicate more precision.
One of the advantages of using CartoDB over Google is that CartoDB knows you are graphing data, so it gives you tools for organizing and presenting that data. Here, we have this helpful color key that indicates how the colors for the different regions have been determined.
Throughout all my graphing endeavors, I try to keep a basic color code consistent so that multiple graphs are legible without having to absorb a whole new framework. I choose primary colors for the three archetypal texts (A is yellow, B is red, and C is blue) so that I can also represent hybrids accurately (thus AB is orange, BC is purple, and AC is green). Any MSS that are outside the basic hybrid structure are assigned colors that are neither primary nor secondary (Z is teal while Hm114, the conflated text, is a maroon color).
There are any number of things to note on these graphs, so I’ll just point out a few for today. First, we can clearly see that of all the MS varieties, A is by far the most widespread. And, that’s A-only, not A portions in hybrid MSS. It’s gone the farthest North and South, and as far East as the other two traveling texts.
Second, we might note that there is a high degree of B activity around London and East of it, activity that almost no other text displays (except for Hm114, which has a B base with C and some A conflation).
Aside from this activity around London and some vague activity of varying types (AC hybrids, an AB hybrid, an A and B text) in the East Midlands,
the vast majority of MSS come from the West Midlands area. In that area, we can see that MS region is stacked on top of MS region in both maps.
In both instances, I tried to draw regional boundaries enough different that if you wanted to click on one to pull up its data, that would be possible.
However, in places of the highest MS density, making sure that all regions are separable presents significant difficulty.
So, there are some drawbacks to this form of mapping, but what it does allow us to do is to put as much geographical information as we have about the MSS onto a map and look for patterns.
Here we see that perhaps the most important pattern is that Piers is by far most copied either in the West Midlands or by a West Midlands scribe. The concentration there is particularly C heavy, and as far as we know we can’t attach C copies to any other region on the map. Second, a significant number of B MSS also indicate that they originate from the same region, while a much smaller proportion of A MSS indicate the same (namely the Vernon and Harley 875). In contrast, fully half the (and indeed the only locatable) A/C hybrids come from outside the high-density region closest to Will’s Malvern Hills.
What kinds of conclusions can we reach knowing Piers‘ existence in space? Well, for starters, we have to re-examine the argument that A was copied after the wide availability of B and C only because the scribes had no access to the “complete” copies. While it may be true for the Piers texts furthest afield, we might find that those MSS, copied at such a late date might have made the A choice for other reasons.
Perhaps, for instance, copying A in TCD 213 in the second half of the fifteenth century was more a matter of how desired (or how not very desired) it was rather than a question of availability. TCD 213 is, after all, a very strange and sloppy MS that contains only the Visio portion of the poem alongside some monastic accounts and two different lives of Alexander.
Second, what it definitely tells is is that Piers was highly desirable in and around the place from which we suspect Langland to originate, and that eventually, all different forms of the text spread out from that locality, except for C, which only ever seemed to make it as far as Oxfordshire in the existing evidence.
Is it conclusive? No. Is it interesting? I hope so. I would strongly encourage you to peruse the full maps (available by clicking on the images above) and start drawing your own conclusions.