In today’s visualization, we have simple a bar graph that helps us to understand the complicated phenomenon of Piers copying over time. The graph below is deceptively simple.
Each of these columns represents* a quarter century of Piers manuscript production according to the extant tradition. It’s true that we have no idea whether or not the remaining manuscripts are representative of Piers copying or even LME copying, but since they are the only evidence we have, we must at least nominally operate on that assumption.
Note that even though the final quarter of the fourteenth century is represented here, the label reads “1390-1400” because none of the existing manuscripts could have been made before the last decade of the fourteenth century. Graphically, it is important to note that not all the bars represent equal time frames. But this is also interesting to Langlandians and because it means that not a single extant Piers manuscript dates to within the author’s lifetime since estimates put Langland’s death between 1385 and 1387.
[UPDATE: a quick conversation with Lawrence Warner on Twitter re Langland’s death and the dating of the C text shows both to be still under discussion.
He also points out a few MSS after the 1550 cut-off date for this graph. I’ll have to check out those ones and add them to the graphic! There are even two from the nineteenth century! How much fun is that?!?]
Another quick note about labels: some manuscripts are dated to “circa 1400” or “circa 1425.” Unless otherwise noted, “circas” will generally be put in the category of the number date rather than before it. Thus MSS dated to ca. 1400 all appear in the 1400-1425 block.
Speaking of which, the second important feature of this graph is its use of color striations. You’ll notice that the Y-axis of the graph is counting thousands of lines of Piers poetry. The striations, then, show you not only distinctions between manuscripts, but roughly how many lines of Piers each one contains.
Now the obviously distinct feature of this graph is the overwhelming volume of manuscripts produced in the first quarter of the fifteenth century. We have over two hundred thousand lines of Piers poetry from this 25-year period, more lines in more manuscripts than in all the other periods up to 1550 combined.
Because the 1400-1425 column so completely dwarfs the other periods of production, it makes it very hard to individuate the different striations and/or compare their size.
However, we do have at our disposal some slightly more precise dates for some manuscripts, specifically for these early fifteenth century MSS. Using the categories that A.V.C. Schmidt uses in his Parallel Text Edition, we can break that block up into three distinct categories: ca. 1400, “early fifteenth century,” and the umbrella category 1400-1425.
Now, because of the constraints of the graphing program, “early fifteenth century” has been translated to roughly 1400-1410. This is simply because the graphing program needs numeric identifiers rather than text. Whether “early” MSS actually date to precisely the first decade or not is not really knowable. I am, however, assuming that “early fifteenth century” has a little more specificity than simply the first quarter, otherwise why distinguish the two.
Additonally, we have to be wary of assuming that the MSS represented in the third of our S.XV1 columns means that all those MSS were actually made later than the others. Some most certainly were, and we know their dates. Others are simply not more precisely datable than that 25-year window. So, if you’re not careful when looking at this graph you might assume that the broken up S.XV1 column was made up of successive categories when in fact we know it was made up of overlapping categories with no clear delineation between “around the turn of the century,” “early” in the century and “first quarter.”
Even with the limitations of this graph, though, we can still learn quite a lot from the now highly distinguishable striations.
First we can see that whether you’re talking about a short text or a long text, they both have varying sizes. While they tend to be a similar size in each category, they are certainly not uniform. Second, you can clearly see some trends in preference for the long version of the poem in particular. But also, we see the short version rise to a “competitive” and then even dominant rate of copying as the copying of the long text drops off after 1450. And then after the turn of the sixteenth century, a somewhat different bookmaking milieu, the long text makes a resurgence and the short text dies out completely.
So, we can clearly see that Piers copying was subject to change over time. We might conclude that it was wildly popular in the first half of the fifteenth century and slowly dropped out of production as other forms and genres became popular, and the reemergence of the short text after popularity in general waned may indeed indicate a copying trend that wants that “good ol’ Piers” but maybe doesn’t want to spend the 7000 lines of copying or reading on it.
We might want to say that. But we have to think carefully about whether we have a chicken-or-the-egg problem. Does this graph look this way because of the changing popularity of Piers, or does it look this way because of the changing landscape of late medieval book production in England?
The fifteenth century saw some major upheavals in both literary and book-making culture. Not only is there a professionalization of bookmaking for an emerging mercantile class, but even monastic book culture is moving away from a contemplative form of copying toward a laureate culture (Chris Cannon, “Monastic Production”). Additionally, the proliferation of paper in England in the middle of the fifteenth century may have influenced the way in which people made choices about what to copy, since copying was no longer quite as dear. Finally, of course, the use of the printing press at the end of the fifteenth century would have eventually impacted book culture in so many ways that accounting for manuscript copying after print can be quite complicated. Nevertheless, there was no print copy of Piers until Crowley’s 1550 edition, after which no new copies of Piers appear to have been made. Again we want to be cautious. It’s not likely that it was simply Piers’ availability in print that led to it not being copied. Indeed, while early sixteenth century copies may well have been novelties–like Royal 18.B.xvii and Sir Adrian Fortescue’s copy (Digby 145)–due to the changing meaning of a manuscript in an emerging print culture, Crowley’s prints demonstrate the changing meaning of Piers in an emerging Protestant culture.
Which is all to say that Piers Plowman is a complex phenomenon unfolding in manuscript over a century and a half, and in print to this day. And even when we flatten this out to show its lifespan as a manuscript phenomenon, it’s temporality, even within the Middle Ages, is not uniform.