DH2015: Modelling the (Inter)national Printmaking Networks of Early Modern Europe

Below are the slides and speaking notes for my DH2015 talk “Modelling the (Inter)national Printmaking Networks of Early Modern Europe”.

This paper was a finalist for the ADHO Paul Fortier Prize for best paper persentation by an early career researcher.

View the official abstract in the published proceedings.

Access the data and visualization code bundled in an R package.

One key point in the history early modern Dutch artistic printmaking is the formation of what might be called a national school: a shift away from internationalized, courtly mannerism such as the Goltzius print above after a design by a German painter, and towards domestic subjects and styles, like the Frisius print after a Matthijs Bril landscape.

Scholars have connected this turn in style and content to the coalescing of a Dutch national identity during the course of the revolt against Spain between 1568-1648. However, little work has been done trying to understand how changing production practices might have effected this shift, both in Holland and in other countries in the early modern period. This is where network analysis comes in.

Engravings are a fundamentally collaborative process (designer, engraver, publisher)

Furthermore, the result of this process – the print itself – serves as a great index of the existence of a professional relationship at a certain period in time…

…even to the point of being recorded on the print itself!

From these surviving prints we can infer a dynamic network of social connections between artistic sources, engravers, and publishers, well suited to computational network analysis.

Better yet, promising datasets have already been constructed for us – by museums! A few years ago the British Museum took the gargantuan step of publishing their collections data as LOD, and this seemed like a perfect dataset to adopt for the print production question. This is obviously a sample biased by the facts of what has survived to us over the last four centuries, and by the historical collecting practices of the BM – and I’ll get to ways towards addressing that by the end of the talk. As collections go, however, the BM does have two useful assets – a particularly broad collection of European prints from this period, and a meticulous set of curatorial data.

It wouldn’t be a DH session without a messy-yet-beautiful network graph of all these nodes and edges. Obviously this won’t work for us, in large part because flattening this dynamic network removes the 250-odd years of history informing its evolution.

Instead, we need to do dynamic network analysis on a series of temporal subgraphs.

The metric we’ll consider today is the EI index - a simple measure of the propensity for nodes belonging to an arbitrary group to connect to members of that group, or members outside that group. This provides a simple measure of how closely print production connections aligned with actor nationality.

Here, I’ve plotted out the changing EI indicies for six major regional printmaking communities – Dutch, French, English, Flemish, Italian, and German.

As a reminder, above the 0 line means members of the community were making the majority of their connections to foreign actors

While below the 0 line means members of a community made the majority of their connections to each other.

There’s some noise in the trends in the early years of the 16th century for several of these charts – that’s why I’ve included size of points as a variable to show how many prints each year’s subgraph is taking account of.

There are a few interesting shapes here. Notice that the Dutch, Flemish, French, and English communities all begin the sixteenth century making most of their connections outward (save for some noise in the Dutch trend caused by the painter and engraver Lucas van Leyden, each one undergoing at different points a relatively swift turn towards more inward connections.

Each of these externally-connected networks underwent their own relatively precipitous shift towards majority-internal connection, but at different points in time. We could make make individual suggestions for what might have precipitated these changes – for example, the beginning of the Dutch revolt in 1566, the French wars of religion in the 1620s, or the English civil war between 1642 and 1651 making it more difficult to conduct international trade.

And yet it is quite easy to locate exceptions to this apparent link between conflict and internal connection. If the onset of conflict spurred Dutch artists to connect more inwardly, the end of the revolt in 1648 and the reopening of borders did not result in a sudden return to more international collaboration. In that same vein, the sudden inward turn of Flemish print production occurred not during conflict, but during great prosperity, as the transfer of foreign trading houses from Bruges to Antwerp around 1500-1510 led to a veritable economic golden age in Antwerp.

This is not to say that these historical events had no significance – they undoubtedly did. But may be more useful to think of these events not as sole motivating agents, but rather as catalyzing agents affected by larger structural forces.

The odd players out in this analysis are the Italianate and Germanic networks. Unlike the swift drops towards relatively stable, internally-connected production, the Italian network remains majority-internally connected for the duration of this study period. And the Germanic network experiences several ongoing fluctuations back and forth across that 0 line, though for at least the first half of the 16th century it remains majority-domestically connected.

It’s tempting to think that the Italians were largely cut off from international European print production just looking at this graph. However, remember we are only looking at the EI ratios of each of these groups, not the actual balance of who was receiving or sending connections across this entire group.

Above is the percentage of all external connections received by each region, below is the percentage of all external links being sent. It’s very easy to risk overinterpretation of such a complex graph, so I’ll limit my remarks to a few quick observations.

Most importantly, we find that, although Germanic and Italianate regions were largely internally-connected in the early 16th century, they were by far the greatest recipients of what international printmaking ties were being sent and received at that time. Italy would remain a strong recipient of international connections through to the mid-17th century as many foreign printmakers and publishers looked to Italian artists for design sources. However Italian printmakers showed little interest in sending out their own international links; that distinction clearly went to the Flemish, and soon later the Dutch and French, and even English printmakers and publishers in succession.

It seems little coincidence that the regional networks that are primarily inward-connecting in the mid-16th century (Italy and Germany) also had some of the longest printmaking traditions, dating back to the late 15th century. The medium of printing demanded a set of artistic and technical skills, not to mention a set of social connections and financial capital, that presented a barrier to new entrants into the printmaking world In the aggregate, these requirements presented a barrier not only to individuals but also to regions and countries. Germany and Italy, the respective origins of woodcut printing in the north and the south in the late 15th century, were able to make mostly internal connections through the 16th century. They would gradually receive an increasing number of foreign connections, as Dutch, French, and English artists sought to connect to expert printmakers. Over time, these externally dependent regions would begin to cultivate more native talent, knowledge, and physical resources, as experienced printmakers trained new students and transitioned from making prints to establishing their own publishing firms. It seems that once a critical mass of designers, printmakers, and publishers had developed within a country, these national networks shifted quickly, rather than gradually, towards increased domestic production.

But I promised I’d speak to the problems caused by looking at just one museum dataset! What about the collecting biases of that particular institution?

This is where the open data from the Rijksmuseum API comes in. The Rijksmuseum’s Prentenkabinet another one of the great European print rooms, although unlike the BM, it was built with a more express goal of creating a definitive repository of Dutch and Flemish prints and drawings. The RKM has documented their prints with a similar level of detail, specifying the particular roles of actors in those artworks, meaning that we can run the same analyses and compare the results. Both institutions are biased in their own ways, but if we see shared patterns then at least we can speak with a bit more confidence.

It’s particularly interesting to see these results in light of the fact that both museums have slightly different ontologies for artist nationality (in the BM, for example, an artist may have multiple nationalities). However, the results are broadly consistent for Dutch, Flemish, French and Italian, mostly consistent for German, but not at all similar for English prints (in which the BM has voluminous holdings, while the RKM has very few). This is reassuring that the trends are not wholly due to the collecting practices of a single institution. More work needs to be done comparing these modern collex against what documentation – like stock lists – survives from the 17th c.

Future Directions

• There’s a lot more descriptive work to be done, and I’ll be exploring that further in my dissertation.
• Ultimately it’d be interesting to evaluate my interpretations of these trends through simulation, to check if they are at least plausible explanations for the observed network behaviors.
• However, we’ve more work to do in modelling print production…

Prints are a tricky fit for art museum CMSes. They are unique objects, but they’re also multiples that can be linked to physical printing plates that have their own states and histories, as well as preparatory materials like drawings. None of these complex relationships fit comfortably into the CMS for a single institution. That’s why we have fantastic reference sources on European prints – they’re just locked into a 50/60 printed volumes of Bartsch and Hollstein that sit on art library shelves.

I close, thus, with a plea for some way to link this data in an open way… say, a list of addressable URIs that museums could reference w/o reinventing the wheel every time. So, get in touch with your local LODLAM afficionados, and let’s make this happen!