Working with Networks

Scott Weingart’s “Demystifying Networks” serves as a great, easy-to-read introduction to Network Theory and Network graphing. Network theory, as Weingart notes, is a growing interest for many scholars– especially art historians! He defines Network Theory/Studies as such:

Generally, network studies are made under the assumption that neither the stuff nor the relationships are the whole story on their own. If you’re studying something with networks, odds are you’re doing so because you think the objects of your study are interdependent rather than independent. Representing information as a network implicitly suggests not only that connections matter, but that they are required to understand whatever’s going on.

Scott Weingart

I find his last point to be key– that relations between people, objects, places, concepts, etc are necessary to understand a complete context (story). For art history, I think this makes sense– in fact you could argue that network theory is woven into the foundation of the discipline (art, after all, doesn’t exist in a vacuum, but rather in many networks– private and public, local and global, etc. My point is not to argue that Art History is a form of network studies, but rather use this idea to say that Network Theory is definitely a viable and at times lucrative tool to art historical scholarship.

So how does this translate into Digital Art History? The rest of Weingart’s article describes and explains the ways one can graph networks– that is, turn the idea of a network into a visual. He outlines the many parts of a network and how they can split into different forms (ig. nodes can have two parts, bimodal). He gives some (very basic) examples of what a graphed network might look like, which I think help to reinforce his description. But when looking at these graphs, I think back to his explanation of Network Theory– “Representing information as a network implicitly suggests not only that connections matter, but that they are required to understand whatever’s going on.” Nodes can be connected with lines, symbolizing an interaction. But do these lines necessarily communicate the necessity of that interaction? Take his example:

This network graphs the relationships between Authors and Books, with the lines indicating that the Book (teal) was written by the Author (red). Again, this is a basic graph to serve as an educational model, but the question still applies– is it necessary? Do we need Network Theory to understand the relationship between an author and his books? Not really. Furthermore, is this a network in which we must understand the connections (the lines) to understand the author and or the book? Of course, books wouldn’t exist without authors, but I think the point here is that the network is not producing anything new about the book, the author, or the relationship between the two. We don’t need a visualization like this to know that a book was written by an author.

In fact, Weingart warns us of blindly applying network theory and visualizations to our research, what he calls methodology appropriation. He cautions us to be mindful about when it is best to apply network methodology to our work and critical in evaluating its usefulness. A fancy graph might look cool in your publication, but could be superfluous.

This leaves me with some questions: How can art historians train their critical eye to know when and when not to graph a network? With a strong push for data visualization, it might seem oh-so-appealing to supplement your text with a network graph, but how can we learn when this is actually the wrong move? Furthermore, how can we begin to reimagine network visualizations to communicate the dependency of networks– not just that relationships exist, but that they are necessary to understand the two (or more) parts/nodes. Lines just don’t seem to work, in my opinion. But what will? All in all, I am for network theory and graphing as a method to art historical scholarship, but believe that there is still room for improvement.

Scott Weingart, “Demystifying Networks”,


  1. Michelle says:

    Taylor, I think we had very similar reactions to networks this week. I also was interested in Weingart’s “methodology appropriation” and thinking about when you really should or need to use networks. My biggest issue with his examples, which I know are simplistic, was that the random placement of them in space seemed to imply a hierarchy or at least some meaning. This made me really interested in how to communicate the relationships/dependency in a network because I agree that lines just don’t seem to work. Although perhaps lines would work in a more regimented way? or arrows? I would be curious to see if you think an arrow would help clarify anything? That would only work in certain situations thought I’m sure.

  2. I enjoyed your post, Taylor. I agree that relations between people, objects, places, and concepts are necessary to understand a complete context. This is certainly true of art history, and many other disciplines as well. I also agree with your assessment that we don’t need Network Theory to understand the relationship between an author and his books. This visualization doesn’t tell us anything that we don’t already know. Weingart’s point about ‘methodology appropriation’ also relates to the issue of beginners using networks that may not be tailored to their project, and since they don’t know how to make changes to the program, it could lead to erroneous conclusions. I think that beginners (like myself) feel that we should do digital humanities projects to be up to date on technology, but as I learn about the field, I realize that it really isn’t something for beginners. There is a lot to consider, and learn before attempting a digital art history project. To be successful, I would need to collaborate with someone with training in digital technology. I feel that on the one hand, we’re being encouraged to embrace digital humanities projects, and on the other, warned about everything that can go wrong with said project if we don’t have the technological savvy necessary. Unfortunately, many of the visualizations we have seen aren’t particularly revelatory, and we are left wondering, so what? What have I learned from this visualization that I didn’t know before? I do think digital technology has tremendous possibilities for the field of art history. 3-D modeling, which we are working on in our class now, is a wonderful way to allow remote viewing of objects from anywhere in the world. What a fantastic resource for scholars! We could see objects in museum collections up close, and from all sides and angles, without any fear of damaging the object. StorymapJS and TimelineJS are also tools which have wide applications to the field of art history. I agree that there is still some room for improvement, and that needs to happen before art historians will embrace the use of digital technology in their research. There is a (sometimes steep) learning curve to using digital technology, and that can make it difficult for those who are interested in it to get started. Couple that with the fact that print is still considered the gold-standard of art historical scholarly research, and many art historians feel there isn’t enough compelling benefits to put the time required into learning digital tools. Will art historians ever completely embrace digital art history? Perhaps. Digital projects would need to be perceived as equal in scholarship to the more traditional methods currently in use, and training in the technology would need to be commonplace.

Start the Discussion!

Your email address will not be published. Required fields are marked *