| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

View
 

Elizabeth Floyd - Social Network Analysis

Page history last edited by Elizabeth 9 years, 4 months ago

A Century of Social Networking: 20th Century Connections

 

While each of the practicums have been useful in various ways, I was by far the most excited about trying social network analysis. While I've previously lamented how 20th century literature has presented its own challenges for digital humanities, multiple people have already attempted to determine the various social networks both within texts and outside of them. For example, Ben Schmidt, Chad Rutkowski, Anastasia Salter, Heather Prescott, and Amanda Visconti have created a social network analysis of the Wandering Rocks chapter of Ulysses. MOMA also created a great interactive social network analysis of Modernist artists, writers, and musicians that were involved in determining Abstraction from 1911-1925. Their website allows you to highlight a name, see its connections, then click on the name and find a more detailed bio and examples of their work in the MOMA collection. There are various other projects that go beyond individuals' connections, trying to map texts upon space, show influence of certain movements geographically, etc. 

 

(Although I don't have any evidence for this theory, it seems as if Modernists like these tools over the other DH forms of analysis because 20th century art and literature is about connectivity with technology and the banding of individuals against the pressures of modernity. However, I'll leave this half-baked idea for now...)

 

While these projects all underscore really fascinating elements of Modernism, they specifically look at High Modernism, or what I like to think of as the T.S. Eliot/Ezra Pound club. James Joyce is a particular favorite to try to parse through various DH tools because of his complexity and "wandering" texts. For this project, I borrowed a bit from the ideas of David Elson, Nicholas Dames, and Katherine McKeon to begin my approach to social networks. Instead of using a 19th century text and thinking about urbanity versus the rural, I focused on one chapter in Evelyn Waugh's novel, Brideshead Revisited, Chapter 5. In this chapter, the narrator, Charles, begins his second year at Oxford, engages in various "adventures" with Sebastian Flyte, and then watches the unraveling of Sebastian. This chapter marks the psychological unraveling of Sebastian and the underscoring of the decay of British aristocracy. I tracked the interactions of the characters, whether it be conversationally, a reference to a meeting or exchange outside of the narrative, correspondence, a reference to an "attachment," a random memory or thought of a person, and emotional influence. These were then weighted from 1-6, respectively.  

 

This was my first run of the data, inputting only the first half of the chapter:

 


 

 

Given this was my first run, I was pretty happy with how things appeared. But of course, more tinkering was necessary. I moved away from the Fruchterman Reingold layout and ran the Force Atlas 2 layout to get this:

 

What if I ran the Expansion layout:

 

Nope. So somehow things went from ok to worse. As Tyler noted in his post, the logic behind some of these layouts is unclear in my initial reading and messing around with Gephi. I decided I needed to add the rest of the chapter for a more complete model and actually learn some of the stylistic buttons for an easier synthesis. After more data and reloading the information, I finally ended with this:

 

 

Ok, now this is what I was hoping for in terms of a readable model. By the way, I have I mentioned how much I love infographics? This finally began to meet my standards in terms of ability to understand, while starting to look somewhat approachable design-wise (I would still change more given the time). What I found is the same complaint others have mentioned time after time - the data is weighted by my own prejudices. I started out labeling things as I encountered them in the text, rather than thinking about how "1" would have the least weight. However, in determining my order, what I also began to realize is that in a novel about society, there are a million interactions, most of which are rather unimportant. For example, the Prince of Wales golfs with Rex Mottram, and yet, is of no significance in the novel. By putting more weight on the emotional/sentimental interactions, the focal characters emerge: Charles, Sebastian, and Lady Marchmain. Then the characters who present the next level of influence are Boy Mulcaster and Mr. Samgrass. In reading the chapter, neither holds a significant role in terms of their narrative position, and yet they both drive the plot through their characteristics. Boy Mulcaster is the typical British party boy and Mr. Samgrass, the irritating historian/family intruder. What also is interesting is that characters who are removed from the narrative and only mentioned in passing, such as Anthony Blanche, have a significant presence. This suggests that while physical interactions are important in a society novel, so are the people on the outside, determining how one might behave or act. On the next run, I would think more about how I wanted to weigh the interactions, but I'm curious to see how much it would actually change.

 

 

Comments (0)

You don't have permission to comment on this page.