Technoscience / Ecomateriality / Literature
Header

DH Critique: Diego Nogales and Greg Lyons

September 22nd, 2014 | Posted by Greg Lyons in Uncategorized

Stephan Thiel’s “Understanding Shakespeare” project succeeds as a digital design project, but it falls slightly short when viewed as a digital humanities project (which, in our opinion, requires effective analysis and original conclusions). Thiel aims to present a “new form of reading drama” (Thiel) to add new insights to Shakespeare’s works through information visualization. The project is broken into five separate approaches, each of which turns the words and events of Shakespearean drama into data and then presents said data in an informative visual display. While Thiel’s intentions (the “new form” stated above) constitute a worthy design goal, they do not serve as a strong thesis to guide the literary implications of his project (or lack thereof – literary conclusions are mostly absent). The separate approaches are not linked to support a core argument.

Each approach display has a small, concise description of its purpose, and presents data in a visual form that is easy for any average reader to navigate and explore. In viewing Shakespeare’s words as information to be processed (by methods described further on), Thiel goes against the opinions of Stephen Marche and others who argue that “literature is not data” (Marche). Marche fears the advent of the digital humanities and criticizes the field for being “nothing more than being vaguely in touch with technological reality” (Marche). He goes on to describe the sorts of algorithms that Thiel uses as “inherently fascistic” (Marche). Most digital humanities scholars will dismiss Marche’s fears of algorithms as irrational and exaggerated. However, there is a danger to the scholarly pursuit of literary analysis when projects claim to serve a literary purpose but instead do relatively little literary research. Although Thiel’s project is primarily a design project, his own self-written goals are a little too ambitious and reflect literary intentions that he does not satisfy. For example, his “Shakespeare Summarized” approach uses a word frequency algorithm to condense speeches from a play into one “most representative sentence” each, which he claims will create a “surprisingly insightful way to ‘read’ a play in less than a minute” (Thiel). This is a far-fetched claim, as the “Shakespeare Summarized” charts each turn out to be more of a disjointed collection of hit-or-miss quotes rather than a coherent narrative. The charts give no detail with regards to plot events or characters, and viewing this data cannot be compared to the experience of reading Shakespeare’s full prose. The data presented is of little value to someone who has not previously read the associated work. Therefore, Thiel falls short in re-purposing the data to create an analytic digital humanities project – instead, he simply gathers the data and presents it visually.

Another of the approaches, “Me, You and Them” (Thiel), serves to identify each character’s role by compiling statements that begin with personal pronouns. Thiel claims that this approach “reveals the goals and thoughts of each character” (Thiel), though the project itself does no analysis of the data. Scholars who are familiar with the work may be able to examine Thiel’s compiled data and draw conclusions from it, but there are no conclusions put forth as part of the project.

Looking at the overall project’s design and technique criteria, it is clear that this digital humanities project really did form in sync with the concept and tool application. Thiel is well aware of the affordances of his tools (the capabilities of each algorithm for useful visualization), and he is effective in organizing the data in a readable manner. The approach titled “Visualizing the Dramatic Structure” introduces Shakespeare’s plays through a fragmented lens. Each lens signifies a major character within the entire play, or simply a character important within one scene. To produce this, while still maintaining an authentic feel to reading a play, this approach has a very inventive page structure. The structure follows that of a novel, however the story is divided by vertical lines that create horizontal portions for each scene/character that summarize their most important lines. This format reveals how this approach properly demonstrates the affordances of the overall project through this particular fragmented, yet organized, display. Thiel focuses on using technology that affords him the ability to examine the scope of an entire story by highlighting smaller, important details. The only major concern or flaw in the design of the media was that the visuals were presented through Flickr. This made it somewhat difficult to zoom in far enough and more so to navigate the vertical Flickr photo. A higher resolution and different media type for the visuals would have pushed the design to a higher level of sophistication.

dhpic

(Hamlet, Prince of Denmark – Understanding Shakespeare Project)

It is not sufficient to only view the final presentation of a digital humanities project. Examining the development of any project is imperative to fully appreciating the level of work and rigor involved within a project’s creation. Studying the design process also can reveal biases or assumptions inherent in the project. The “Understanding Shakespeare” project was successful in recording and documenting the entire process, from the digitalization of the plays, to the coding manipulation of the data, to its fruition. The process is presented through a series of Youtube videos fast-forwarding through the various mini-projects. This is a great tool to observe and, to an extent, understand the coding algorithms that were used to organize the words or lines of the play by frequency. The major dilemma with this entire process, however, is that without a Computer Science major, it may be impossible to understand the process of the coding by looking at the video. What is missing in this page is verbal dialogue, walking someone through the process as the video is playing. Therefore, even though the documentation is there, the transparency of the project’s development isn’t present.

This Shakespeare project not only documents the entire process to the final product, but it also thoroughly credits the different platforms and software used within the project. In the “About” tab, all the acknowledgements are made. It certifies that the data being used was based from the WordHoard Project and Northwestern University. In addition, it reveals that the software processors called “Toxicilbs” and “Classifer4J”, were the ones used to manipulate the data into an interesting visual arrangement based on frequency. In terms of project visibility, the open web accessibility of this project allows for any academic scholars to examine Thiel’s charts. Furthermore, it is also open and simple enough that it accommodates for the layman who may only be attracted to the visuals of one play that he or she may have read. It is worth noting, however, that Thiel does not make the raw data available to the public – he only displays the data visualizations.

To sum up “Understanding Shakespeare” as a digital humanities project, it helps to look through the lens of a prominent digital humanities scholar like Katherine Hayles. In her book “How We Think”, Hayles describes how “machine reading” processes like Thiel’s algorithms could supplement traditional reading experiences by providing a “first pass towards making visible patterns that human reading could then interpret” (Hayles 29).  However, this relationship implies that machine reading could inform readers who have not yet read the work traditionally, and in the case of “Understanding Shakespeare”, the data is not of much use without previous familiarity with the drama. As of yet, no scholars have taken advantage of Thiel’s project to make literary arguments, and thus it still sits idly as what Mattern would describe as a “cool data set” (Mattern). Standing alone as data, the project leaves lingering questions: Could these techniques be applied effectively to the works of other authors, and more importantly, what are the literary implications of this type of data?

 

Citations:

Hayles, Katherine. How We Think: Digital Media and Contemporary Technogenesis. Chicago: U of Chicago, 2012. Web.

Marche, Stephen. “Literature Is Not Data: Against Digital Humanities – The Los…” The Los Angeles Review of Books. N.p., 28 Oct. 2012. Web. 15 Sept. 2014. <http://lareviewofbooks.org/essay/literature-is-not-data-against-digital-humanities/>.

Mattern, Shannon. “Evaluating Multimodal Work, Revisited.” » Journal of Digital Humanities. Journal of Digital Humanities, Fall 2012. Web. 22 Sept. 2014.

Thiel, Stephan. “Understanding Shakespeare.” Understanding Shakespeare. 2010. Web. <http://www.understanding-shakespeare.com/>.

You can follow any responses to this entry through the RSS 2.0 You can leave a response, or trackback.

Leave a Reply

Your email address will not be published.