An Experiment in Progress
As the final exam began my students broke into groups, huddling around laptops, reams of numbers, and books of poetry. Low conversations rumbled through the classroom, punctuated by raised hands and urgent questions about what on earth we were doing. My job: demonstrating how the visualization software worked again, pushing them to think further about the emerging results, and reminding them that this was just another experiment.
“Experiment” had become a watchword in “Introduction to Digital Humanities,” a course that asked students to transform their interpretation of literature through the application of specific technologies. In the weeks prior to the exam, the students had read two volumes of poetry—Mean Time (1993) and The World’s Wife (1999)—by the British Poet Laureate, Carol Ann Duffy. Drawing on the collection of Duffy’s papers in MARBL, we had also read a letter in which she suggests that the latter volume “is not a ‘normal’ poetry collection by me.” Motivated by this assertion, the students read the volumes with an eye toward identifying any differences between them. When writing papers on this subject, the students drew on the themes, imagery, and language present in each volume to argue if Duffy’s own assessment was accurate.
The final exam asked the students to go beyond these typical avenues of inquiry and to engage with Duffy’s two books on a quantitative level: the average number of words per poem, the frequency of particular words, and what groupings words appeared near each other in the corpus, to name a few. The students collectively produced a digital transcription of the books and then set about their investigation using the text-analysis suite Voyant Tools and Excel. The task facing the students was not to reach a pre-conceived conclusion about what the data would reveal but rather to experiment and see how—or even if—such tools added to our already considerable discussion of Duffy’s poetry. By the end of the exam, each group had collaboratively written an explanation of the phenomena they observed and how such results related to the larger question of the uniqueness of The World’s Wife.
While this exam was an experiment, so too was the class itself: a joint endeavor of the new, Andrew W. Mellon Foundation-sponsored Digital Scholarship Commons (DiSC), where I’m a post-doctoral fellow, and the English department to offer the first course at Emory dedicated to the digital humanities. My students were wrestling with the quantitative analysis of poetry, and Emory’s libraries and college were experimenting with the future of humanities research—and the liberal arts.
Libraries have been central to humanities research for centuries. They provide the raw materials for scholars’ investigations and often serve as the space in which these materials are painstakingly converted into original arguments and insights. It’s not whimsy, then, to compare the relationship between a humanist and the library to that between a scientist and the laboratory. For the last several decades, the library has turned to digital technology to augment the raw resources it makes available to scholars: books, journals, and archives have been joined by electronic catalogs, databases, journals, and books.
Today, academic libraries—and the Emory Libraries in particular—are seeking to support new technological methods for engaging with these materials. If digital resources are the equivalent of adding to the lab’s supply of chemicals, then such new methods are the equivalent in investing in a different microscope, which allows work that simply could not be done previously. The computational text analysis that engaged my class is one such “microscope.” Other new tools include using geospatial tools for visualizing historical data, topic modeling for identifying common subjects in textual corpora, and network analysis for displaying the relationships between ideas or individuals. Along with these new approaches, the Library has also furnished “lab assistants” in the form of its subject librarians, software developers, and the team working in DiSC.
Collaborations between faculty and the library’s team of scholars have already begun making use of these new methods. In the “Lynchings in Georgia (1875-1930)” project, Roberto Franzosi of the Department of Sociology is partnering with DiSC to visualize data from the records of almost 400 lynchings. Franzosi built a database of these events from newspaper accounts, and the library’s team is creating network relational maps that show how different actors in the events relate to one another. Such research should help answer familiar questions about historical events but do so through an entirely new, digitally enhanced fashion.
Other experiments underway in the library are connecting technology with Emory’s physical collections. The “Views of Rome” project draws on Pirro Ligorio’s 1562 map to create an interactive display for research. Not only will people be able to explore high-resolution images of Ligorio’s work, but students in Sarah McPhee’s and Eric Varner’s art history classes will be able to embed their research about Roman monuments into the map. They will augment this research with other images and texts from MARBL to show how the same locations have been depicted differently throughout Rome’s history. In this manner, the tools and materials in the library enhance not only faculty research but also the work of Emory’s undergraduates.
As was the case with the Duffy project in my class, deploying such new technologies in the libraries shifts not only humanities research but also what a liberal arts education means, putting a renewed emphasis on primary evidence. Undergraduates working on such projects—even within the context of a class assignment—make new contributions to a scholarly discourse as they deploy new digital methods. Further, students involved in such research broaden their understanding of professional scholarship; as a result, their liberal arts education prepares them for lifelong engagement with the work of academics, regardless of their ultimate career paths.
Yet as Emory creates new avenues for humanities research, we must remember that future scholars will want to engage with today’s scholarship. This presents a challenge, as the technologies that drive many innovations of the present will quickly become obsolete. What, in other words, is the future of the liberal arts if the output of today’s scholars becomes inaccessible? This question is critical for today’s libraries, which preserve the work of previous generations. One reason, then, to have the libraries and DiSC help develop new methods of research is that it provides us a chance to advise scholars on matters that will increase the longevity of their digital work. For example, a major goal of the DiSC project “Tracking Samothrace” is devising a semantically enriched database that can continue to evolve over the coming decades of the ongoing archaeological excavation. Alongside digital scholarship, the Libraries are also preparing to launch OpenEmory, an open-access repository for faculty research. Providing Emory researchers the technological infrastructure for storing their finished work will also insure its continued availability for years to come.
Ultimately, the role that technology and libraries will play in the future of humanities research—and the future of the liberal arts—is a lot like my class’s final exam: an experiment in progress. We’re not yet entirely sure what we’ll learn as we implement new tools in traditional disciplines, but we do know that the collaborative investigation will leave us all the richer for having made the attempt.