Digital Dissertation Workshop

Silhouette Letter WriterAmid the whirl of data visualization, digital pedagogy, network analysis, text mining projects, and big data vs. small data debates that energized Boston’s inaugural “Days of Digital Humanities” conference last week, The Junto caught up with three early Americanists using new media to complete traditional dissertation work: Erin Bartram (University of Connecticut), Jean Bauer (University of Virginia), and Lincoln Mullen (Brandeis University). Here’s what they had to say about the process, and how digital research will shape their work after the dissertation.

JUNTO: At what point in researching your dissertation topic, did a “digital” light bulb go off? Can you describe how you chose to incorporate digital research methods or new technology in your work, and how your colleagues reacted to that idea?

BAUER: I was already working on other digital history projects, designing prosopographical databases for The Dolley Madison Digital Edition and Documents Compass when my original plans to do an intellectual history of early U.S. Foreign Policy began to unravel. As I became more and more interested in the communication networks and day-to-day experience of American diplomats and consuls, I realized that the work I was doing for my summer jobs could translate into my own research interests.

BARTRAM: While I think I was always looking for ways that technology could make my note-taking, organization, and writing processes smoother (and keep me from forgetting things), I have been pretty conventional in terms of the product of my research. A few of my colleagues at UConn are working on digital projects separate from their dissertations, so when Lincoln and I first started talking about creating the database, I had some people to bounce ideas off of.

MULLEN: In graduate seminars I read any number of social histories that used a rich set of sources that were completely inaccessible to the reader, because the data sets were not published along with the book. One of the promises of digital work is the ability to make our sources more available to readers, so that was one of the goals for my dissertation. Since I wanted to write about 19th-century conversions to as many religions or denominations as possible, I also knew that I would need some method of keeping track of the many people. My prospectus promises a methodological appendix to the dissertation explaining the digital methods I would use in the course of research. I held off constructing the database until I better knew the research questions I wanted it to help me answer. But what really spurred me to get started was the opportunity to collaborate with Erin and other historians. But for me, there isn’t a binary between digital and traditional. As I’ve suggested elsewhere, unless you’re taking notes on 3″x5″ cards and use card catalogs exclusively, we’re all using digital methods to one degree or another. It’s just a question of where each historian feels comfortable working on the spectrum of digital methods.

JUNTO: You have each developed a significant new tool for other researchers to use. Erin and Lincoln, can you describe the American Converts database, and any features our readers should check out right now? Jean, can you outline the Early American Foreign Service database and say a little bit about the different kinds of software you’ve experimented with?

BARTRAM: The American Converts database is a collaborative project that allows scholars to, in Lincoln’s words at the DH conference on 3/15, share their research notes. The goal is that scholars will contribute their information on American converts and conversion experiences, building on and helping refine the contributions of others. Researchers can sort converts and conversion experiences by a variety of categories and can discern patterns and connections among the records contained in the database. Lincoln and I chose Omeka rather than build a database from the ground up primarily for ease and speed of getting the project going. While we both worked together to conceptualize the way information would be organized, Lincoln’s done much more of the behind the scenes work, like his recent creation of plugins that allow you to search things like WorldCat and Google Scholar from any record in the database.

MULLEN: If Junto readers want to check out the database, I’d suggest that they start with some of the featured people on the front page (here is Warder Cresson, who makes for a great story). They might also look at the collection of conversion relations from Pawtucket Congregational Church from 1856-68, which are my notes from an unprocessed and previously unknown manuscript collection that the archivists at the Congregational Library were kind enough to let me look. They might try some sample queries, such as this search for conversions to Roman Catholicism. Finally, I hope they’ll send us an e-mail and start contributing their own research.We have a long way to go with the American Converts Database. It will never be comprehensive, but there are many thousands of records yet to add. Erin, our collaborators, and I are working on adding converts we’ve found in archival sources and in our secondary readings. I’m also working on searching databases like the Library of Congress’s Chronicling America to add other converts (for example, here is an interesting conversion of an African American Baptist to Judaism). We also have a number of features to develop for the database. We intend to link database records together, for example, connecting converts who knew one another or who read one another’s works. Soon we’ll create a timeline and map of the conversions. Eventually we hope to do network analysis and make other kinds of use of the data. Many of these features come courtesy of the amazing developers at the Roy Rosenzweig Center for History and New Media at George Mason or the Scholar’s Lab at the UVA Library. Others we will have to write ourselves, such as an Omeka plugin that I’m developing to the people in our database to sources about them in library catalogs, ArchiveGrid, JSTOR, and other repositories.

BAUER: The Early American Foreign Service Database (EAFSD to its friends) is an open access repository for information about the members of the American diplomatic and consular corps from 1775 to 1825. You can search for people, assignment titles, and locations or visit the Explore page to play with visual indexes based off live data. The EAFSD is also a publication platform for long-format essays which rely heavily on data visualizations to make their arguments (the latter part is a new feature, so it isn’t fully implemented yet). The EAFSD runs on Project Quincy , an open source Django and MySQL software package I wrote to trace historical networks through time and space. Most of the current data visualizations are written in D3.js , a javascript library. I am indebted to many, many people at the University of Virginia (especially Scholars’ Lab and the Brown University Library for their expert advice and assistance. If you do visit the site, please read the Acknowledgments page to see how many people it takes to build something like this.

JUNTO: How does presenting your research in-progress online–rather than in chiefly traditional settings like conferences and seminars–change your notion of audience as you write?

BAUER: Not all that much. As the daughter of a novelist, I have always tried to write for the intelligent non-specialist as well as the academic insider. But I do have two separate “professional” voices: the historian and the digital humanist. My historian voice is third person, avoids buzz words like the plague, and relies heavily on literary styles to convey the epic and human qualities of my diplomats and consuls and the world they lived in. In contrast, my digital humanist voice is more conversational, almost always first person, and can become very technical. Hopefully both are funny. As more and more of the database comes into the dissertation I am learning to blend these voices into one.

MULLEN: The records or notes for the American Converts Database tend to be factual rather than interpretative: dates of conversions, transcripts of sources, and the like. I’d be less comfortable posting my first shots at interpretation online. The main benefit for me of putting research online is getting an audience sooner in the process. Dissertations are tough because writing them can feel lonely and there are long periods without feedback from people outside your committee. But it’s tremendously energizing to post work and have people respond to it, even if all they say is that they want more of the same. As a side benefit, posting work online is a way to challenge the absurd professional tic of stamping our drafts “do not cite or circulate,” as if citations and more readers were the worst possible fate that could befall our scholarship.

BARTRAM: Conferences are excellent, and quite often the research presented is still in progress in many ways. The database itself was conceived of at this past AHA, where Lincoln and I were on a panel together on the history of conversion in America, and came out of a realization of how many good connections were made in that room. I think the database simply allows scholars to connect even earlier in the process; in some ways, this puts all of us together in the same reading room, the same study carrel, and the same coffee shop, sorting our notes out together. In that sense, it’s a bit more intimate than we’re used to, despite the fact that we’re all at computers miles away from each other. But I think it’s also a very friendly space, because the best thing that can happen to one of your records is to have someone correct it or complete it.

JUNTO: According to recent criticism, the traditional dissertation is “broken,” a prolonged academic rite that is more than ready for reinvention. Do you think adding a digital component streamlines/slows down the dissertation process? How does the current academic infrastructure need to change in order to accommodate the needs and results of the digital dissertation-writer?

MULLEN: Among the views in the article you cite, I come down squarely in Anthony Grafton’s camp. Grafton and Jim Grossman make similar arguments in their well-known article, “No More Plan B.” I’m experiencing what Grafton describes—the learning that comes from creating the extended research and argument that is a dissertation. I’m not sure that purely digital work could replace that: in any case, a purely digital dissertation isn’t something that I want to do. Just as laundry machines and dish washers don’t really reduce the amount of housework we need to do, digital components to dissertations don’t reduce the total amount work. If you did an accounting of time, the net result is probably more work. But they do allow for better work, or work of a different kind. The case can be made for regarding a digital project as a separate piece of research, or even a publication, in addition to the dissertation. The American Converts Database is currently hosted on a commercial server, but the fine folks at Brandeis’s Library and Technology Services are working on humanities research computing. Once that’s available, we’ll migrate to their servers. That kind of institutional support can be indispensable to a digital project, especially if it involves software like Geoserver. Though not everyone want to learn this way, there is a strong do-it-yourself tradition in digital humanities, so one of the simplest things an institution can do is have an up-to-date stock of technical books in the library.

BAUER: Creating the EAFSD certainly extended my time-to-degree, but I wouldn’t give it up for anything. I need the tools I’ve built to perform my analysis of correspondence networks and geographic expansion, but I have truly enjoyed building it and look forward to improving it for years to come. On the most basic level, I don’t know of any university libraries equipped to store digital dissertations in their institutional repositories. I’m part of conversations on how to make this happen at Brown University and the University of Virginia. It will happen, but until universities can accept and maintain a non-print format dissertation, widespread adoption will be tricky.

BARTRAM: I think digital advances on the back end of the dissertation, like research organizing systems, software for analysis, and even our database, can do a lot to streamline a dissertation. But feasibility and acceptance of a digital product, even a digital component to a larger project, is much spottier. And I do think there’s an idea floating around out there that technology makes it easier, universally, so that graduate students using digital methods are seen as “not putting in the hard work.” I think that dismisses the conceptual labor that goes into a digital project, and even the conceptual labor that goes into choosing the note-organizing software you want to use. I think the main issue is that the dissertation is understood by most to be part of a recognized, repeatable progression towards tenure (the concept of which isn’t looking to healthy either), and “disrupting” this part of the chain has ramifications that lots of people need to deal with. But deal with it they must, because this is something that enriches our work. I think many advisers, even those who are interested in digital methods themselves, pause at encouraging this kind of work for their students because they don’t want those students to lose out further down the road simply because someone they encounter doesn’t consider the work “valid.” That needs to change.

JUNTO: Finally, can you reflect a bit on how you’ll carry these digital skills forward after the dissertation, i.e. in future projects and teaching?

BARTRAM: I was actually interested in digital methods on the teaching end before I became interested with them on the research/scholarship end. That being said, thinking through a collaborative database like this has given me a lot of good ideas for wrangling a similar collaborative project in the classroom. One thing I want to keep in mind, though, is that I don’t want the lack of access to technology to ever prevent a student of mine from learning the historical skills that are at the core of what we do.

BAUER: I am the Digital Humanities Librarian at Brown University. I got this dream job based on the EAFSD and the skills I learned to build it. I reference it in almost all of my workshops on digital humanities, database design, and network analysis. I also have a few ideas for how to teach American history by having students add their research to the system, and I look forward to putting that into practice some day…

MULLEN: A digital project is different than a dissertation in that at least you can finish a dissertation. (Or so I’m told …) Though the second research project in the back of my mind is a history of children’s religious experience, or maybe a biography of Lorenzo Dow, I hope that this database will be a project that extends beyond the dissertation and book so that it becomes a base for other scholars’ work. In spring 2014 I’ll be teaching a course with the somewhat ponderous title, “Nineteenth-Century American Religion: A Digital History Seminar” (draft syllabus). As I’ve described the course, the students’ main work will be to create a digital map of religion in the greater Boston area over the course of the nineteenth-century. I’m hoping that this “history class as shop class” will be fun and an experience that will teach students both digital and traditional historical skills.


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: