Why not build another digital humanities tool?

One of the recurrent issues I noticed when our Praxis cohort began discussing the meaning of the digital humanities was the field’s need to justify its existence. At the beginning of the semester, we read articles about digital humanities as a “tactical term” and the kind of institutional, financial affiliations necessary to sustain DH labs and staff. All of this background on the history of the field proved useful in understanding where DH stands within academia and why it has received so much recent criticism as an instrument in the further neoliberalization (insert your personal definition here) of the university. Scholars within and outside the field have noted, with some justification, how DH might further the financialization and quantification of, well, everything and how that might lead us further away from the inquiries that should ostensibly drive our intellectual pursuits in the humanities.
So, if there is this “dark side” of DH, then what does it look like and what can we do to push against it?
An emphasis on empirical, quantitative projects and the “mining” of data and texts has characterized many (most?) DH projects, creating easy fodder for critics concerned with the ever-growing reach of neoliberalism. These types of projects have served as the very foundation for many DH literary studies and have yield generative results into topics like genre devices, authorial style, and the history of the novel. But beyond yielding new insights into literary texts, these types of projects offered a way to justify the field of DH. If humanities scholars could use computational analysis to produce projects illustrated with graphs, charts, and numerical data, then perhaps the humanities could maintain its relevance in a political/academic climate that seemingly values STEM over the arts. 
The other way that DH seems to claim its relevancy is by emphasizing the production of tools. Our own Scholars’ Lab staff has created new tools like Neatline and previous Praxis cohorts have developed Prism and Ivanhoe, making useful, open-sourced tools and bringing attention to the work being done here at UVA. Of course, there’s no problem with producing such tools and many will say that’s exactly what DH should do. The problem will arise if every DH lab in the world starts producing new tools for every new project. At that point, the future of DH might look a lot like the iTunes app store, more a marketplace of things rather than ideas.
So, will this year’s Praxis cohort create a slick new tool for the DH marketplace? You guessed it, no.
For one reason, our cohort simply does not have the time or expertise to build the coolest new tool on the web. Several of us entered this program with little to no knowledge of coding, web design, or software development. My goal is to gain an introduction to if not a handle on those skills while I’m here.
The other reason is political.
During a DH presentation at the 2016 American Studies Association conference, I heard a panelist describe the Internet as the “island of misfit toys.” I took this to mean that in a push for innovation and the need to justify the field’s existence a proliferation of DH projects have led to a mass of broken, unmaintained, and otherwise unusable tools. This stress on tool creation has fed into the critiques of DH as a neoliberal undertaking bent on the production of quantifiable and marketable results rather than the humanistic analysis that might yield more ideas than products.
Our cohort won’t be producing a fancy new tool, and that’s okay. But we will use existing digital tools in innovative ways and maybe offer inspiration for other scholars with similar amounts of time and DH experience to pursue projects of their own.

Joseph Thompson is a doctoral candidate in the Corcoran Department of History. His dissertation, "Sounding Southern: Music, Militarism, and the Making of the Sunbelt," uses music to examine the cultural impact of the military-industrial complex since the 1950s. This analysis draws on methods of cultural history, political history, and sound studies to consider the shifting meanings…


  1. I have always felt that the Praxis Program was too focused on the building of tools than the using of tools, and this is mainly the result of the skill sets and interests of the Scholars’ Lab developers. Web application skills might be useful to some people, but HTML and CSS are largely irrelevant to most humanities scholars using computational methodologies in their research or publications. On a more abstract level, one of the most useful skills you can learn in DH is how to model knowledge into information/data. What are the sorts of questions that you want answer in your research? How do you model information in a way that enables to to ask these questions? What tools are already available that will enable you to query or visualize your data to answer your questions and pose new ones that hadn’t occurred to you? These tools vary from discipline to discipline, but the fundamentals of knowledge representation are broadly applicable. So as you think about the type of project you want to work on in the spring, consider using existing data and existing tools (with just enough coding to string together an array of tools) to offer new interpretations. You will end up learning methodologies that will benefit you more in the long term.

  2. Perhaps fix some of the broken toys?

    My view is that many DH software programs end up broken because of design weaknesses that make them hard to maintain. They are usually too integrated and insufficiently modular. They lack clearly defined architural layers. They try to do too much and end up doing many things poorly.

    Generally these tools are built originally for a specific project and lack organisational support for ongoing maintenance. But some of them are worth keeping, if they can be refactored into a more reusable and maintainable shape. It’s easy to knock up a prototype, but there’s more value long-term in critical analysis and rework.

    • Thanks for your comment, Conal. You make great points. We may, in fact, end up fixing a discarded tool if we find one that fits what we want to do. Honestly, we are in the beginning stages of our work and are looking around for what might be of use to us. Announcing that we’re not building a tool is more about defining the parameters of our project than anything else. Check back with us soon!