The Digital Humanities: An Introduction for Techies

A guest post by Aditi Muralidharan


I'm Aditi, and I'm a PhD student at the department of computer science at UC Berkeley. I work in the digital humanities, but when people ask me what I do, I rarely use those words. I say "natural language processing", or "HCI". This is partly because nobody in the Berkeley CS department knows what the digital humanities are, but mostly because there's a strong stigma associated with the word 'humanities' around here, digital or otherwise: it's the soft option, not 'real' research, has no substance. Nevertheless, I'm going to come out and admit it here - I do digital humanities. I enthusiastically work on applying computational tools to problems from the humanities.

My expertise is in the areas of natural language processing and human-computer interaction, and I spend my time as a researcher thinking about user interfaces for exploring large collections of text. For decades now, the dominant way of thinking about design in human-computer interaction is as a user-centered process. We begin the design process by first observing users, understanding their needs, and then successively prototyping solutions, taking user feedback at each iteration.

This is the angle from which I approached the digital humanities: I want to build interfaces for exploring text, and my users, real people who explore text on a daily basis, are journalists, historians and literary scholars. I could just as easily have ended up in the field of computational journalism, which is the application of computational methods to the problems of journalists and journalism, but I happened to get interested in the humanities first.

The digital humanities brings together a lot of different kinds of people, solving all kinds of different problems, but there are a few main challenges that regularly come up.

1. Managing digital collections e.g. library catalogs, image collections, electronic records of archeological artifacts, and other kinds of digitized data.
2. Using technology in education.
3. Augmenting existing sources with structured electronic information e.g. creating linked datasets, geo-tagging, and mapping.
4. Analyzing already digitized collections on a large scale e.g. data mining, text mining.
5. Getting tenure.

Number 4 above is where I come in, and I'm not the only traditional computer scientist interested in this stuff. Google recently announced 12 awards of $50,000 to proposals for large scale digital humanities analysis centered around Google Books.

An example of a digital humanities project is the one I'm working on right now. I'm focusing on a set of slave narratives published before the civil war, a collection of significant interest to literary scholars and historians. By careful examination of the texts, they have identified overwhelming stylistic similarities among these narratives attributable, perhaps to the political abolitionist agendas of the white publishers of these works. My goal is to build an interface and text mining tools that will allow the exploration of the stylistic features of these texts on a much larger scale. One that will hopefully allow the scholars I'm working with to spot other patterns not visible through manual examination.
The text-analysis sub-sphere I inhabit is getting increasingly populous: other research groups I know are finding thematic trends in 18th century novels, extracting the names of places, people and things from ancient Greek and Latin texts, and automatically classfying a huge collection of Scandinavian folklore into categories based on text content.

The digital humanities seems to be fueled by a lot of excited people staying in touch with each other through twitter, blogging, and un-conferences. The "digital humanities" search stream on twitter is a good place to join the conversation, and the aggregating blog at Ditigal Humanities Now is the best way to keep current with what digital humanities bloggers are up to. I've also joined the crowd with my own blog, where I write about digital humanities from a computer scientist's point of view. The most popular avenue for meeting other researchers with similar interests is THATCamp the Humanities and technology camp. There's a main event in Virginia every year, and regional, and even international versions almost every month of the year in all kinds of places. The annual Digital Humanities conference is a more serious venue for scholarly work.

I have found the digital humanities a very welcoming community. And even though I still have no idea what "hermeneuetics" is (despite patient explanations from many wise people), it has been no barrier to productive collaboration.