The human in digital humanities

I find I often think of the digital humanities as a discipline in which scholars work with digital stuff to answer questions about texts. I know that’s not the full scope of the field, but that’s how I’ve most often interacted with it. And when I do this, I subconsciously apply only a Big Data connotation to the word “digital” — querying and analyzing massive datasets of all the words written by an author, for example, or publishing wiki-style so that the masses can contribute or change what exists. And this habit of defining the digital in these terms alone leads, I think, to Stanley Fish’s worries that the digital humanities are a dehumanizing field. He wrote an opinion piece called “The Digital Humanities and the Transcending of Mortality” in which he talks about the blurring of the individual into the digital collective, resulting in a concept of the individual that is no longer “an entity that can be said to have ownership of either its intentions or its effects.”

Perhaps dissociation from the human — the particular, the subject — is part of the post-postmodern project. But perhaps this worry about dehumanization is an early stage of understanding the digital humanities. One of the greatest changes that happens when you work with digital information is that you gain the capacity to perform large scale computation. That’s what the human brain is less good at, but that’s what computers do without breaking a sweat. So it’s only natural for many of us to subconsciously or consciously define “digital humanities” as highly computational and as primarily preoccupied with massive aggregates even when this definition does not describe all that is going on in the field.

Timothy Burke rejects Fish’s dehumanizing view of the digital humanities in his wonderful essay, The Author is Human. “In DH, authors are not dead, just brought down to human scale,” he writes. Authors aren’t the cultural monoliths of the pre-digital humanities — great minds that dominated the minds and imaginations of their times or ours. Authors are just… authors. In a sea of other authors, each one authoring away, any one author is uniquely that author. It’s just that that author’s intentions and effects do not completely overshadow other authors’ intentions and effects. In a way, I think Burke is saying something similar to what I said about focal flexibility.

My sketch of a Mechanical Turk.

My sketch of a Mechanical Turk.

So maybe the object of study is still fundamentally human, or at least not necessarily dehumanized, but maybe the praxis itself is less authorish than it was before. Maybe there’s often too much resistance in the materials, or too much of the wrong sort of resistance, making the whole enterprise feel less like communing with other minds over time and space and more like fighting with computer code and mechanics. Of course, the real work of human-computer interaction draws on the complementary strengths of mind and algorithm. Shyam Sankar’s TED talk on human computer cooperation demonstrates this beautifully leaves no doubt that even highly computational work still relies heavily on human intelligence and design.

Perhaps the very nature of the work itself, relying as it so often does on teamwork rather than a lone thinker, reduces the sense that the scholarship is as purely Authored. So much of the ethos and identity of the humanities is wrapped up in the idea (the myth?) of individual scholars thinking deeply and carefully about their subjects. With that as the backdrop, assertions like Beth Nowviski’s, that “the great project of humanities computing is the development of a hermeneutic—a concept and practice of interpretation—parallel to that of the dominant, postwar, theory­ driven humanities: a way of performing cultural and aesthetic criticism less through solitary points of view expressed in language, and more in team­based acts of building” sounds uncomfortably de-centering (Said in the Chronicle of Higher Education, and on her blog). It’s easy to feel as if this type of work places less value on the self of authorship and scholarship.

And in this case, any individual’s ownership of intentions and effects really is open to more disruption and dilution by the very nature of working as part of a team. Traditional humanities relied on more than one scholar as well, of course. Acknowledgements sections attest to this, but even without them we know from experience that knowledge doesn’t spring full formed from a brilliant mind in a vacuum. But more and more projects, particularly in the digital humanities  promote team members from the acknowledgements to the author line both in form and in deed. The selfhood of individual authors can no longer be seen (in fact or in myth) as fully owning their intentions and effects.

Even so, truly collaborative projects rely heavily on the spark of individuality that each partner brings, the rich tapestry of backgrounds, experiences, and predispositions that allow one partner to see facets of a project that the other partner would never have seen, or seen too late. This is what we wanted to spark off in our Digital Humanities Speed Dating event.

The human, it seems, is anything but an amorphous digital collective in the world of digital humanities. Both the subject and object of scholarship in the field rely on and celebrate the individual. This does not mean that things are the same, though. Authors are less monolithic things now, it seems, less singular. And right now that feels unsettling, particularly as we’re getting used to the whole idea and while the digital piece is so relatively new that it needs modify the term “humanities.” Perhaps as things settle into normalcy the word “digital” in the name won’t feel so inhuman, or perhaps we’ll dispense with the modifier altogether. However things evolve, though, it will be fascinating to watch the role of the self, the author, the human as it navigates the space of the digital humanities.

Digital Humanities Speed Dating

The director of our Humanities Center instituted a First Fridays informal gathering at our local pub geared toward people who are interested in digital humanities. It’s only met three or four times so far, and I had to miss the first one, but it’s been really great to sit around and listen to people say what projects they’re working on.

At the December gathering one of the profs there said what I think a lot of people were thinking when she admitted that she wasn’t speaking up because she didn’t know enough about the digital side of things to envision what she might do for a digital humanities project. She’s well aware of great digitization projects in her area, and web archives, but beyond digitizing things, what kinds of questions would she want to ask that computers could help her with? On the other side, the CS types said that they knew that computers could be great thesauri, but they didn’t really know enough about what keeps humanists awake at night to really suggest projects.

This seemed like such a fundamental question that we hatched a plan. We set up a “digital humanities speed dating” session. Humanists would come with some description of their own research, and CS-types would listen with their CS-perspective and would also talk about what they were working on. And maybe the more we knew about each other the more we would be able to see collaboration potential.

So this evening we conducted our Speed Dating session, and whatdyaknow, some CS profs and students had highly humanistic interests, and some humanists got really helpful ideas from the CS profs and students. My favorite moment was when one of our classicists was telling us about a set of fragmentary Roman documents that she is really interested in. They’re documents that record the auction prices of quite a few people’s property (they’d been convicted of something or other, and the proceeds were to be in some way presented to whichever god they’d offended). And people were discussing what she might or might not be able to learn from the fragments she has, and one CS prof said it’d be interesting to see if they could infer something about the data that’s missing. You could use each sale as a kind of user rating of the item, he said, and the missing items would be something like the things in Amazon or Netflix that haven’t yet been rated, and you could infer something about their potential value based on what had been rated/priced already. Well, that blew our humanist minds.

So after the gathering the CS prof went to talk to the Classicist a bit more, and then I happened to be there in the hall when he debriefed another CS prof. “Yeah,” he said, “I’m just going to take her spreadsheet and run a ……” and here he and the other guy went into a dialog that I assume wasn’t actually in Greek, but it might have been.

I love it when I come face to face with expertise I can’t even parse. And also, I think that’s part of the point of this whole thing.

THATcamp Denver Takeaways

I’ve been watching blog posts trickle out of THATcamps for years, and this fall I finally got a chance to attend one. As luck would have it, just as Carleton signed onto a Mellon Digital Humanities planning grant with Macalester and St. Olaf, a THATcamp popped up focused specifically on Digital Humanities and Libraries.

What a rich experience. It was populated by such a range of participants: people from large research universities, university presses, centers for digital humanities, OCLC research, major digital libraries, regular old librarians like me, and some disciplinary faculty, and some graduate students. It was a great opportunity to hear from people who are a few years ahead of where we are, to hear what questions they have now, what confusions exist that maybe we can head off before they become issues here, what workflows they’ve arrived at, what kinds of projects may be coming down the pike.

It was also an exhausting experience because the whole time I had to listen not so much to what, exactly, people were saying (since that was usually mostly irrelevant to me) but to patterns underlying the conversations: what kinds of services do institutions provide, are people mostly working in teams or alone, are the teams institutionalized or ad hoc, how do they negotiate the class differences between disciplinary faculty and library or IT partners, where do student research assistants fit into current work, what terminology do people use and where do confusions in terms exist… So many things to listen for and translate into my own context. And often I’d realize that information I’d dismissed earlier as totally irrelevant to me suddenly fit into a new pattern I was noticing, so by the end of the day I was pretty fried.

So what were some of the things that stood out as interesting or important themes or patterns? Well…

Have a Process — Allow for Exceptions

One session was about libraries becoming more involved as publishers since they often house the tools or manage the preservation and access to digital humanities projects. In my case, those tools would probably come from our IT department rather than the library. It was still really interesting, though, to hear the conversation slowly gravitate toward the consensus that in an ideal world institutions would have both a formal publication/preservation process and then also a sort of sandbox publishing process. This second process would accommodate projects that didn’t fit well into the formal containers (journals, books, wikis) or that couldn’t be evaluated in the same way prior to publication (peer review, for example). As one person said, often you have to build a thing before you know if it’s going to be worth anything, so the before-hand evaluation for publication process is largely impossible in these cases.

And in fact this became one of the major themes of the day for me. It seems that in order to support digital humanities projects, the goal is always to have the project based in tools, support structures, and data that are already present and ready for that kind of project. But there are also a lot of projects that simply won’t fit into existing structures, and so many programs provided capacity to work on these “boutique” projects. One woman had a phrase I really liked, saying that she looked at these as “first of a kind projects, rather than one of a kind projects.” The idea is to use these boutique projects to develop tools or services that the become part of the standard suite of support options, therefore taking less effort the next time.

Collaboration is Difficult and Necessary

This has been a theme not just of this gathering, but of every digital-humanities-related gathering I’ve attended or heard of. One humanist at St. Olaf said point blank that the humanities have traditionally been a realm of solo work so humanists don’t really know how to play well with others. They lose control of the process, the timeline, and the outcome, and the conclusions aren’t necessarily fully their own. Another humanist at a conference hosted this fall at Carleton spoke bluntly about how this “go it alone” attitude can easily translated into almost a master/servant relationship when collaborations mix faculty and staff because this relationship fits more easily into disciplinary practices that privilege solo work. “This is my project. Please make this thing happen in the way I want.”

Library Disciplinarity

While I was at THATcamp, I started thinking that one way to think about these projects so that they maintained their collaborative nature would be to think of it almost as interdisciplinary work, where the library is a discipline more than it is a service. The library has hundreds of years of disciplinary experience with finding, organizing, using, preserving, and making accessible digital and physical items. We study the rich complexity of the interstitial spaces between discrete pieces of instantiated knowledge. We study the socio-cultural practices that develop in these spaces and that draw dynamic connections between the items. To think of ourselves as “just” a service organization is to sell ourselves short and to relegate ourselves to perpetual outsiders in these ideally collaborative pursuits.

For this reason, I also really appreciated the ways that some institutions have “Digital Scholarship” programs rather than “Digital Humanities” programs. From the CS and Library perspectives, we bring very similar portions of our disciplinary training to projects from the Humanities and the Social Sciences, so there is really no need to have separate programs to support the various disciplines. (No need, that is, after the humanities have had a chance to acclimatize themselves to this new realm of scholarship. I can absolutely see how having dedicated energies directed to the humanities is important when first getting digital humanities off the ground.)

Some Practicalities

People at THATcamp were justifiably concerned about the sustainability and scalability of their programs. Some had decided that they would provide consulting support only or that they would cap the number of active digital humanities projects they could support at any given time (with a wait list for people once an active project graduated or died). Several found that they were in high demand to help scholars develop data management plans now that humanities scholars were applying for grants that require those, and that this had been an unanticipated and heavy demand on their time. And other libraries pointed out that rather than being all things to all projects, they would specialize — so Library A might specialize in text analysis and Library B might specialize in GIS applications for humanities. Most make use of graduate student work (we intend to employ undergraduates here).

It also became clear that initial conversations in the planning stages for a project should include some discussion of how permanent the end result is intended to be. Is this something closer to an item in our “circulating collection” that can be weeded when use dwindles and space gets tight? Or is this more like an item in “special collections” that requires greater commitment to long-term care? Who will be responsible for the server space and upgrades? Who is the intended audience and how will we get it to them? If it is interactive, how much of a change log should be available, and how much control do we exert over the kind and type of interactivity? Questions like this are not just procedural, it turns out, but can actually shape the project itself on a fundamental level.

And finally, it turns out that most projects need (but do not have) good graphic design. This is especially important for projects that involve crowdsourcing. And if you’re after interactivity, you’d best have thought about how to get people to participate (I’m reminded of Riedl’s research for this problem).

Once is a fluke, three times is a trend: Digital humanities arrive in my world

I’d kind of expected a slow ramp-up. After all, this is the humanities, and as a humanist myself I know that change does happen, but only as long as you don’t watch for it too hard, kind of like watching water boil.  This term, the water boiled.

Within the space of two weeks, three students from three different language & literature majors came to me for advice on scanning, OCR, conventions of web design, hypertext literature, digital annotations, and (of course) copyright. For two of these, the students were developing senior thesis projects outside of the usual templates of a research paper or essay exam.

I foresee discussions about a coordinated support model for student work in the digital humanities in my near future. Luckily, the data services group has been meeting for a couple of years already and probably has some good advice about mounting a service that requires input from multiple departments on campus, so hopefully we can crib from their work.