I have a confession: I’m not good at reading the directions. That’s not always a bad thing, because sometimes I learn more, and better, about how to use a tool by teaching myself than by following the instruction manual to the letter. However, this week was different. Using Voyant, an online tool for digital text analysis, without a concept of distant reading, text analysis, visualization, and their implications for studying history only resulted in me not getting the point. At all.
Voyant has a very basic setup, and was clearly built by someone with a very different brain cell structure than mine. There is no obvious tutorial, but only a “find out more” link to a fairly technical user manual. It’s not really a tool for a casual user; you have to be willing to sit down and read through the methodological theory and case studies for using Voyant before you really get the point. At least if you’re me. However, after I read through the user manual (actually, it’s a book), and after reading Moretti’s Graphs, Maps, and Trees, what I got was this: Heremeneuti.ca, the overall project that Voyant is a part of, is about applying computing to humanities research, allowing scholars to use the kind of tools previously thought to be the mien of NASA scientists and macroeconomists to research and think about fields like history and literature.
The whole field of digital humanities sometimes reminds me of those perspective-bending images where at first you see one thing, but then, if you look closer, it appears to be something completely different. The rabbit-duck illusion specifically comes to mind when thinking about distant reading, digital humanities, and the possibilities of using big, powerful technology to research history. I think many mainstream academic historians are accustomed to looking for the metaphorical ducks when doing research that the only tools they use are breadcrumbs and the only place they look are ponds. So often, historians miss the big fat rabbit sitting in the middle of their garden, because all they’re looking for is ducks.
This analogy is getting away from me, but what I’m trying to say is what most digital humanists have been (once again, metaphorically) shouting at the top of their lungs: “We built this for you! You can use it! It does really cool stuff if you’ll just read the manual!” And it’s true, Voyant does really cool, really valuable stuff. For example, I looked up the lynching data that the Equal Justice Initiative recently released, and found that the EJI had not only their data, but links to several different websites that reported on their findings. To test Voyant out, I entered the URLs for the lynching report as well as several news stories from a variety of different media outlets, mainly liberal and conservative and U.S.-based and foreign (though all English language). Voyant returned a fascinating analysis of the words used by the news stories as compared to the original report, tracking and comparing frequency of language usage. Though my little project didn’t constitute a hugely comprehensive analysis on how the news media reports on a really atrocious piece of American history, it did demonstrate how useful a tool like Voyant is to humanities scholars.
I have to say, I’m pretty excited about the possibilities open to historians who aren’t afraid to explore digital resources and look for rabbits where their colleagues are seeing ducks. Tools like Voyant offer a paradigm shift that’s badly needed if historians want to stay (become?) relevant to this brave new world where previously unimaginable loads of data are available to almost anyone with an Internet connection and enough spare time. Truth is, we need historians to get on board, as they have the training and skills to really do something beautiful using digital technologies. So come on, historians: let’s hunt some rabbits.
I think the next question is how do we assuage the fear? How do we get the historians and literary theorists on board with these really cool things that have been built for them? Do we have to wait a generation for the old guard to die off and some hypothetical bright and shining new vanguard to arise?
I think its an Albert Einstein quote, that if you can’t explain it simply you don’t really understand it. How do we explain it simply? Things like Voyant are complicated structures and they do complicated things – how can we simplify the entre so that the tentative can become the acolyte?
Excepting Wikipedia (and Flickr, but not flickr commons), everything we’ve examined has been new to me, does anyone know how these resources are marketed? I’m curious to see how the developers are trying to gain their audience – what, if any, training sessions they offer, outreach opportunities? How do they engage with their audience outside of the interface?
I really think it’s up to those of use who walk between two worlds, as it were, to introduce a lot of this stuff. Dr. Richard Marciano, who just joined UMD as the head of digital curation, is a great example. His digital curation lab works with programmers and historians to use big data for historical analysis and research. I’m very excited to be at UMD right now, because it seems like the iSchool is doing a lot to cross those boundaries (inter-departmental woes notwithstanding) and to get historians to see the value of digital resources.
I completely agree that more people—and scholars–need to get on board with these digital resources; and I fully admit that while excited by their possibilities, they often confuse me immensely and that confusion largely keeps me away from investigating/using them. Because of this, it never really dawned on me that a reason some scholars choose not to engage fully (or at all) with these resources is that they don’t believe digital methods will uncover new evidence (not just because, like me, they’re wary of complicated technology). Underwood brings this up in his post, “We Don’t Already Understand the Broad Outlines of Literary History.” He argues, instead, that it’s precisely that–new discoveries and understandings of history and our own scholarship–that makes digital methods and resources so valuable. He writes: “I think we’ll discover something similar as we explore categories like point of view and genre. We may start out trying to recognize known categories, like first-person narration. But when you sort a large collection into categories, the collection eventually pushes back on your categories as much as the categories illuminate the collection.” That last sentence really resonated with me. Simply exploring sources and history in these ways, and with these new digital tools, can–and should–reveal a different story, in exactly the same way that any category of analysis employed by historians in reading their sources should. Just by doing the work to “define categories” for text mining or digital exploration, we are revealing the rabbit!
I also agree though, that this still doesn’t resolve the technological complications and fear—I’m still overwhelmed myself. When it’s put into the terms outlined by Underwood (and Julie–thank you, super helpful), however, my interest builds and begins to override my fears. When it’s directly applied to the kind of history I do, I’m completely convinced. I have also been reading the EJI “Lynching in America” report, related articles, and Jim Crow violence/lynching mapping projects avidly. These kinds of projects–and even more so the public response on social media, etc. they have brought on–directly shows how these digital technologies can engage both scholars and the public by illuminating new historical understandings AND encouraging public debates about the legacy of racial violence within America. Are these public conversations, social media debates, etc. marketing in themselves for these kinds of projects? Just providing the public and scholars with direct, tangible examples of what these digital methods can do in itself invites exploration of digital history techniques. So how can we learn from this? I like @C.E. Bloom’s question about marketing, and it made me wonder, to what extent can we engage scholars themselves in creating these projects and applying these methods? If we want historians and other scholars to get on board with these techniques, should we first ask them what topics, questions, etc. they have? How can we take advantage of that information to make these digital history projects bridge the gap between academic scholarship and public discourse–and thus encourage new understandings and debate?