Blog Post

Reading, Close and Distant

Image of three books open on a table with stacks of books in the background.

Matt Shoemaker of the Temple University Digital Scholarship Center was giving us newbies an intro to Voyant last week. At one point, he talked about using digital text analysis tools as a form of "distance reading," in contrast to the "close reading" of old school, manual texual analysis we are so used to in the humanities and humanistic social scientists. Distance reading and close reading provide us different types of information about our sample and both are incredibly useful--I'd argue necessary--to understand what's "going on" in a corpus. Distance reading provides us with the bird's eye view, the lay of the land, the forest; close reading gives us the chipmunk's perspective, the feel of the soil, the trees. Neither are incorrect, but nor are they complete on their own. 

I'm not sure what it was (Matt's easy delivery of the information? My self-absorption? Mars conjuncting Pluto in Capricorn?), but something finally clicked for me. I've been struggling recently with this tension between the distant and the close views. My academic grounding in Philosophy fostered my love of close, textual analysis. I could (and have) rambled on for many, many minutes about the placement of a particularly punctuation mark or a single word choice. The humanities have definitely given me space to love language and think about the choices that we make to best express our thoughts not only in the content, but the form of our writing. That said, after five years in Sociology, my interpellation as a social scientists is more encompassing than I realized. I've been bristling at how such close readings can miss so much context outside the texts that are also imbued in the text itself. How does this text compare to others like it? To others unlike it? What happened in the world the day those words were written? What happened after that has changed how we understand those words?

This tension is also at the heart of my dissertation. I'm curious about the shape of transgender literature and how it has evolved. I'm particulalry curious about the influence of the Lambda Literary Awards in this process. To satisfy these curiousities I need to know what is in the books themselves--close up and from a distance--but I also need to know what's happening at an organizational level with the Lambda Awards, as well as what's happening in the wider LGBTQ movement. I'd been struggling to articulate this tension in a way that made sense, but was also easy to explain to others. Thank you, Matt.

On a less existential note, I'm excited to explore Voyant more. I plan on using topic modeling as my main method for digging in to my corpus, but I love the different options that Voyant provides, especially for visualizing the data (I'm a sucker for word clouds). The combination of so many tools makes it useful for getting into the data and just mucking around a bit. I'll be sure to post about what I dredge up once I get in there and get my hands dirty.  



Great ideas, AJ. Even as a sociologist, you don't have to give up your close reading appreciation of language and its subtleties, of course. And much of the really experimental work being done now in textual analysis involves a recursive pattern of distant reading to identify patterns, close reading of passages to validate or modify the patterns, leading to an adjustment of the distant reading parameters, then more scrutiny of passages ... I share you frustration with analyses that limit themselves to close reading, but those skills have served us well in the many decades when they were the only option. The situation is changed, and to me, combining them with qualitative methods is very exciting.


Two thoughts about this:

1. Voyant is indeed a great set of tools to experiment with; the results can be thought-provoking.

2. "Close" in "close reading" is probably not the opposite of "distant." See "Re: Search and Close Reading" at


Thanks for the resource suggestion. I'm looking forward to reading it!