Posts in category: DH

I recently returned from the DH 2014 conference in Lausanne. I went to give my brief (my brief brief) for getting serious about social science methods in digital humanities under the title “Let DH Be Sociological!” The conference offered plenty to think about on this theme. Also I got accused of wanting to “dumb down” an entire field of study…so that’s got me thinking too.

Update, July 11, 2014: my slides are here: slides.pdf. The R code used to generate material in the slides is available in a gist.

Update, July 2026: the online archive of abstracts from the conference has been ruined, so I removed links my and others’ abstracts. They can be found, however, in the PDF of the proceedings.

I’m attending the Digital Humanities 2014 conference in Lausanne next week. If you’re around Lausanne next Friday, my talk, titled “Let DH Be Sociological!,” is part of a panel at 11 a.m. in 415 Amphimax. This paper builds on my work using topic models to analyze the history of literary scholarship, but takes a more polemical tack. I’ll use my ten minutes to argue that the moment of “DH” represents a critical opportunity for bringing literary studies (and perhaps other humanities) closer to the social sciences.

My article with Ted Underwood on the long-term history of literary studies is now available in a preprint form (pdf). This essay will appear in NLH, but the publisher has kindly given permission for us to make a submitted manuscript available now. Please keep your eyes open for the final published version. It’s been a long (though enjoyable) journey from our initial exploratory blog post in late 2012 to this essay.

The article’s primary evidence is a topic model of seven scholarly journals in literary studies that span the last century and a bit more. In order to make this form of evidence more accessible, I designed a website that allows you to explore the model. Ted and I spent a long time exploring the many, many different ways you can slice and dice topic models before arriving at the claims we make in the essay. To get a sense of how we might have gotten there, spend some time surfing the model, seeing both its power to reveal and its inevitable froth of randomness and approximation. And, hopefully, you’ll discover patterns of your own.

What I’m most proud of in our essay is our effort to show why our approach responds to central questions in literary history and literary theory. We are interested in topic modeling in itself, but I see our essay as aiming primarily at two things. We try to revise literary scholarship’s conception of its own history. And we argue that quantitative methodologies in literary history, far from dispensing with interpretation, attempt to scale up the interpretation of cultural texts while keeping the major problems of interpretation in view. This scaling-up tends to bring humanistic scholarship into closer proximity to the challenges and methodologies of the social sciences.

I’m planning to expand on that last claim in my talk at DH 2014…. Ted’s and my joint formulations are in the essay. Read it, and browse the website.

After the jump, the rest of this post is all geeking out about programming and workflows.

Tomorrow (Wednesday, January 29) is the Rutgers Digital Humanities Showcase. It’s a packed schedule of presentations demonstrating digital humanities work from all around the university. If you’re in the area, stop by Alexander Library’s Teleconference Lecture Hall. We’ll run from 2 p.m. to 6:30 p.m. Full schedule at dh.rutgers.edu/showcase. Free food at the end!

I’ll be giving a highly compressed demo of some of my work on topic-modeling and the history of literary studies. But the main reason to come is the revelation of the broad range of exciting digital and humanistic work that is in full swing at Rutgers.

Let’s talk about quantitative literary history and where you can find the best tacos al pastor.

I’ve been spending some time recently researching trends in the history of literary scholarship. For MLA this year, for a panel on “Seeing with Numbers” with Hoyt Long, Richard So, Matt Jockers, and Amy Hungerford, I was thinking about the drive to include more texts among our objects of study. On the panel we focused on efforts to see more of literary history by using quantitative methods.¹ But as I was working on my analysis of the changing (or rather unchanging) distribution of scholarly attention in modernist studies, it occurred to me that it is not only the quantifiers in literary studies who articulate the impulse to include more texts.

Tomorrow at MLA, I’m participating in a panel on quantitative literary history and the sociology of literature. You can call it digital humanities if you want! You can call it modernist studies if you want! Personally, I call it…

Seeing with Numbers: Macroanalytic and Sociological Approaches to Literary Exclusion

Friday, January 10, 2014
10:15–11:30 a.m., Chicago F, Chicago Marriott
Presiding: Richard Jean So, University of Chicago

1. Modernism’s Limits: Patterns of Exclusion in Scholarly Reading
   Andrew Goldstone

2. Finding Thomas Curtis Clark: Topic Modeling the Rules of Exclusion in American Modernist Poetry
   Richard Jean So and Hoyt Long

3. Response
   Matthew Jockers

4. Response
   Amy Hungerford

I will post my slides on this site before (shortly before) the talk at andrewgoldstone.com/mla2014.

Over on dh.rutgers.edu, I’ve posted extensive notes from last week’s Empowerment Part II workshop on plain text formats like markdown, HTML, and LaTeX. I’ve also made available the set of sample files I used in the workshop, plus a few more. Those might be useful for others who might wish to learn any of those three languages or see what files converted from one to the other look like. I’ve put some time into making the workshop notes into something someone could work through on their own. For more about the files, see the workshop notes. For anyone interested in LaTeX, the template.tex file contained in that set of sample files supplies a starter template for a LaTeX document (it’s derived from the templates in my git repository of tex files). As a bonus I have also made available the markdown source used to generate both that workshop-notes webpage and the slideshow for the in-person version.¹

You might reasonably ask: Why spend any time on this? Computer typesetting and the ins and outs of plain-text composition gadgets are very far from the center of the digital humanities. HTML is an important computer language in DH, but LaTeX is no such thing, and markdown is pretty rudimentary-seeming. Some thoughts on this over the fold…

Tomorrow I’m leading a digital humanities workshop at Rutgers.

Empowerment Part II: Digital Text

Wednesday, November 20, 1:10-3:10 p.m.
Alexander Library, Room 413
169 College Avenue, New Brunswick, NJ

This workshop aims to expand our horizons for thinking about how we handle text on our computers. In order to attain liberation from Word, we will explore the difference between text editors and word processors, discuss the ways computers represent text as content or form, and experiment with some key technologies for digital document preparation. We will dally with three related computer languages in rapid succession. We will begin with markdown, a minimal but versatile set of plain-text conventions. Then we will learn to convert markdown into equivalent HTML markup using Pandoc. Finally, we will introduce the LaTeX document-processing system, which elegantly typesets plain text into PDF files. No prior experience with any form of markup or other computer coding is required. E-mail Vishal Kamath to register (vkamath at scarletmail dot rutgers dot edu).

Naturally I will be riding my markdown and TeX hobby horses. But the actual “empowerment” (modest but real) comes in getting a more detailed understanding of the way the systems we already use handle text, and in learning more ways to manipulate that text, beyond the confines of any single program. The business of plain-text-slinging, a minor craft on its own, nonetheless forms a natural starting point for thinking more deeply about analyzing digitized texts, expressing yourself in “code” of various kinds, and composing in the digital medium.

A note of caution about working with text as data in the age of digital archives, and a note of caution about notes of caution.

“Raw data is an oxymoron”—as the title of a recent collection edited by Lisa Gitelman has it. Indeed yes. But what should this mean, in practice, for those who still want to aggregate data for analysis? Especially when the data in question is textual, and the analysis is meant to serve the scholarly study of culture? What do we do when the data is not only “cooked”—as all data is—but flawed? In this post, I want to suggest, as a twofold rule of thumb, one of Beckett’s favorite quotations: do not despair: one of the thieves was saved; do not presume: one of the thieves was damned.

I have pushed up a new version of dfr-browser, the d3-based topic-model browser I put together while working on topic-modeling journals from JSTOR. This version has more visualization options and a somewhat richer interface, thanks to bootstrap CSS.

Though I do hope others may be able to adapt the code to share and explore their own topic models, the main use of this project is to help me, and I hope others in the humanities/social sciences who are interested in this technique, to develop a stronger sense of what a topic model is and how it slices and dices a corpus of documents. For a fuller discussion of what dfr-browser does, see the new project page hosted on the handy-dandy Github Pages. Also part of the presentation is a demo of the tool, showing a (ragged though not totally uninteresting) 64-topic model of PMLA articles since 1889.

d3 is kind of addictively fun, and goes very well with the game of writing Javascript as “Lisp in C’s clothing” (as Douglas Crockford says). I wouldn’t have done this at all, but for the pleasure I had working through Crockford’s Javascript: The Good Parts and Scott Murray’s introduction to d3, Interactive Data Visualization for the Web.