Posts in category: DH

I’m on sabbatical, but this post goes out with an affirmation of solidarity with everyone getting down to fall teaching in this low dishonest decade. It’s a bad time for teaching. On top of social and political crisis, there’s AI. Or rather, as part of said crisis, there’s AI. It is exhaustingly present everywhere you turn, even when you have retreated to your sabbatical bunker—perhaps especially there, if you are in your sabbatical bunker, as I am, to study the institutions of genre fiction. I am studying how, for a century, Anglophone culture and society developed a system of (apparently) formula-bound text production which, whatever its limitations, many people have turned to for its seeming copiousness and its openness even to those with few literary and economic resources. Meanwhile, in the past decade, Anglophone culture and society have developed a system of (apparently) formula-bound text production which, whatever its limitations, people are turning to in order to fill the internet with garbage text while filling the atmosphere with carbon dioxide and filling the pockets of some astonishingly awful people with more money than you can imagine. So here’s my idea: like the development of genre fiction, “AI” is the product and beneficiary of a significant transformation of literacy—that is, the production and use of writing—and its role in the distribution of social rewards.

This past Tuesday, at the kind invitation of Wen Xin, I gave a talk at the University of Kansas, under the title “Data Is a Sandwich: Lessons from the Computational Literary Field.” My slides (PDF) might only be of use for pictures of the titular sandwich, but they do also contain R code for reproducing my figures demonstrating some of the lessons I taught out of my dataculture package of cultural data in last year’s Data and Culture course. The talk itself was a bit of a retrospective on different things I’ve tried to do in literary data analysis since I first started dabbling some 14 years ago (!?!), with gestures towards what I think are the lessons I learned about how useful data gets made and turned into evidence for meaningful scholarly arguments.

Manual pin: My article, “Origins of the US Genre-Fiction System, 1890–1856,” is just out in Book History. Read all about it.

I made an R package with some “cultural” datasets of various kinds that might be of pedagogical use. It is available on github as agoldst/dataculture. See the repository page for a summary of the datasets, which I used to teach introductory analyses of:

1. cultural tastes over time and social space (names, music genres, recipes)
2. textual/paratextual signs of fictional genre (text-mining science fiction and crime)
3. historical and fictional social networks (Hamlet characters and eighteenth-century Bostonian troublemakers)

If that sounds interesting, take a look at the package and the lecture slides and lab exercises from the course I created it for, “Data and Culture” (Fall 2022). I’m hoping to save someone somewhere a few steps of wheel-reinventing.

I have two undergraduate courses this fall. I thought I’d post the syllabuses and also give away my hidden agenda.

English 358:358 Early Twentieth-Century Fiction

Syllabus (pdf)

What do James Joyce, Dashiell Hammett, Mulk Raj Anand, and Zora Neale Hurston have in common? All significant writers of English-language fiction, all active in the first half of the twentieth century, these writers lived through an epoch of global social upheaval—world wars, revolutions, mass migrations, the rise and decline of empire—and their work registers and responds to a world of crisis. Yet Joyce, the Irish experimentalist, writes nothing like Hammett, the pioneer of hard-boiled detective fiction; Anand, the committed Indian leftist, adopts very different perspectives from Hurston, the supreme Harlem Renaissance novelist. This course is a study in what is and is not shared in the fiction of these four writers and others of their era. Students will learn to analyze the forms and themes of exemplary fictions of the early twentieth century and to understand the variety of these fictions as a result of social contestation and collaboration. Readings include case studies in literary modernism (Joyce, Virginia Woolf, William Faulkner), detective fiction (Dorothy Sayers, Hammett), Harlem Renaissance fiction (Jean Toomer, Hurston), and Indian writing in English (Rabindranath Tagore, Anand).

English 359:207 Data and Culture

Syllabus (pdf)
Syllabus bibliography (pdf)

The digitization of wide swaths of the print record has opened up new challenges and opportunities for researchers in the humanities. This course introduces students to some of the key techniques used by humanities scholars to organize, manipulate, and analyze digital sources—attending both to longstanding scholarly institutions and practices that shape our understanding of digital texts (critical editions, brick-and-mortar archives, and quantitative methods within social, political, and cultural history) and to new methods for studying texts, cultural geography, and relations between and among producers and consumers of culture.

Students who complete this course will develop facility in the use of digital tools for the representation, curation, and analysis of digital text. In each case, however, we will place these relatively new tools within a longer history of humanistic inquiry and will ask: what insights can these tools provide, and what questions (and texts) do they marginalize or occlude? Our aim throughout is to examine how digitization and data science have changed the questions that humanists can ask of their sources. What does it mean to think of culture as data? What new histories do these tools and methods help us uncover? In what ways has digitization helped and hindered the ability of humanities disciplines such as history, literary studies, and art history to provide an understanding of the past that can speak to urgent questions in the present moment?

Hidden agenda after the jump (as they used to say in the Elder Days of Blogging).

The other day I was talking with an innocent bystander about some of my past work in the digital humanities. It occurred to me to wonder what a person who went looking for that work would find. The abyss also looks into you. Anyhoo, once upon a time I spent a lot of time working with data from JSTOR’s Data for Research service, a thing that no longer exists, and I produced two fairly elaborate programming projects related to topic models of text: my dfrtopics R package and my dfr-browser topic-model visualization. I am writing this post to announce that those things are still available and continue to shamble on, zombie-like, into the coming apocalypse. But I don’t plan to develop them further.

I caught wind of some of the examples of GPT-3 answering PhD-exam-style questions plausibly. It seems to me an elegant if indirect proof that Wikipedia entries on many topics are written by current or former graduate students, or people with an excellent ability to imitate them. But it also called to mind the famous arithmetic scene in Eugène Ionesco’s La leçon, which I am sure I am not the first to think of in connection with today’s debates over “stochastic parrots.”

I have an extended review of John Thompson’s new study of publishing, Book Wars: The Digital Revolution in Publishing, forthcoming in Contemporary Literature. Tempting as it is to publish an adaptation of my review as a cozy historical mystery novel on Kindle Unlimited, I prefer to keep adding to my personal weblog like it’s 2009. The journal kindly permits authors to share accepted manuscripts, so here it is: behold the digital revolution in publishing! The final edited version will appear in Contemporary Literature 62, no. 3 (Fall 2022). Spoilers: I liked the book and think that literary scholars should read the work of this eminent sociologist. Also spoilers: the digital revolution in publishing does not consist of putting LaTeX’d PDFs on your website.

Update: added the link to the published version.

For a troublingly long while I have been almost finished with an update to my software for exploring topic models, which consists of an R package, dfrtopics, and a JavaScript thingy, dfr-browser. This update is focused on making it easier to look at a single topic model more than one way or to toggle between topic models of identical or related corpora. Experience suggests that producing and interpreting even one useful topic model of a corpus is so much work that many people, myself included, are tempted to stop short of comparing the outputs of multiple models, even though this is an important way of checking that patterns revealed by any model are not purely artifacts.

About a decade ago—painful words to type—I started working on an empirical study of the bibliography of scholarship on modernism. I intended it to be a quick side-project, maybe a piece of an essay or a book introduction, that would bolster my theoretical argument for alternatives to “modernism” as a period concept for studies of the twentieth century. It was also my first foray into quantitative work, which possibly explains why it has seemed to need so many further iterations since then. In any case, I now have an essay version, which I call “Modernist Studies without Modernism,” which I am ready to make public. I have posted it to an Open Science Framework repository which also points to a reproducibility archive with R markdown source code for the essay and (processed and cleaned) data.

I was invited by the new journal Cultural Analytics to respond to an essay by Sarah Allison, “Other People’s Data: Humanities Edition.” Sarah’s essay makes an important argument for what she calls “data recycling,” and she touches on my collaboration with Ted Underwood on the way. My comment expands on the conditions for effective data reuse. I will be happy if the conversation does not get bogged down in the technicalities of “reproducibility,” and we instead insist on asking how we carry on cumulative research programs. My short essay is open access: From Reproducible to Productive.