I spent two weeks of the summer of 2017 participating in the Kaplan Humanities Institute Digital Humanities Workshop, learning from such generous and brilliant individuals as Josh Honn and Matt Taylor. My goal was to take some preliminary steps toward creating a website that would serve to introduce visitors to the rich, prodigious literature of Spanish-language newspapers in the late nineteenth and early twentieth centuries. This blog is the first iteration of such a website.
We learned a lot during the two weeks about different platforms for digital humanities scholarship, and had lively discussions about issues such as audience, accessibility, and evaluation. Each participant also worked on an individual project to showcase their research or teaching. For my project, I collected data from seven different newspapers published in 1917, all of which are digitized in the Readex Hispanic American Newspapers online archive. I put the data into a relational database consisting of three different spreadsheets: one tracking individual literary texts (poems, short stories, crónicas, and serialized novels), one tracking authors (with biographical information, if available), and one tracking the newspapers themselves.
I then used the data to create network visualization via the Stanford Humanities Lab tool Palladio. Here is a screenshot of the graph created in Palladio:
The large nodes in the graph represent the seven newspapers I collected data from: Estrella (Santa Fe), Cosmopolita (Kansas City), Mercurio (New Orleans), Revista Mexicana (San Antonio), Gráfico (New York), Demócrata Fronterizo (Laredo), and Regeneración (Los Angeles). Conspicuously missing here is La Prensa (San Antonio), for the very lame reason that in the week I was able to focus on data collection, I knew I would not have time to record the many, many literary texts published in Ignacio Lozano’s newspaper, the most prolific publisher of the era. The smaller nodes in the graph represent authors, and the edges connect the authors to newspapers that published their work. Linking edges indicate publications that circulated among more than one newspaper. The area that interests me most in the graph is the space in the middle of the four largest nodes. (Note: in Palladio, you can manipulate the placement of the nodes by dragging them with the mouse, which I did here to get a better view of that field.) Within that space we see authors who were published in three or sometimes all four of the newspapers, including nineteenth-century heavyweights like Leo Tolstoy and the Spanish writer Juan Nicasio Gallardo and twentieth-century modernistas like Amado Nervo, José Santos Chocano, and Rubén Darío.
Two quick thoughts about the graph. First, it suggests that Spanish-language newspapers in the United States mostly reproduced the middle-class literary tastes of the broader Spanish-speaking world, reprinting the same writers one would find in a newspaper in Mexico City, Caracas, or Madrid. Second, however, the profusion of edges that don’t serve as links between two newspapers indicates a dynamic archive of non-circulating texts. Many of the author nodes with no linking edge are local writers about whom we know little to nothing, and their literary production is at least as important to this field of Latinx literature as are the connecting nodes in the middle of the graph. As we are able to collect more data and create more of these visualizations, it will be interesting to test whether we can see any patterns by, for example, region or genre over time. We plan on sharing our csv data, too, so that other researchers can begin doing their own reading and visualizations with this archive. In the meantime, I’m happy to have this project up and running!