This week our reading discusses “the call for methodological transparency in historical writing” as presented by Gibbs and Owens in their essay within Writing History in the Digital Age (Gibbs and Owens). Their primary argument is exactly this, a call for transparency. They request for discussions of data queries, workflows, the production and interpretation of visualizations, explaining productive failures, and a whole host of other things. Historians have been presented with a new hurdle to pass, as the sheer volume and ease of collection of data in this digital age leads to projects that comes to a strong conclusion, but the methods that hey used to get there are obscured within hundreds or thousands of data points. The need for methodological transparency allows for responses and inquiries on how the author came to their conclusions as well as their project itself.
Gibbs and Owen also discuss the uses of data and harken back to the cliometricians, claiming this is not what they are calling for, While they wish for data and methods to be transparent when writing findings, they also wish for the data to show how they reached different stages in their project. They call for historians to treat data as text, not as a reason or evidence to support a claim. You should draw your conclusions from your data, allowing it to speak for itself instead of finding data that corroborates the hypothesis you have created. In their conclusion, Gibbs and Owens request that “just as historians learn to find, collect, organize, and make sense of the traditional sources, they also need to learn to acquire, manipulate, analyze, and represent data” and this is a fitting final statement.
As a student, their call to arms for transparency is welcome. They mention Google Ngram and use it to show how vast the data located at your fingertips on a personal computer can be. A transparency with scholars would aid me in the construction of my own projects, not just digital but traditional, by illustrating how professional scholars go about chosing their topic, arguments and finding and presenting their results. The most difficult part in my writing process is finding topics. With transparency regarding how one might move from interesting data to a historical question and argument would be invaluable.
From here, our reading focuses on Cameron Blevins’ Mining and Mapping the Production of Space. Blevins provides us with a Methodology section that is exactly what the previous authors called for. Blevins addresses the problem of abundance, with the newspapers he examines containing nearly 2500 issues with 130 million words. Blevins employed “distant reading” to track trends over long periods of time and vast amounts of issues, and luckily the University of North Texas Library had digitized these newspapers. This digitization allowed Blevins to use Optical Character Recognition to mine the data of all of the newspapers. Blevins also addresses the problems with his particular methodology, in this case they were smudges, tears, and changing typefaces. All of these reduce the accuracy of the OCR. Advertisements and stories that continue later int he issue also present a problem to the computer, as they cannot distinguish between their location and the adjacent text, sometimes combining words. Blevins then removed high frequency words such as “a” and “the” and began to count the occurrences of placenames in the newspapers. This data set was used to create a map that visualize the city’s changing position in the space of the midwest, as well as the cities that Houston frequently addressed and spoke about.
All in all, I believe Blevins answered Gibbs and Owens’ call to arms regarding transparency and provided valuable insight into the mind of a scholar and the production and completion of a massive project. His use of OCR can be transferred to my research when Emancipation Day, Memorial Day, and discussions of Olustee are discussed in Floridian newspapers.