Working with the HathiTrust dataset was initially overwhelming; the metadata categories were unfamiliar and not all of them were intuitive. Even after reading through the metadata descriptions, deciding what questions I could ask of the data and how effectively those questions could be answered took a lot of time. Like building the website last week, getting to the visualizations I ended up with took a lot of trial and error. Ultimately, I landed on two questions
- How many texts in this dataset were published in the United Kingdom and Republic of Ireland?
- How many unique authors were published in each year covered by the dataset?
Initially, I tried to use the metadata category “imprint” and filter it by country and city to make this chart, which was hugely ineffective. Using the “place” field was much more effective. It took some time to track down the MARC country codes, but I was finally able to make what seems to be a fairly comprehensive chart. This visualization was the most fiddly, as after I figured out how to implement COUNTUNIQUE for my other pivot table, I was able to fairly quickly land on a visualization that would effectively display the data. I imagine this process would be at least a bit easier with data that I was more familiar with or had collected myself. This exercise really highlighted the necessity of intentionality of data collection for me. Being attuned to the questions that the data might be used to answer will help the longevity of the project.
Turning to the Trans-Atlantic Slave Trade Database data, the pivot table that troubled me most in light of our discussions and readings was the one charting African resistance. The stark category of “African resistance” marking whether or not a slave ship’s journey was disrupted by the captives or their communities removes the nuance of other kinds of resistance undertaken by enslaved people. As Chapter 4 of Data Feminism reminds us, “What Gets Counted, Counts.” In large scale records like those tracking slave voyages, individual acts of resistance are not likely to be noted in records. Larger-scale acts that significantly impacted the voyage itself is what gets counted, eliding instances of individual resistance from the historic record. Because the emphasis of the original record keepers was on profit margins and tracking the “cargo” of human beings, it was the events that damaged the bottom line that were counted. This is where some of the issues that Johnson tackles in “Markup Bodies” come into play. Reading these numbers in the tradition cliometrics might suggest that there were relatively few acts of African resistance during the slave trade (a mere 576 recorded instances among thousands of recorded voyages.) Johnson says of the cliometrician’s work: “Statistics on their own, enticing in their seeming neutrality, failed to address or unpack black life hidden behind the archetypes, caricatures, and nameless numbered registers of human property slave owners had left behind”1. This visualization suffers in similar ways. Context is fully needed to understand the scope of different kinds of resistance available to captives, and not reduce the potential for individual actions to the recorded possibilities. Perhaps a change in the wording of the metadata category can partially accomplish this.
The aesthetics of these data visualization also troubled me. When I made the pivot table and the pie chart visualization, the chart was first generated in bright colors. I adjusted the colors to seem less cheerful, but the problem still seems to remain of how to effectively visualize data like this. Bright colors may be more engaging, but they may have the effect of minimizing the affective import of this data. Of course, minimizing human lives to data points is problematic in itself, and reproducing acts of agency by enslaved people as a pie chart is likely not the most effective way to communicate the importance of these actions. Of course, we are working in Google Sheets for now, and there are other options to visualize this data. Data visceralization may be an alternative to the route of straightforward charts and graphs, but retraumatizing participants in such an experience is also a risk. The Trans-Atlantic Slave Trade Database itself has done an excellent job of adding dimension to their data and adding context that does not reduce Black life to data points, but there are still many conversations to be had once that data is imported out of the context of the original site.
D’Ignazio, Catherine and Lauren F. Klein. Data Feminism. MIT Press: 2020.
Johnson, Jessica Marie. “Markup Bodies: Black [Life] Studies and Slavery [Death] Studies at the Digital Crossroads.” Social Text 137 36, no.4 (December 2018): 57-79.
Jessica Marie Johnson, “Markup Bodies: Black [Life] Studies and Slavery [Death] Studies at the Digital Crossroads,” Social Text 137 36, no.4 (December 2018): 61. ↩