Moving forward from the topic of narrative mapping to data mapping, this week we looked at the geospatial projects “The Spread of Slavery in the US” (https://lincolnmullen.com/projects/slavery/) and “Visualizing Emancipation” (http://dsl.richmond.edu/emancipation/). Much like how Ayers and Nisbet pointed out in this week’s readings, projects such as these bring the issue of “scale” to the forefront in our understanding of history. In writing on the history of slavery, for example, people will write on anything from individual slave narratives, to the networks of trade and finance the institution was built upon, to state and national policies. Yet events at all these different scales play into one another; the “macro” has no meaning without a plurality of the “micro” to constitute it, whilst the boundaries within which the “micro” takes place are shaped by the “macro” context. The “Visualizing Emancipation” map embodies this point quite well: the project plots in a time series the locations of major Union troop positions and over 3000 documented “emancipation events.” On a large scale, this shows how the emancipation of slaves in the Civil War was correlated with the expansion of Union troops into the south. Yet also for each “emancipation event” represented by a red dot on the map, one can click on it to view the details of how such a broad-sweeping observation actually played out at the level of people’s lives on the ground (i.e. did people at a particular space and time self-emancipate by fleeing toward Union lines? Did Union troops tear a town apart and conscript the slaves there? Or something else?). Meanwhile, “The Spread of Slavery in the US” project uses census data to represent the proportion of free and enslaved populations at the level of territories and state counties across the US. Through this, we can observe trends like how enslaved populations were growing in the run up to the Civil War or how transport routes like the Mississippi River tended to have high concentrations of enslaved compared to free people (indicating a lot of large plantations). We can potentially achieve a greater depth and nuance to our understanding through such geospatial data visualizations.
At the same time, data mapping can open new questions based on the various levels of scale contained within a map. When one looks at the “Visualizing Emancipation” project, its depictions of national scale can illuminate the significance of microcosms within it. For example, well before Union troops expanded into the deep south, we see clusters of emancipation events around the coast of South Carolina, which prompts further investigation as to what was so distinct about emancipation in South Carolina. In the same sense, we can use these visualizations to greater illuminate investigations into more “localized” histories. Were one to research the history of slavery in Charleston county, would it not be important to know from these visualizations that Charleston had more slaves than any other US county since 1800? We can sharpen our research and the questions we ask by seeing how the “micro” relates with and compares to the “macro” in these visualizations.
Of course, we must remember that these visualizations are not infallible, and depending upon how they are used, have the potential to skew our image of the past. What is inherent in any visualization is that it is a simplification of whatever it represents. For example, in the emancipation map, “re-enslavement of African-Americans by Confederates” and “African-Americans helping the Union” are both listed as “emancipation events.” In this sense, the sea of red dots we witness on the map can be very misleading, with two very different types of experiences lumped together as the same thing. And sometimes, the very data one uses to create a visualization can be flawed. According to the census data, Michilimackinac and Detroit (Wayne, Northwest Territory) had no slaves in 1800, yet we know from Tiya Miles’s work that this was far from the case. Thus, we must maintain our critical eye toward “what is the source?” when examining geospatial data just like when we examine any other history.
For this exercise in creating a narrative map, I chose to use Storymap JS to trace the path of a beaver pelt took from its moment of acquisition from Native trappers, through its transport by enslaved labor, and to its final consumption as an alienated commodity across the Atlantic. The “on the rails,” step-by-step nature of Storymap JS seemed to make perfect sense for this style of commodity history. However, I quickly discovered the matter was not as straight forward as I originally assumed: beavers were hunted and entered the commodity chain at numerous points of entry. In the real world, there was no single, teleological path leading from Wisconsin to England. Some traders went east to New York, while others went north to Quebec. I wanted to show multiple branching and intersecting paths, yet the “rails” began to feel constricting. Additionally, when it came to my document methodology, the very structure of the “rails” felt shaky. Ideally, this singular-path narrative would be perfect if I had the documentation to identify the purchase of a singular beaver pelt and follow it through every point of exchange. But aside from such a perfect set of documentation existing being slim, the odds of finding it digitized are even slimmer. Thus, the path I “reconstructed” felt a little anachronistic in how it pulled documents ranging from 1763 to 1835. It was also disappointing to discover that much of the rich documentation Tiya Miles drew from for the intersection of the slavery and the fur trade simply were not digitized.
Nonetheless, after completing this assignment, I could not help but feel a sense of pride that was rather unique compared to past assignments. While it is far from polished, what I created is still a piece of presentation-friendly material that I wish I had had the knowhow for in the past. And even on a more intellectual level, this exercise forced me to more closely grasp how history transpires over space (something I did not realize I was lacking before). I have dealt with commodity histories in past classes, but this narrative mapping enabled me to grasp the complexity of commodity chains in a way that simple textual narratives are unable to.
This week, I learned about text analysis using Voyant and topic modelling with Mallet. Admittedly, even though this was not my first time seeing Voyant, the tool and methodology involved still can feel like an overwhelming experience. The very idea of “reading without reading” a whole corpus of texts does not always feel intuitive, and it still took some time for me to get my barings toward understanding “what” exactly is being represented in the various windows in Voyant.
Nonetheless, after downloading the corpus of plain text files off of the Documenting the American South website, I ran the corpus through Mallet to organize the texts into 40 different topics. Like everyone in the class, the first topic I examined in Voyant was the Haitian revolution (distinguished by the appearance of words like “Hayti” or “Ouverture”), but I quickly realized that text analysis is subject to similar limitations as primary reading, but in a scaled up manner: it is hard to comprehend what you are looking at without enough prior knowledge. Thus, after taking another look at the topics identified by Mallet, I noticed that one appeared to deal with the slave trade (“distinguished by words like, “Africa”, “ship”, and “boarded”). I then took the top 25 documents in that topic and labelled them with their respective publication date, as I was interested in seeing if there was any way rhetoric surrounding the slave trade had evolved over time. After loading the 25 documents into Voyant, I noticed the “correlations” feature said that there was a correlation between the words “Africa” and “great” in the corpus. I was then curious to see if this was a rhetorical correlation that evolved over time: were people more inclined to use the word “great” while referring to Africa as time progressed, and the slave trade increasingly became seen as an “antique of the past”?
The answer turned out to be a resounding “no”. The figures at the bottom of this post were generated using the “StreamGraph” and “Trends” tools in Voyant, where blue in the graphs represents the word “great” and the brown represents “africa”. While it is clear to see that someone who was speaking of Africa was also more likely to use the adjective “great”, by no means is there a clear chronological pattern. In hindsight, this should not have been surprising: while I was going through all the CSV file containing the metadata for each document, I noticed that most documents identified by Mallet in this topic were slave narratives. In that case, the real correlation might be as simple as, the more someone talked about Africa, the more likely they were to use a celebratory word like “great.” This also raises an issue that might arise from using topic modeling to guide text analysis: people that were indicting the slave trade were the most likely to be talking about it extensively in the first place. But as a consequence, topic modeling carries the risk of cherrypicking your document sample if done carelessly.