Network Analysis Reflection Blog

This Digital Humanities project studied the paths taken by runaway slaves in three states: Mississippi, Texas, and Arkansas. The main purpose of this project seems to be to visualize the escape patterns of slaves by analyzing the runaway ads slave owners put out. It also seems to be of interest to the creators of this project the differences between escape routes in different states. They hypothesized that routes through/from Texas and Arkansas would vary from the “standard” Mississippi due to their proximity to southern borders. 

Nodes and Edges: 

In this project they created various visualizations to represent the data. Below is an example of one of the finalized maps they created using Palladio, a digital geography tool: 

In this image, both nodes and edges can be identified and can help a viewer understand the map better. This map specifically shows where slaves who were caught and held in the Bexar County Jail were escaping from. The blue-ish dots at the ends of the lines are “nodes”, and mark the vertices of the data. The lines that connect these nodes are the “edges” and in this project they denote the direction and distance of travel of runaway slaves. From these details we can see that slaves in that area tried to escape toward the southeastern area of Texas. 

How Was the Project Created?

Apart from the mapping project, they also did some work with Text Mining where they took 2,500 individual ads from all three states and looked at them on a larger scale. They did this using tools on Voyant that allowed them to create visualizations of commonly used words and recognize overarching patterns that humanists might not catch when doing close-reads of individual texts. 

Other Notes

I also really appreciated how the authors of this project acknowledged some of their limitations that I think are important. They say that their corpus– the body, or main information– come from runaway slave advertisements. I think it’s particularly interesting that they decided to use this corpus because it analyzes slave patterns based on the perspective of the slave owner. They therefore mentioned how it was very possible and likely that slave owners didn’t actually know the locations of their runaway slaves and the map their information can create is not the whole picture. I just think this is important to keep in mind when interpreting their visual models.