Project URL: http://stephanietmartinez.com/projects/mappingcentraltexaslynchings/

This project aimed to visualize a list of known lynchings in central Texas in order to reveal previously unseen patterns and stimulate new questions. It also aimed to fill a gap that exists regarding this sort of visualization – numerous lynching visualizations exist regarding the US and Texas, but none that I am aware of displays to great extent the lynchings in the central Texas region.

Before I began my work on this project, I anticipated that the presentation of data would make an overwhelmingly strong case supporting the history of racialized lynching in the region. Once the map was created though, I noticed that more than half of the lynching victims were white. It occurred to me that someone viewing and using the map might conclude that race was not a major contributing factor to these lynchings since more white victims were lynched than non-whites. I don’t believe this to be the case, but it might be a good idea to include more context with the project to support this theory. I think in future iterations of this project, I’d like to create an exhibit with Omeka based on this topic and incorporate this map into that larger project. I’d also really like to find a way to include county borders on the map.

I used information from Appendix A in The Making of a Lynching Culture: Violence and Vigilantism in Central Texas, 1836-1916 by William Carrigan. I selected this material because because it contains a comprehensive list of more than 100 lynchings in the area over a span of less than 100 years, along with county locations. I used kepler.gl to easily map and visualize the locations of the lynchings. I really admire the filters kepler.gl offers in its demo version.

I really wanted the user to reflect on the race of the lynching victims, so I made it the default view by color-coding the points. I could have chosen to make all the points the same color and then allowed users to filter the points by race. I also could have added multiple layers (of the same data) and then filtered the data by race so that a layer displaying white/Black/Mexican victims could more easily be made visible or not visible based on the users preference. In the end, I decided that I wanted to make it relatively difficult to “turn off” the option to view points by race, which is why I only included one layer of data with color coded points. I am unsure whether a DH project should reflect the perspective of the creator or should be as close to a blank slate as possible. I wouldn’t necessarily categorize this as a problem – just food for thought and something to take into consideration as I move forward with DH.

I saw an opportunity to use OCR when entering the information from Carrigan’s book into an excel spreadsheet, so I did that. I then had to review and correct the OCR entries. The entire OCR process took much longer than anticipated. I then determined basic geographic coordinates for each county, and used a random location generator to create modified coordinates for each lynching so the points wouldn’t display on top of each other on the map (which would result in only a handful of points on the map instead of 100+). I also modified the alleged crime categories so they would be a bit more uniform to make that category easier to work with in the filter function (example: combining “horse theft” and “cattle theft” into “theft, horse/cattle”). Finally I uploaded the csv spreadsheet to kepler.gl, adjusted the settings of the map, and embedded the map in my post.

I was really shocked by how much time was spent in what felt like “prep” work: scanning the text, using OCR, prepping the spreadsheet, finding coordinates that would display appropriately on the map, etc. Compared to that, the actual work with the map and the post felt like a breeze. There were some creative decisions that had to be made when adjusting the settings on the map that I had not extensively considered beforehand. Overall, there was a lot of trial and error involved in the process.