Week 3 – Mapping History

This weeks content has been the most exciting for me so far, and if I had to pick one digital history topic, is probably the one that I have been most interested in learning. Besides history/teaching, I have considered 2 other fields of study: library science, which my experiences in have already proven beneficial for this class; and geography. I have taken one geography class, in undergrad, and when I got a 99% in the course the professor told me I should go into GIS. His lesson on GIS intrigued me. Being able to map human activity and draw conclusions with the data always fascinated me, and now I am learning that I am able to merge that with history! It is a prime cross-section of my interest.

I was confused by QGIS at first, even as I followed the directions provided. I think my confusion merely arose from inexperience though; as I kept toying with the settings and doing the same steps over and over, I began to understand how the data was connected with each step. It is one thing to just follow the directions; it is another to understand WHY we are taking the steps. As I began to understand why each step was taken, I moved onto the Fairfax data. Once I understood that there were two SHP files (one with 3 districts and one that gave the precincts of 1 of those districts) I began to consider how I wanted to display the data listed.  

Each precinct lists a multitude of data: total ballots cast, total votes for each of 5 candidates, percentage of the total ballot earned for each candidate, etc. I began my project by incorporating the percentage of total votes earned for each candidate, split into 5 levels of percentage, with each candidate assigned a specific color based on party (Fig. 1). After creating 5 separate layers, one for each candidate, I realized that Barbara Comstock dominated the map because she was the top layer; this excluded the other candidates votes. I tried to exclude specific values on each layer, only showing the most votes each candidate got, but that left specific precincts value-less as they did not have an extreme majority of votes. I was left to ask, how can I show multiple layers of data?

Fig. 1 – Barbara Comstock layer, showing legend of each of the five candidates having a separate layer.

Rather than displaying multiple layers of data, I realized that I can change the filter of a new layer to display new data. I set up a new layer that displayed the winner of each district: while dominated by Barbara Comstock, it became clear which of the handful of districts John W. Foust won (Fig 2). While I was happy to find a way to display this data, I decided I wanted to go one step further. While my last layer displayed who won the election in each precinct, I was curious how close the election was in each precinct. In order to display this data, I edited the excel table which held the data I was working with. I calculated the difference between the total percentage of votes won by Foust and Comstock (I excluded the other 3 candidates, who accumulated about 10 percent of the vote between them across all precincts). I then set up a layer with the percentage different, and assigned blue and red to each end of the spectrum, with a purple center. Setting a graduated display, I was able to demonstrated which districts Comstock clearly won (dark red), which Foust clearly won (dark blue) and which was a much closer race (purple) (Fig. 3). I intend to continue working on the data, to see if I can find any other interesting patterns, but I am content that I was able to create a layer that displayed the closeness of the race. One purpose for this specific kind of map could be for political parties to know where to campaign in the next election. If a district is solidly red, the Republican party would need to spend less money campaigning there than in a district that was purple, where they barely scraped out a majority.

Fig. 2 – Winner by district, by distinctive majority
Fig. 3 – Winner by district, shaded by size of majority vote

I will briefly touch on some of the readings and HGIS examples that we looked at this week as well. One of the most succinct quotes to summarize the use of GIS in history was found in the chapter “Using GIS to Visualise Historical Data”: “The use of maps, therefore, presents a challenge to the historian, as it demonstrates the patterns within the data and challenges him or her to explain them.”[1] This shifted my perception of mapping, as previously I had considered it to be visual statistics laid out on a map. However, in statistics you manipulate the data and run tests to prove cause-and-effect. GIS, on the other hand, is just adding another layer of data to a regular table: geography. This lets historians notice the patterns within the data without having to run a statistical analysis by associating geographic area with a number. But ultimately, pure statistics lacks the context that is at the core of historians’ work: it can provide a “what?” but not a “why?”. I think that emphasizes the usefulness of GIS in historical analysis though. HGIS as a tool can provide a framework for a historical question, guiding historians to answer questions that they draw from noticing patterns in data.

Each of the databases we looked at this week demonstrates a unique way to approach GIS. In ORBIS: The Stanford Geospatial Network Model of the Roman World, you have the option to shift the routes to be reflective of the “cost” of getting to that location (based on distance, time, or expense). This shows common patterns that redefine what geographical distance means. This is pretty common today’s world, where people measure distance by time or cost based on mileage, drive times, or plane ticket costs. However, at least in my experience, people tend to think about ancient travel in terms of distance or exaggerated time. To be able to identify how quickly Romans could travel opens the door to questions of trade, expansion, and comparisons to the modern world.

The Smithsonian map of “The Decisive Moments in the Battle of Gettysburg” introduces a unique feature that displays the view of the Confederate Army during the battle. With a normal map of the battle, it is simple to see how the armies marched up to each other and could see each other across the field. However, I have never considered the ability of the armies to be able to construct that map at the time. Obviously each army would know the formation of their own forces, but knowledge of the enemy is based on what scouts and other soldiers were able to see. Utilizing this database could provide evidence for why specific actions were taken, based on available knowledge. It adds a new layer of analysis to the battle.

This post is already too long, so I will cut it off here, but I have been very invested in this week’s content. The mapping of Nazi POW camps in the USSR, the Bankes expedition files, and the mapping of urban Tokyo also each demonstrated a specific talent for how to utilize GIS and the ways that it can contribute to a deeper analysis of historical research.

[1] “Using GIS to Visualize Historical Data,” 90.

2 Replies to “Week 3 – Mapping History”

  1. Your discovery of the two SHP files and sharing that knowledge was also a life-saver in our group discussion this week, and it was so useful to watch and practice a few things alongside you!

    Also, I responded similarly to your comment on my post about the same, but I appreciate how concisely you summed up the limitations and capabilities of GIS here — that it can provide a framework for inquiry, and potentially questions that won’t be asked for many years, by other historians.

    1. i liked the fact that you admitted to trying to work on the software and repeated the steps of over and over until you understood how the software worked. I know that I tend to be impatient if I cant learn something right away. I also appreciated your summaries and analysis of the examples that we read about GIS. And I had two shape files and thought I had made a mistake when I downloaded the software. I now understand what Dr. Otis meant when she said we were going to be breaking things. Instead of just looking for the right answers, we are actually learning by practice and experimentation,

Leave a Reply

Your email address will not be published.